dtrace discuss - Mar 2007 - Questions for OGB Candidates from DTrace

OGB Candidates,

We in the DTrace Community have a couple of specific questions and 
concerns that will affect how our Core Contributors vote in the upcoming
OGB elections.  To see where you stand on the issues that are important
to us, we would like you to answer the following questions before
campaigning ends on Sunday night.  Please cc: dtrace-discuss at opensolaris.org
on your answers, and thanks in advance for helping us make an informed
decision!

	- Bryan

--------------------------------------------------------------------------
Bryan Cantrill, Solaris Kernel Development.       http://blogs.sun.com/bmc


Licensing Questions:

- DTrace is one of only a small handful of OpenSolaris technologies that
  has actually been incorporated into other operating systems.  Thus,
  your position on dual-licensing is very important to us; what is your 
  position on dual-licensing in general?

- Do you agree with the conclusions and decrees of CAB/OGB Position 
  Paper # 20070207?

- The OGB is responsible for the representation of OpenSolaris to 
  third parties.  If a third party were to inquire about incorporating
  DTrace into a GPL''d Program, what would be your response or position?

Constitutional Questions:

- DTrace is currently a Community Group, but some could argue that it would 
  make more sense for DTrace to be a Project in (say) the Observability
  Community Group.  In your mind, what is (or should be) the difference
  between a Community Group and a Project -- and where should DTrace fall?

- The Draft Constitution says next-to-nothing about where the authority 
  lies to make or accept changes to OpenSolaris -- only that Projects
  operate at the behest of Community Groups, and that Community Groups
  can be "terminated" by the OGB.  In your opinion, where does or
should
  this authority lie?  And do you believe that the Constitution should
  or should not make this explicit?  Finally, under what grounds do you
  believe that a Community Group should be "terminated"?

- The Draft Constitution says that Community Groups (and in particular,
  the Community Groups'' Facilitators) are responsible for
"communicating
  the Community Group''s status to the OGB"; what does this mean to
you?

- According to the Draft Constitution, "nominations to the office of
  Facilitator shall be made by the Core Contributors of the Community
  Group, but the OGB shall not be limited in their appointment to those
  nominated."  Under what conditions do you believe that the selection of
  a Facilitator would or could fall outside of the nominations made by
  a Community Group''s Core Contributors?

- According to the Draft Constitution, "non-public discussion related to
  the Community Group, such as in-person meetings or private communication,
  shall not be considered part of the Community Group activities unless or
  until a record of such discussion is made available via the normal meeting
  mechanism."  In your opinion, in the context of a Community Group like
  DTrace -- where a majority of the Core Contributors spend eight to ten
  hours together every work day -- what does this mean?  Specifically, what
  does it mean to be (or not to be) "considered part of the Community
  Group activities"?  And in your opinion, what role does the OGB have in
  auditing a Community Group''s activities?

Potpourri:

- Historically, binary compatibility has been very important to Solaris,
  having been viewed as a constraint on the evolution of technology.
  However, some believe that OpenSolaris should not have such constraints,
  and should be free to disregard binary compatibility.  What is your
  opinion?

- If a third-party were to use and modify DTrace in a non-CDDL''d
system,
  whose responsibility is it to assure that those modifications are
  made public?  To put it bluntly:  is enforcing the CDDL an OGB issue?

- Do you have an opinion on the patentability of software?  In particular,
  what is the role of the OGB -- if any -- if Sun were to initiate legal
  proceedings to protect a part of its software patent portfolio that
  is represented in OpenSolaris?

- When you give public presentations, do you run OpenSolaris on your laptop?
  Have you ever given a public demonstration of OpenSolaris technology?

- And an extra credit question:  Have you ever used DTrace?  When did you
  most recently use it, and why?  The answers "just now" and "to
answer
  this question" will be awarded no points. ;)

Bryan Cantrill writes:
> - DTrace is one of only a small handful of OpenSolaris technologies that
>   has actually been incorporated into other operating systems.  Thus,
>   your position on dual-licensing is very important to us; what is your 
>   position on dual-licensing in general?
I''m opposed to it unless there''s a compelling explanation
showing how
and why it (on balance) benefits OpenSolaris.  So far, I haven''t seen
that compelling explanation.

The issues are quite complex, and the additional burden for
contributors seems fairly large.  The benefit is unclear and mostly
theoretical.
> - Do you agree with the conclusions and decrees of CAB/OGB Position 
>   Paper # 20070207?
Essentially, yes.  I don''t see a good reason to jump the gun here, and
I do think that speculation about licensing is harmful to the
community.
> - The OGB is responsible for the representation of OpenSolaris to 
>   third parties.  If a third party were to inquire about incorporating
>   DTrace into a GPL''d Program, what would be your response or
position?
It''s the responsibility of that third party to evaluate and obey the
licensing restrictions of the software he''s using.  I think it would
be a terrible mistake for anyone (including an OGB member) to attempt
to provide legal advice to any third party.

If you''re asking my _own_ understanding of the legal issues involved,
I think the situation is murky.  It''s not clear to me what types of
use fall under the "mere aggregation" clauses of GPLv2 versus those
that form a dependency requiring release under GPL.

I suspect (again, my private opinion, and not intended as legal advice
to _anyone_) that private use in this manner would probably be quite
safe, but that you''d be in an impossible position if you tried to
redistribute the result unless you could convince one of the authors
to license in a different way.

Your concern here seems to be the use of DTrace on Linux.  I have no
apriori position on that matter.  I''m not convinced it''s
something
that is really relevant to the OGB -- one way or the other.
> Constitutional Questions:
> 
> - DTrace is currently a Community Group, but some could argue that it would
>   make more sense for DTrace to be a Project in (say) the Observability
>   Community Group.  In your mind, what is (or should be) the difference
>   between a Community Group and a Project -- and where should DTrace fall?
I think projects should have a clear set of objectives and a lifetime
-- after which the work is done and the project is retired.
Communities corral multiple projects and should set direction for
them, and likely have hazier objects and perhaps no set lifetime.

It thus makes more sense to me that DTrace remain as a community
rather than as a project, as it seems intended to direct projects in
various areas to improve the operation of DTrace.
> - The Draft Constitution says next-to-nothing about where the authority 
>   lies to make or accept changes to OpenSolaris -- only that Projects
>   operate at the behest of Community Groups, and that Community Groups
>   can be "terminated" by the OGB.  In your opinion, where does or
should
>   this authority lie?
I think _responsibility_ should be as high as possible, while
_authority_ is as low as possible.  In other words, I would like to
see projects that have the authority to make changes to OpenSolaris,
sponsored by communities that delegate that authority, and an OGB that
delegates to the communities.
>  And do you believe that the Constitution should
>   or should not make this explicit?
I don''t think it should be explicit because I suspect that a fair
amount of flexibility is needed.
>  Finally, under what grounds do you
>   believe that a Community Group should be "terminated"?
I think it should be extraordinarily rare.

Groups that have taken no substantive action in a prolonged period of
time or that are either overly broad or too narrow to serve their user
base should be encouraged to discuss their goals and report to the
OGB.  I''d rather see consensus on how to deal with those problems than
"rulings" from the OGB.  I''d be surprised to encounter a case
where
the OGB sees a problem but the community itself doesn''t.
> - The Draft Constitution says that Community Groups (and in particular,
>   the Community Groups'' Facilitators) are responsible for
"communicating
>   the Community Group''s status to the OGB"; what does this
mean to you?
It means reporting (as necessary) on the progress of endorsed
projects, identified community goals, and future work -- anything that
can help the OGB members understand the scope and purpose of the
community.

I don''t think it means that any sort of formal notice is required, if
that''s the underlying question.
> - According to the Draft Constitution, "nominations to the office of
>   Facilitator shall be made by the Core Contributors of the Community
>   Group, but the OGB shall not be limited in their appointment to those
>   nominated."  Under what conditions do you believe that the selection
of
>   a Facilitator would or could fall outside of the nominations made by
>   a Community Group''s Core Contributors?
I have a hard time crafting a case in which I could see such a thing
happening.  I suspect that there would have to be much deeper and more
dire problems afoot with the functioning of the community for this to
be true, and that it''d be better to address those problems rather than
talking about this one role.

I think it''s important to have a place where (in the extreme) the buck
stops moving.  I think the OGB needs to perform that function.
> - According to the Draft Constitution, "non-public discussion related
to
>   the Community Group, such as in-person meetings or private communication,
>   shall not be considered part of the Community Group activities unless or
>   until a record of such discussion is made available via the normal
meeting
>   mechanism."  In your opinion, in the context of a Community Group
like
>   DTrace -- where a majority of the Core Contributors spend eight to ten
>   hours together every work day -- what does this mean?  Specifically, what
>   does it mean to be (or not to be) "considered part of the Community
>   Group activities"?
I don''t think hallway conversations, no matter how well-intended,
substitute for participation in the open community.  I think it''s very
important that we build community involvement, and that doing so is
extremely difficult if the "real" decisions are made in secret, and if
external participants can tell that this is being done.

When it comes to community-related decisions, such as deciding whether
to endorse a particular project, or recognizing someone as a core
contributor, or proposing a new project, I''d very much like to see
those discussions held on the open mailing lists.

Yes, even if you''re all on the third floor of MPK17.  ;-}

Dumping those decisions as a fait accompli on the rest of the possible
contributors outside of that hallway is discourteous.  Ultimately,
it''ll result in the failure of the community to gain any real traction
outside of Sun, and that would be a real shame.
>  And in your opinion, what role does the OGB have in
>   auditing a Community Group''s activities?
Unless someone comes to the OGB with a complaint about a group''s
activities, I don''t think it would be wise for the OGB to attempt to
play the role of "group auditor."

So, no, I don''t expect the OGB to step in and say "you
can''t endorse
that project until you have an open discussion."  However, I''d
like to
see communities recognize the value of open discussion.
> Potpourri:
> 
> - Historically, binary compatibility has been very important to Solaris,
>   having been viewed as a constraint on the evolution of technology.
>   However, some believe that OpenSolaris should not have such constraints,
>   and should be free to disregard binary compatibility.  What is your
>   opinion?
I think backward compatibility is crucial to the success of
OpenSolaris.

Independent distributions are free to make such decisions on their
own, but when it comes to accepting code into OpenSolaris, I believe
we need to follow more constrained rules.

An unresolved issue with OpenSolaris in general is what to do with
changes that have a Major release binding.  I don''t think we have a
very good story when it comes to dealing with release binding skew
between the distributions.  I suspect that we may have to deal with
this issue (perhaps even by _forking_ the source) at some point in the
future.  (That''s just speculation, though.  I know of no such need to
worry about it now.)
> - If a third-party were to use and modify DTrace in a non-CDDL''d
system,
>   whose responsibility is it to assure that those modifications are
>   made public?  To put it bluntly:  is enforcing the CDDL an OGB issue?
Legally, my (again, non-expert) understanding is that the copyright
holders and the original authors are the ones who have an interest in
enforcing licensing matters.

I don''t think that enforcing the CDDL (or any other license on the
code -- OpenSolaris has a diverse set of licenses, not just CDDL)
should be an OGB issue.  I think that the community (at large) could
possibly decide to _make_ it an OGB issue, but I don''t see how
it''s an
OGB issue now.
> - Do you have an opinion on the patentability of software?  In particular,
>   what is the role of the OGB -- if any -- if Sun were to initiate legal
>   proceedings to protect a part of its software patent portfolio that
>   is represented in OpenSolaris?
I''m not thrilled with the way in which the US patent system works with
regard to software.  I don''t think that position, though, has anything
to do with the OGB.

My (again, non-expert) understanding of the patenting issues is that
the CDDL provides a limited grant for users of the software.  I''d
expect Sun to honor that commitment.  Other than that, lawsuits among
users of OpenSolaris (whether involving Sun or not) do not by
themselves sound to me like OGB matters *unless* the issue threatens
the operation of the community.  It''s only the community operation
that''s at issue for me.
> - When you give public presentations, do you run OpenSolaris on your
laptop?
I always run OpenSolaris on my laptop.
>   Have you ever given a public demonstration of OpenSolaris technology?
No; I''ve never been in a position to do so.  I don''t hide it,
though.
> - And an extra credit question:  Have you ever used DTrace?  When did you
>   most recently use it, and why?  The answers "just now" and
"to answer
>   this question" will be awarded no points. ;)
I''ve used it many times.  Most recently was in investigating CR
6479968.  I used it to narrow down the possible culprits.
Unfortunately, as the actual problem involved interplay between
multiple CPUs -- and DTrace upset that applecart -- I had to resort to
hacking up the code (with mutexes) to pinpoint the problem.

Other times, I''ve been more successful.  ;-}

-- 
James Carlson, Solaris Networking              <james.d.carlson at
sun.com>
Sun Microsystems / 1 Network Drive         71.232W   Vox +1 781 442 2084
MS UBUR02-212 / Burlington MA 01803-2757   42.496N   Fax +1 781 442 1677

Hi Bryan,

Some really great questions, thanks for putting these together. I''ve
purposely
avoided reading any other candidate replies before replying to this mail, and
hope these will give you an affirmative good/sour taste in your mouth rather
than going away undecided. Sorry that they appear long winded, but they are
hellishly difficult to answer in a few short lines.
> Licensing Questions:
Let me first say, licensing very quickly gets me out of my comfort zone.
It''s a
long way from my core competency, and definitely not a skill that I would bring
to the OGB nor expect anyone to think I could.
> - DTrace is one of only a small handful of OpenSolaris technologies that
>   has actually been incorporated into other operating systems.  Thus,
>   your position on dual-licensing is very important to us; what is your 
>   position on dual-licensing in general?
There is no one license that is better than others. They''re all
different, they
all have their pros and cons, and everyone has their favourite. Nor do I believe
that choices X years ago should not be re-evaluated if environmental conditions
change. That said, there''s obviously been a mixture of opinions about
dual-licensing, whether the fact that it complicates things from a developer
point of view, or opens up new opportunities for a different group of
developers. I''d personally much prefer to maintain a single license,
unless
there was a *significant* benefit to the majority of stakeholders (you
can''t
please everyone) to going the dual-license approach.

However, that being said, I am not a current stakeholder in the code and
I''m not
a lawyer, and I would personally step out of the discussion from those counts.
Furthermore, I don''t think the OGB should be making any of these types
of
decisions without those types of people coming to the table.
> - Do you agree with the conclusions and decrees of CAB/OGB Position 
>   Paper # 20070207?
I think the timing of the paper was poor, and I''m not sure I entirely
understand
why the OGB felt they had the responsibility to publish such a paper at that
stage. In terms of content, there are points in it that I fully agree with, in
that the GPLv3 *is* an unfinished license as it stands right now; however, there
are elements that I disagree with and I don''t believe that the OGB
should have
any right to discourage discussion.

Would I have handled this better? I''d like to think I would have pushed
for
being a *lot* more transparent communication wise. I believe I would have pushed
(and plan to do so if elected) for a license working group of people to be
formed comprised of a selection of informed people across the community with
different skills and backgrounds who have earned the trust of the wider
community.
> - The OGB is responsible for the representation of OpenSolaris to 
>   third parties.  If a third party were to inquire about incorporating
>   DTrace into a GPL''d Program, what would be your response or
position?
To be absolutely honest, I''d have no idea. As a start, I''d do
a lot more
research and weigh up the benefits, including discussion with the stakeholders.
I''m not sure I can give you a black or white answer on this. As a wild
example,
if you got the opportunity to have DTrace embedded on every DVD player in the
world, would you take it? Would you still take that opportunity if it was based
GPL''d software?

The OGB needs to act as a bridge for many of these types of questions. However,
I don''t believe it is solely responsible for making these types of
decisions. If
the discussion is road-blocked, and a decision needs to be made because it
deeply effects the growth of the community, then and only then should I think
the OGB needs to step in. We were no way close to this with the recent GPLv3
discussion.
> Constitutional Questions:
You''re assuming that I actually agree with the current draft of the
constitution. As I''ve mentioned elsewhere, I''m tied between
wanting to accept it
or reject it during the upcoming vote. There''s parts of it that I
completely
disagree with, and think are vastly over complicated. We grew up way too quickly
IMO, put too much structure in at the start and now struggling with the
consequences. However, I''m also pragmatic to know that we''ve
sat on this for 2
years, and sometimes you need to compromise a little to make progress.
> - DTrace is currently a Community Group, but some could argue that it would
>   make more sense for DTrace to be a Project in (say) the Observability
>   Community Group.  In your mind, what is (or should be) the difference
>   between a Community Group and a Project -- and where should DTrace fall?
I''ve struggled with dealing with the split between Community and
Project myself
when working on the various desktop related proposals. JDS started off as a
''sub-community'' of the Desktop community, and then later moved
over to a Project
only because of the constraints of the SCM system in place. It''s a
curse quite
frankly, and badly thought out. From a desktop point of view, having a
''Desktop
Community'' hasn''t really gained us anything other than
providing a forum that is
jointly shared between JDS project development, XFCE, Looking Glass, KDE and
others. We could just have easily created mailing lists for each project and
have the generic desktop list go quiet.

We''ve also seen how much the division has confused the elections in
terms of the
creation of a Core Contributor voting list when a given project hasn''t
been
associated with any existing community.

It''s all heavily broken. If I had to make a call right now, that call
would be
made on the technical barriers. Clearly being a Project gives you the
opportunity of having access to an SCM implementation. That wins it for me, and
would give Brendan the opportunity to host his helper-monkey project there.
> - The Draft Constitution says next-to-nothing about where the authority 
>   lies to make or accept changes to OpenSolaris -- only that Projects
>   operate at the behest of Community Groups, and that Community Groups
>   can be "terminated" by the OGB.  In your opinion, where does or
should
>   this authority lie?  And do you believe that the Constitution should
>   or should not make this explicit?  Finally, under what grounds do you
>   believe that a Community Group should be "terminated"?
Definitely not with the OGB. Authority should lie with the community of
developers who are core contributors to their respective component. You have the
invested interest, the experience and the maturity to be able to make that call
- I''d hope you do it based on community (within your component)
consensus where
possible.

There are a lot of things that the constitution should make more explicit, and a
lot of things that the constitution should be purposely vague about to allow
things to organically develop - however I think it''s important that we
also
don''t get to a stage where developers are constantly quoting from the
constitution when trying to resolve an argument.

Community groups should be terminated when some good discussion for how to
better organize the OpenSolaris project as a whole has happened and been agreed
to. There are some very obvious examples of communities *and* projects that have
failed miserably. I partially blame that both on the complexity of the initial
structure we all had to follow, and on the relative inexperience of people
creating something new and the move towards an open development model.
It''s
clear that we have some hard yards to cover still.
> - The Draft Constitution says that Community Groups (and in particular,
>   the Community Groups'' Facilitators) are responsible for
"communicating
>   the Community Group''s status to the OGB"; what does this
mean to you?
I would rather like to think that this would only be if something specific came
up in which they looked to the OGB for advice and/or resources.

However, I would expect to have board appointing working groups to communicate
their status eg. licensing working group (if desired), membership and elections
committee, release team, marketing/press releases, employee etc. In my
experience with being on the GNOME Foundation and involved in many groups there,
I''ve found this works really well, and I''d be pushing for this
division as
something that the OpenSolaris community might move towards.
> - According to the Draft Constitution, "nominations to the office of
>   Facilitator shall be made by the Core Contributors of the Community
>   Group, but the OGB shall not be limited in their appointment to those
>   nominated."  Under what conditions do you believe that the selection
of
>   a Facilitator would or could fall outside of the nominations made by
>   a Community Group''s Core Contributors?
I don''t believe an OGB appointed Facilitator sounds like a good idea,
and would
much rather one being elected by informal consensus from the Core Contributors.
In practice it''s likely to be the person that cares and has time to do
so.

My gut feeling is that this again smells like we''re trying to
mico-manage every
possible interaction within the community, without letting it naturally evolve.

FWIW, I really don''t know how best to solve these issues in the
constitution.
Clearly scrapping the entire draft is the right way to go forward, but
I''d love
to work towards making some improvements to simplify things.
> - According to the Draft Constitution, "non-public discussion related
to
>   the Community Group, such as in-person meetings or private communication,
>   shall not be considered part of the Community Group activities unless or
>   until a record of such discussion is made available via the normal
meeting
>   mechanism."  In your opinion, in the context of a Community Group
like
>   DTrace -- where a majority of the Core Contributors spend eight to ten
>   hours together every work day -- what does this mean?  Specifically, what
>   does it mean to be (or not to be) "considered part of the Community
>   Group activities"?  And in your opinion, what role does the OGB have
in
>   auditing a Community Group''s activities?
I have no problem with discussions that happen outside the public forums. That
type of stuff happens all the time whether it''s in an office, in the
corridor,
during social nights, at conferences, whatever - it''s hard to avoid. I
think in
general people need to be more aware that there''s a wider community out
there
that isn''t privy to these discussions and may feel left out if anything
*big* is
decided.

I know for a fact that all the current Core Contributors for DTrace are very
aware of this fact, and I''m a lot less worried about you guys than I am
for
other communities or groups. You''ve already had the opportunity of
seeing open
development work for you and the obvious benefits there are in that.
I''d love to
see the responsibility and guidance that has been shown in the DTrace community
been transferred across the other groups.

In summary, the OGB isn''t a baby sitters club. It''s up to the
various
communities and projects to be responsible about their activities. They are the
ones that will lose out if they lose the community trust, because people will
feel less inclined to contribute ideas, code or bug reports.
> Potpourri:
> 
> - Historically, binary compatibility has been very important to Solaris,
>   having been viewed as a constraint on the evolution of technology.
>   However, some believe that OpenSolaris should not have such constraints,
>   and should be free to disregard binary compatibility.  What is your
>   opinion?
I think binary compatibility is important, and Solaris engineers should pride
themselves on being able to maintain it while continuing to advance the
technology. However, I think there are compromises that can be made so that
OpenSolaris can have a healthy mixture between compatibility and evolution.
There are benefits to both, and I believe it''s fundamentally important
for us to
be aware of that if we want the project to grow.
> - If a third-party were to use and modify DTrace in a non-CDDL''d
system,
>   whose responsibility is it to assure that those modifications are
>   made public?  To put it bluntly:  is enforcing the CDDL an OGB issue?
Yes, I believe so. However, often governing boards like the OGB are too busy
with their respective work to know about a situation where someone was violating
the conditions of the license. I would hope that community members would be
vigilant and inform OGB of any violations in the licensing.
> - Do you have an opinion on the patentability of software?  In particular,
>   what is the role of the OGB -- if any -- if Sun were to initiate legal
>   proceedings to protect a part of its software patent portfolio that
>   is represented in OpenSolaris?
To be honest, again, I don''t know enough about software patents and the
software
patent system. I can understand why we have them historically, but I
don''t
believe they are healthy for free software. In this case, I''d look for
advice
from the rest of the board and/or other people that are specialized in this
area.
> - When you give public presentations, do you run OpenSolaris on your
laptop?
>   Have you ever given a public demonstration of OpenSolaris technology?
I currently run Solaris Express build 53 on my Toshiba laptop. I try to make an
effort to make sure I''m no greater than 3 or 4 builds behind, but
it''s a little
hard when you''re working at home full time, and living in a country
where
broadband still has its problems.

I''ve demoed DTrace previously during the annual GNOME conference, but
present
regularly at conferences and user groups.
> - And an extra credit question:  Have you ever used DTrace?  When did you
>   most recently use it, and why?  The answers "just now" and
"to answer
>   this question" will be awarded no points. ;)
Yes. I''ve been using DTrace pretty much since you demoed it during an
internal
Solaris Desktop summit a couple of years ago. I know enough to randomly probe
userspace applications and figure out what they''re doing. I
wouldn''t classify
myself as a regular DTrace user, nor would I classify myself as a regular
developer either. What little I''ve used recently was simply to see
whether a
function in the gnome-panel was getting called or not while I was debugging some
branding related code. I have yet to use it to find a long outstanding bug.


Glynn

On Thu, 8 Mar 2007, Bryan Cantrill wrote:

I''ll keep my replies as brief as possible (I''m preparing to
exhibit
at a trade show tomorrow); requests for more detail are welcome!
> Licensing Questions:
> 
> - DTrace is one of only a small handful of OpenSolaris technologies that
>   has actually been incorporated into other operating systems.  Thus,
>   your position on dual-licensing is very important to us; what is your 
>   position on dual-licensing in general?
In general, it strikes me as a nightmare.  As far as [Open]Solaris
technologies like DTrace are concerned, I think the CDDL is more
than sufficient, and therefore dual-licensing is not necessary or
desirable.
> - Do you agree with the conclusions and decrees of CAB/OGB Position 
>   Paper # 20070207?
I helped write it, so I agree with it!  (Though in retrospect, the
word "decree" was poorly choses; "declare" might have been a
better
choice...)
> - The OGB is responsible for the representation of OpenSolaris to 
>   third parties.  If a third party were to inquire about incorporating
>   DTrace into a GPL''d Program, what would be your response or
position?
My response would be along these lines: we welcome the incorporation
of OpenSOlaris technologies like DTrace into 3rd party programs, provided
the license of that 3rd party program permits it.  A GPLed program
would therefore need to be relicensed or dual-licensed using the CDDL.
> Constitutional Questions:
> 
> - DTrace is currently a Community Group, but some could argue that it would
>   make more sense for DTrace to be a Project in (say) the Observability
>   Community Group.  In your mind, what is (or should be) the difference
>   between a Community Group and a Project -- and where should DTrace fall?
To me, a community is the "meeting place" of one or more projects with
a
common or complementary theme, a project is something which is expected
to produce deliverables in some form or other (e.g., code or docs).  With
that in mind, it seems to me that DTrace would indeed fit into the "project
within the observability community" model.
> - The Draft Constitution says next-to-nothing about where the authority 
>   lies to make or accept changes to OpenSolaris -- only that Projects
>   operate at the behest of Community Groups, and that Community Groups
>   can be "terminated" by the OGB.  In your opinion, where does or
should
>   this authority lie?  And do you believe that the Constitution should
The authority should lie with those most qualified to make the decision,
perhaps with the OGB be the "final court of appeal".  Thos most
qualified
would presumably the core contributors of the community into which the
change is trying to be integrated.
>   or should not make this explicit?  Finally, under what grounds do you
>   believe that a Community Group should be "terminated"?
When it has outlived its usefulness, or maybe when the community in question
doesn''t align with the goals of the OpenSolaris community at large. 
But
only in extreme circumstances; I expect that in most cases, communities
would voluntarily terminate.
> - The Draft Constitution says that Community Groups (and in particular,
>   the Community Groups'' Facilitators) are responsible for
"communicating
>   the Community Group''s status to the OGB"; what does this
mean to you?
Is the community still alive?
> - According to the Draft Constitution, "nominations to the office of
>   Facilitator shall be made by the Core Contributors of the Community
>   Group, but the OGB shall not be limited in their appointment to those
>   nominated."  Under what conditions do you believe that the selection
of
>   a Facilitator would or could fall outside of the nominations made by
>   a Community Group''s Core Contributors?
Probably fairly extreme conditions, when it was "obvious" that the
facilitator chosen was unsuitable for the role.
> - According to the Draft Constitution, "non-public discussion related
to
>   the Community Group, such as in-person meetings or private communication,
>   shall not be considered part of the Community Group activities unless or
>   until a record of such discussion is made available via the normal
meeting
>   mechanism."  In your opinion, in the context of a Community Group
like
>   DTrace -- where a majority of the Core Contributors spend eight to ten
>   hours together every work day -- what does this mean?  Specifically, what
>   does it mean to be (or not to be) "considered part of the Community
In nutshell, anything that impacts others.  So, day-to-day design decision
may not qualify, but "big picture" decision probably do.
>   Group activities"?  And in your opinion, what role does the OGB have
in
>   auditing a Community Group''s activities?
I would hope that communities would be largely self auditing, so again
the OGB would be a last resort.
> - Historically, binary compatibility has been very important to Solaris,
>   having been viewed as a constraint on the evolution of technology.
>   However, some believe that OpenSolaris should not have such constraints,
>   and should be free to disregard binary compatibility.  What is your
>   opinion?
Binary compatibility is one of [Open]Solaris'' great strengths, and as
such should NOT be disregarded lightly, if at all.  The fact that OpenSolaris
contains many great innovations, without sacrificing backwards compatibility,
supports this position I think.
> - If a third-party were to use and modify DTrace in a non-CDDL''d
system,
>   whose responsibility is it to assure that those modifications are
>   made public?  To put it bluntly:  is enforcing the CDDL an OGB issue?
No; there are non-OpenSolaris projects that use the CDDL, and it is not
the OGB''s job to interfere with them.  I''d say the
responsibility lies
with those making the changes.
> - Do you have an opinion on the patentability of software?  In particular,
I''m not a fan of software patents, and I''m not sure I see
their relevance
in an open source project.
>   what is the role of the OGB -- if any -- if Sun were to initiate legal
>   proceedings to protect a part of its software patent portfolio that
>   is represented in OpenSolaris?
I''d say it''s none of the OGB''s business what Sun does
with it''s IP assets,
though a consultation with the OGB might be advisable.
> - When you give public presentations, do you run OpenSolaris on your
laptop?
Is there another option?!
>   Have you ever given a public demonstration of OpenSolaris technology?
Yes.
> - And an extra credit question:  Have you ever used DTrace?  When did you
Yes.
>   most recently use it, and why?  The answers "just now" and
"to answer
To answer this question.
>   this question" will be awarded no points. ;)
Oh.  In that case, yes.  I was investigating some problem that I can''t
recall right now, and I''ve also used it for pedagogical reasons if that
counts.


Hope these answers have been useful,

-- 
Rich Teer, SCSA, SCNA, SCSECA, OpenSolaris CAB member

CEO,
My Online Home Inventory

Voice: +1 (250) 979-1638
URLs: http://www.rite-group.com/rich
      http://www.myonlinehomeinventory.com

On Thu, Mar 08, 2007 at 10:01:17AM -0800, Bryan Cantrill wrote:
> Licensing Questions:
> 
> - DTrace is one of only a small handful of OpenSolaris technologies that
>   has actually been incorporated into other operating systems.  Thus,
>   your position on dual-licensing is very important to us; what is your 
>   position on dual-licensing in general?
As I noted in my position paper:

	(a) the OGB does not control licensing, and

	(b) to the extent that the OGB would be consulted on the
	matter, I''m opposed to dual-licensing.

The well-known opportunity it offers for license-based forks is a
significant drawback that would have to be balanced and more by
compelling benefits.  No one has yet articulated such benefits, and I
have found no evidence myself that they exist.  The advantages
presented by proponents of such a licensing scheme appear to be
predicated on the idea that the second license would be GPLv3 (which
is not complete), and that its use would dramatically increase the
size of our community by drawing in the FSF as a partner in our
technical work.  Those are two large ''ifs'' for a
''maybe'' we''re poorly
positioned to handle.
> - Do you agree with the conclusions and decrees of CAB/OGB Position 
>   Paper # 20070207?
Generally, yes.  See above.
> - The OGB is responsible for the representation of OpenSolaris to 
>   third parties.  If a third party were to inquire about incorporating
>   DTrace into a GPL''d Program, what would be your response or
position?
I would note that my lay reading of the GPL would preclude that party
from distributing the resulting product without violating the terms of
that license.  I would also advise that party to seek legal counsel,
as with any licensing concern.  That''s as far as I''d go,
however; the
OGB does not hold the copyright to DTrace and is not in a position to
warn or litigate against infringers.
> Constitutional Questions:
> 
> - DTrace is currently a Community Group, but some could argue that it would
>   make more sense for DTrace to be a Project in (say) the Observability
>   Community Group.  In your mind, what is (or should be) the difference
>   between a Community Group and a Project -- and where should DTrace fall?
These two questions are not necessarily well-linked.  The difference
between a Project and a Community Group is straightforward.  A Project
owns one or more gates and does direct technical work with the intent
to add or improve a specific aspect of the software they contain.  A
Community Group is the unit of independent government as defined by
the Constitution; it is responsible for directing and guiding Project
teams and others doing work that affects a broadly-defined set of
interests.

Others have suggested that a Project is defined to have a defined life
span (presumably terminating upon integration into a consolidation).
I disagree with this definition - a project (like DTrace) which
provides a large and useful set of functionality will never be fully
complete unless and until it is replaced wholesale.  So long as the
Project''s work remains in use, it is important that some collaborative
unit exist to provide a home for those using and improving it.

DTrace is unquestionably a Project.  Whether it deserves a Community
Group of its own[0] depends on the granularity at which we wish to
distinguish among Community Groups and the amount of overlap among
them.  That is, if Observability is held to be a Community Group
distinguished from others at the correct granularity, DTrace should
not be a separate CG, as its function would be a strict subset of
another valid CG''s.  Instead, the DTrace leadership would be expected
to participate in the Observability Group''s activities, offering
guidance and advocacy for consumers of its work.  As part of that
transition, mutually acceptable agreements regarding contributorship
grants and leadership structure would need to be in place regarding
the merged community (much like any corporate merger).  Alternately,
however, I could envision a finer-grained set of Community Groups with
some overlap; DTrace might fit alongside, for example, a Debuggers
Community Group in such a scheme.  My personal preference is for a
smaller number of larger Community Groups, some of them controlling
the long-term maintenance of consolidations and others providing
guidance to project teams (and the consolidation owners) based on
their particular areas of technical expertise.  I believe this would
promote a vision of our software as an integrated whole.  Just as
importantly, even under such a system, any large and ambitious Project
would fall inside the scope of several Community Groups'' areas of
interest.  Expecting project teams to interact with dozens of Groups''
leaders would seem to introduce excessive and unneccessary complexity.

[0] The existence of DTrace as a Project ought not preclude the
existence of other Projects which seek to enhance it.
> - The Draft Constitution says next-to-nothing about where the authority 
>   lies to make or accept changes to OpenSolaris -- only that Projects
>   operate at the behest of Community Groups, and that Community Groups
>   can be "terminated" by the OGB.  In your opinion, where does or
should
>   this authority lie?  And do you believe that the Constitution should
>   or should not make this explicit?  Finally, under what grounds do you
>   believe that a Community Group should be "terminated"?
As I noted in my position paper, I believe the authority for code
acceptance should reside with Community Groups responsible for the
targetted consolidation.  Those CGs would be expected to delegate some
or all of that authority in turn to specific individuals forming the
C-Team for a particular release.  While some minor changes will be
needed to this strategy to accomodate open development, the basic
process has worked well for a long time, and I see little reason to
alter it radically.

As I''ve noted in several messages, I would prefer that the
Constitution had made at least some of this more explicit.  The
absence of this specification leaves the OGB with a set of
illegitimate Sun entities excercising effective control over matters
the Charter clearly leaves to the OGB, and offers no transition plan,
timetable, or framework in which to take over these functions.  This
will present an additional challenge to the first elected OGB.

Community Groups formed under a coherent and comprehensive strategy
such as the one I hope the OGB will provide should generally be
terminated only for inactivity or other clearly self-induced act of
dissolution (such as a voluntary merger with another Community Group,
approved by the OGB).  Unfortunately, we also have a large number of
existing Communities which do not fit well within any strategy one
could retroactively imagine, and the OGB will be obligated to
rationalise this situation.  The process of doing so will likely
involve terminating a number of these Communities and/or merging them
with other Communities to form strategically valuable Community
Groups.  In the process, it is not unrealistic to suppose that some
Communities may be terminated without the consent of their leaders.
The OGB should seek to offer reasonable accomodation to the leaders of
such Communities and work with them to find acceptable solutions that
fit the strategic plan.  My hope and expectation is that events of
this type would occur very rarely after the initial realignment.
> - The Draft Constitution says that Community Groups (and in particular,
>   the Community Groups'' Facilitators) are responsible for
"communicating
>   the Community Group''s status to the OGB"; what does this
mean to you?
My understanding was that the Working Group introduced the position of
Facilitator for the purpose of maintaining a single first-line point
of contact for each Community Group.  The OGB should expect each
Community Group to provide its membership list as required by the
Constitution on a regular basis, and for proposing desired changes in
structure or termination (if any).  Beyond that, I believe this
requirement has little meaning to the OpenSolaris community; it seems
to make more sense in the context of an Apache-like organisation in
which many completely disjoint software engineering efforts are
undertaken simultaneously by likewise disjoint groups overseen by the
Foundation.  Since the OGB is not responsible for technical decisions,
it makes little sense to expect Community Groups to provide detailed
information about the work they oversee in the absence of a specific
conflict or other matter requiring the OGB''s attention.  In short, it
makes no sense to sample data which you cannot usefully consume.
> - According to the Draft Constitution, "nominations to the office of
>   Facilitator shall be made by the Core Contributors of the Community
>   Group, but the OGB shall not be limited in their appointment to those
>   nominated."  Under what conditions do you believe that the selection
of
>   a Facilitator would or could fall outside of the nominations made by
>   a Community Group''s Core Contributors?
The only example I can imagine is one in which the designated
Facilitator has a proven history of unreliability or deception.  It
seems unlikely that such an individual would be nominated by a
responsible Community Group, so in practice I doubt this clause will
ever be exercised.
> - According to the Draft Constitution, "non-public discussion related
to
>   the Community Group, such as in-person meetings or private communication,
>   shall not be considered part of the Community Group activities unless or
>   until a record of such discussion is made available via the normal
meeting
>   mechanism."  In your opinion, in the context of a Community Group
like
>   DTrace -- where a majority of the Core Contributors spend eight to ten
>   hours together every work day -- what does this mean?  Specifically, what
>   does it mean to be (or not to be) "considered part of the Community
>   Group activities"?  And in your opinion, what role does the OGB have
in
>   auditing a Community Group''s activities?
I choose to interpret this as a Blue Sky provision, requiring that
important decisions be undertaken in public with the opportunity to
participate for all those whose input might be considered useful.
Since the Constitution provides no definition of "Community Group
activities" other than voting, by implication this works in the same
way as similar provisions in Municipal charters.

In the context of the DTrace Community Group, I take it to mean that
matters which require a Community Group to vote must be presented on a
public list with reasonable opportunity for comment before such a vote
is taken.

Outside of bootstrapping activities around organising and
rationalising Community Groups, I see little proactive role for the
OGB in auditing CG activities.  The OGB should generally handle only
conflicts which cannot be resolved within one or more CGs, and then
only when requested by a party to the conflict.  The Constitution does
preclude the OGB from interfering with a CG''s internal governance.
> Potpourri:
> 
> - Historically, binary compatibility has been very important to Solaris,
>   having been viewed as a constraint on the evolution of technology.
>   However, some believe that OpenSolaris should not have such constraints,
>   and should be free to disregard binary compatibility.  What is your
>   opinion?
Those people are wrong.  Binary compatibility is a great strength, one
which can in nearly all cases be preserved without retarding progress.
To the extent that binary compatibility requires deeper thought on the
part of engineers, it also directly enhances the quality of new work.
Solaris customers praise and appreciate this engineering philosophy
and the results it offers them; we should offer the same benefits to
customers of other distributions as well by maintaining compatibility
and architectural consistency within all recognised consolidations.
Naturally, consumers of OpenSolaris are free to incorporate the
technology into their own products in whatever manner they choose,
including the introduction of changes that violate these constraints.
Such activities are outside the scope of the OGB to regulate.
> - If a third-party were to use and modify DTrace in a non-CDDL''d
system,
>   whose responsibility is it to assure that those modifications are
>   made public?  To put it bluntly:  is enforcing the CDDL an OGB issue?
The answer to the first question is "No one."  Neither use nor
modification triggers the requirement that those modifications be made
distributed in source form (and additions, in particular, need not be
distributed at all).  Only distribution triggers this requirement, and
it is extended only to those to whom binaries are provided.  If such a
party did distribute the binaries containing DTrace, it is that
party''s responsibility to ensure its own compliance with the license
terms.

Enforcement of the CDDL is not an OGB consideration.  The OGB does not
hold any copyrights and has not issued any licenses.  If the OGB is
notified of a license violation, it should (as a group of good
citizens) pass the information along to the copyright holder, if
his/her/its identity is known.  For much of the code in OpenSolaris
including DTrace, that copyright holder is Sun Microsystems, Inc.
Further action is at the discretion of the copyright holder.

It may well be within the scope of the OGB''s activities to help
educate contributors about the terms of the CDDL, but such a campaign
would require the OGB to obtain legal counsel.
> - Do you have an opinion on the patentability of software?  In particular,
>   what is the role of the OGB -- if any -- if Sun were to initiate legal
>   proceedings to protect a part of its software patent portfolio that
>   is represented in OpenSolaris?
The OGB does not own software patents (or any other property), and I
have no position on the patentability of software in general.  Sun has
the right to enforce its property rights under the laws of the
countries in which it operates, and the OGB has no authority to
interfere with that enforcement.  Since community members who adhere
to the terms of the licenses offered for OpenSolaris have limited (but
adequate for all uses permitted under the CDDL) licenses to patents
represented within that body of code, there is no reason for the OGB
to worry about this.  If such an event were to occur, the OGB might
profitably offer a simple statement to this effect, clarifying the
facts of the situation and denying incorrect rumours.  Whether such an
action would be necessary or appropriate would depend on the specific
circumstances.
> - When you give public presentations, do you run OpenSolaris on your
laptop?
>   Have you ever given a public demonstration of OpenSolaris technology?
Yes, I use OpenSolaris exclusively with the exception of
interoperability testing.  Yes, I have demonstrated new technology in
Solaris 10 (now in OpenSolaris as well) at OSCON in 2004 and 2005, and
the early OpenSolaris build system technology at SVOSUG in 2005.
> - And an extra credit question:  Have you ever used DTrace?  When did you
>   most recently use it, and why?  The answers "just now" and
"to answer
>   this question" will be awarded no points. ;)
Yes, I''ve used DTrace.  I most recently used it earlier this week
while diagnosing the behaviour of two machines in an HA cluster.  I''ve
also written a (never-integrated) System V IPC provider for
OpenSolaris and introduced USDT probes to enhance the observability of
several aspects of daemon behaviour.

-- 
Keith M Wesolowski		"Sir, we''re surrounded!" 
FishWorks			"Excellent; we can attack in any direction!"

Hi Bryan & the DTrace Community,
> Licensing Questions:
> 
> - DTrace is one of only a small handful of OpenSolaris technologies that
>   has actually been incorporated into other operating systems.  Thus,
>   your position on dual-licensing is very important to us; what is your 
>   position on dual-licensing in general?
In general, I have yet to see a project where dual-licensing has helped.
It didn''t work for OpenOffice certainly.  All licenses have evils -
dual
licensing only doubles the evils.

Dual-licensing only increases the likelihood of driving a rift in
between the community.  Distributors could choose to license their
changes solely under once license thus rendering it unavailable to the
distributors who license under the other.  It seems to me that it would
be doing a major disservice to the OpenSolaris Community.
> - Do you agree with the conclusions and decrees of CAB/OGB Position 
>   Paper # 20070207?
In general: yes.  I agree with the spirit of the paper.
In practice: I abhor the fact that the CAB felt the need to publish such
a paper.  I felt that it was unwarranted and unnecessary.
> - The OGB is responsible for the representation of OpenSolaris to 
>   third parties.  If a third party were to inquire about incorporating
>   DTrace into a GPL''d Program, what would be your response or
position?
Depends... has OpenSolaris become GPL?  If it hasn''t, then I would say
"piss off".  Though perhaps more nicely than that.

In all seriousness, I believe simplicity is better.  Simplifying the
OGB, simplifying licensing, and simplifying the community is a good
thing.  Part of this means an unwavering line-of-action on things where
people want "exceptions".  

Granting license exceptions is a slipper path we should not be walking
down.
> Constitutional Questions:
> 
> - DTrace is currently a Community Group, but some could argue that it would
>   make more sense for DTrace to be a Project in (say) the Observability
>   Community Group.  In your mind, what is (or should be) the difference
>   between a Community Group and a Project -- and where should DTrace fall?
A project is a means to which develop code in a project-separate
repository until such time that projects are ready to integrate into
larger consolidations.  Communities are a more abstract grouping of
goals, ideas and projects.  It seems to me that DTrace should be a project
within the Observability group, yes.  This should not preclude
project-specific mailing lists (which are clearly still valuable after a
project has integrated), nor should it preclude additional related
project work from creating new successor projects.
> - The Draft Constitution says next-to-nothing about where the authority 
>   lies to make or accept changes to OpenSolaris -- only that Projects
>   operate at the behest of Community Groups, and that Community Groups
>   can be "terminated" by the OGB.  In your opinion, where does or
should
>   this authority lie?  And do you believe that the Constitution should
>   or should not make this explicit?  Finally, under what grounds do you
>   believe that a Community Group should be "terminated"?
I''ve always believed the OGB should aspire to "govern less". 
:-P
Community groups should only be terminated if they are actively doing
harm, or if they have gone dormant and have had no activity.

Yes, the OGB should wield that power - since communities are the basis
for governance.
> - The Draft Constitution says that Community Groups (and in particular,
>   the Community Groups'' Facilitators) are responsible for
"communicating
>   the Community Group''s status to the OGB"; what does this
mean to you?
It seems like a bug. 
How is the OGB to know that a group has gone dormant (as in the above
example)?  I don''t envision a scenario where the community
groups''
facilitators send the OGB an email saying "hey - we''re not doing
anything here. just letting you know"

I think it perhaps makes sense to have an OGB-designated group of
volunteers (perhaps rotating) who periodically poll the various
communities to make sure things haven''t gone idle.

But in terms of communicating status upstream from a presumably active
community to the OGB, the community group leaders makes the best sense.
i.e.: If the community wanted input from the OGB for some issue, then
it seems the most logical people to communicate said issue/status would
be the leaders.

(or did I misunderstand ''status'' here?)
> - According to the Draft Constitution, "nominations to the office of
>   Facilitator shall be made by the Core Contributors of the Community
>   Group, but the OGB shall not be limited in their appointment to those
>   nominated."  Under what conditions do you believe that the selection
of
>   a Facilitator would or could fall outside of the nominations made by
>   a Community Group''s Core Contributors?
I can''t imagine a scenario where the OGB should want to appoint someone
outside of the nominations provided by the community group''s core
contributors.  It might make suggestions or provide suggestions for
mentors (i.e.: in the event that a newly created community has leaders
who might not have had leadership/facilitator experience and want some
guidance).
> - According to the Draft Constitution, "non-public discussion related
to
>   the Community Group, such as in-person meetings or private communication,
>   shall not be considered part of the Community Group activities unless or
>   until a record of such discussion is made available via the normal
meeting
>   mechanism."  In your opinion, in the context of a Community Group
like
>   DTrace -- where a majority of the Core Contributors spend eight to ten
>   hours together every work day -- what does this mean?  Specifically, what
>   does it mean to be (or not to be) "considered part of the Community
>   Group activities"?  And in your opinion, what role does the OGB have
in
>   auditing a Community Group''s activities?
That''s tough... the reality is the majority number of contributors
currently are employed by Sun - hallway conversations are a fact of
life.  I think the best effort we can make is to try and post some
periodic status/minutes of such conversations if they would involve or
influence the community at large.

Certainly I don''t expect every engineer to post official minutes of all
such technical conversations in the hallway - but if it''s a larger
issue
or topic, then I think the community should be informed.

But overall - I would leave it up to the participants'' judgement as to
what needs to be posted or not.  People will usually do the right thing.
> Potpourri:
> 
> - Historically, binary compatibility has been very important to Solaris,
>   having been viewed as a constraint on the evolution of technology.
>   However, some believe that OpenSolaris should not have such constraints,
>   and should be free to disregard binary compatibility.  What is your
>   opinion?
The OGB should never decide such an issue.
But my personal opinion is that innovation can be had while still
maintaining binary compatibility.  I don''t see that it has stifled us
in
the past (witness DTrace, Zones, ZFS, etc.)

That being said, more radical issues such as package management,
desktop, etc. can and should be given free rein to innovate and break
things if they want.  I don''t see why such work couldn''t
happen on
opensolaris.org, and be layered on top of more "stable" components
like
ON to form the basis for additional distributions to Sun Solaris.
> - If a third-party were to use and modify DTrace in a non-CDDL''d
system,
>   whose responsibility is it to assure that those modifications are
>   made public?  To put it bluntly:  is enforcing the CDDL an OGB issue?
No.  As long as Sun insists on owning the trademarks and protecting the
code, then Sun''s lawyers should police that.  The OGB does not have the
means by which to sic a team of lawyers on someone.  

The OGB is a "governance board".  Not an "enforcement
board".

If OpenSolaris split off into a foundation with its own team of crack
lawyers, then sure - maybe it becomes an issue for the OGB to discuss,
though I would think in the event of such a foundation there would be
a separate "enforcement"/legal team for that.
> - Do you have an opinion on the patentability of software?  In particular,
>   what is the role of the OGB -- if any -- if Sun were to initiate legal
>   proceedings to protect a part of its software patent portfolio that
>   is represented in OpenSolaris?
It entirely depends on whether those legal proceedings involve, harm, or
help the community.  If they are enacted against someone in the
OpenSolaris community and would induce harm against the great community,
then I would think the OGB should stand up and protest or do whatever to
protect the community.

If said proceedings were enacted against someone who was violating the
CDDL or the terms of use for such code involved in the patent, then I
would say the OGB should give a "+1" to Sun legal.  ;-)
> - When you give public presentations, do you run OpenSolaris on your
laptop?
>   Have you ever given a public demonstration of OpenSolaris technology?
Yup.  I''ve alternatively given presentations on Solaris Express
(currently running build 57) and Nexenta (dist-upgraded as of last
week) using StarOffice or OpenOffice.

I give demos predominantly on building and deploying OpenSolaris, how to
develop (i.e.: use the build environment), and on how to use Mercurial.

I''ve also in the past given demos of zones, BrandZ, ZFS, and of course
DTrace.
> - And an extra credit question:  Have you ever used DTrace?  When did you
>   most recently use it, and why?  The answers "just now" and
"to answer
>   this question" will be awarded no points. ;)
Yup, though all my recent work has been involving SCM or tonic
operations stuff, so haven''t had to use it for my day job recently.  My
probably most recent use has been in demos showing people how useful it
can be in drilling down to find an issue (i.e.: how to go from using
larger analysis tools like prstat, vmstat, truss to focusing down on
where to place more specific probes to get more info)

cheers,
steve

-- 
stephen lau // stevel at sun.com | 650.786.0845 | http://whacked.net
opensolaris // solaris kernel development

Bryan Cantrill wrote:
> Licensing Questions:
> 
> - DTrace is one of only a small handful of OpenSolaris technologies that
>   has actually been incorporated into other operating systems.  Thus,
>   your position on dual-licensing is very important to us; what is your 
>   position on dual-licensing in general?
That dual-licensing needs to serve a real benefit to the community.   Doing
it for pure publicity sake serves no one in the long run, and so far
I''ve seen
no compelling arguments for dual-licensing the whole of the CDDL''ed
body of
code (which is far from all of OpenSolaris) that would outweigh the cost and
risks of such an approach.
> - Do you agree with the conclusions and decrees of CAB/OGB Position 
>   Paper # 20070207?
Not entirely.  For OpenSolaris when viewed as a whole, I mostly agree with the
conclusions, though individual parts may fit differently and come to different
answers.

I disagree with the decrees that attempt to impose discussion limits on the
community on this issue.  I agree that much of the discussion that occurred was
a distraction and should be discouraged, but not forcefully eliminated.
Discussing GPLv3 before it is completed is not a complete waste of time, if it
is being done in such a fashion as to provide feedback to the GPLv3 drafting
process.
> - The OGB is responsible for the representation of OpenSolaris to 
>   third parties.  If a third party were to inquire about incorporating
>   DTrace into a GPL''d Program, what would be your response or
position?
That it''s something we''d have to discuss with the DTrace
Community and with
the powers-that-control-the-license at Sun.    Licensing DTrace under a GPL
compatible license (or a BSD compatible license for that matter) could be a
win for DTrace if the communities involved were willing to work together to
make it successful.
> Constitutional Questions:
> 
> - DTrace is currently a Community Group, but some could argue that it would
>   make more sense for DTrace to be a Project in (say) the Observability
>   Community Group.  In your mind, what is (or should be) the difference
>   between a Community Group and a Project -- and where should DTrace fall?
I''ve always thought Projects should be for doing a specified set of
work,
and should have a clear answer to "How will you know when the project is
complete?" but that is at odds with only projects being able to host
source.
The next OGB definitely needs to clarify what the separation is.
> - The Draft Constitution says next-to-nothing about where the authority 
>   lies to make or accept changes to OpenSolaris -- only that Projects
>   operate at the behest of Community Groups, and that Community Groups
>   can be "terminated" by the OGB.  In your opinion, where does or
should
>   this authority lie?  And do you believe that the Constitution should
>   or should not make this explicit?  Finally, under what grounds do you
>   believe that a Community Group should be "terminated"?
I don''t think the Constitution needs to make everything explicit, just
enough
to provide a framework for making the decisions.   Putting too much into the
Constitution just makes it harder to make changes as time goes on.

In reality, each gate needs to have it''s own decision making authority,
and
they''re not going to be identical, just as they aren''t today. 
Gates who
pull large chunks from other bodies, such as JDS, X, and SFW, are going to
operate differently than those containing mostly original work, such as ON
and Install.

However, the constitution only recognizes Communities and Projects, so the OGB
is going to have to somehow either reconcile the mapping of consolidations/gates
to those structures or define a new one for those.
> - The Draft Constitution says that Community Groups (and in particular,
>   the Community Groups'' Facilitators) are responsible for
"communicating
>   the Community Group''s status to the OGB"; what does this
mean to you?
Clearly the current list of Core Contributors & Contributors needs to be
provided when elections or other things requiring validation of membership
occur.   Beyond that, I''m not sure the OGB needs much in the way of
status
reporting other than a ACK/NAK to occasional "are you still alive?"
pings
if it''s not clear from a simple viewing of the community
website/mailing lists.
> - According to the Draft Constitution, "nominations to the office of
>   Facilitator shall be made by the Core Contributors of the Community
>   Group, but the OGB shall not be limited in their appointment to those
>   nominated."  Under what conditions do you believe that the selection
of
>   a Facilitator would or could fall outside of the nominations made by
>   a Community Group''s Core Contributors?
Given that I don''t expect a Facilitator to do much (and honestly until
you
asked, hadn''t noticed that section of the Constitution in the first
couple
readings I did - clearly another few readings are in order), the only reasons
I can see to reject them are if they''ve proven themselves untrustworthy
to
report the contributors lists accurately, or are unable to communicate
effectively with the OGB.   In either case, asking the core contributors to
nominate another would be the first step to resolving the issue, and we should
never get to the case where the OGB selects someone on their own.
> - According to the Draft Constitution, "non-public discussion related
to
>   the Community Group, such as in-person meetings or private communication,
>   shall not be considered part of the Community Group activities unless or
>   until a record of such discussion is made available via the normal
meeting
>   mechanism."  In your opinion, in the context of a Community Group
like
>   DTrace -- where a majority of the Core Contributors spend eight to ten
>   hours together every work day -- what does this mean?  Specifically, what
>   does it mean to be (or not to be) "considered part of the Community
>   Group activities"?  And in your opinion, what role does the OGB have
in
>   auditing a Community Group''s activities?
It means that if all the decisions are made in the hallway, then it''s
not
open development.   And no one expects you to have a public debate over every
design decision or the strategy to fix a bug or implement a new feature, but
if all important decisions are made by the core contributors out of public view,
then there will never be any way to grow more core contributors out of the
community.

I don''t expect the OGB to audit any Community Group activities unless a
group
member complains to the OGB and it is necessary to resolve the situation.
> Potpourri:
> 
> - Historically, binary compatibility has been very important to Solaris,
>   having been viewed as a constraint on the evolution of technology.
>   However, some believe that OpenSolaris should not have such constraints,
>   and should be free to disregard binary compatibility.  What is your
>   opinion?
I believe that much as Solaris has always done, OpenSolaris needs to make
careful decisions about which interfaces need to remain compatible and which
don''t --  (pardon the pun) binary compatibility is not a simple binary
issue
of yes or no, which is why the ARC has 4 levels of interface stability.
The most important part remains making it clear what developers can and cannot
rely on, and designing the system to be likely to meet their needs.
> - If a third-party were to use and modify DTrace in a non-CDDL''d
system,
>   whose responsibility is it to assure that those modifications are
>   made public?  To put it bluntly:  is enforcing the CDDL an OGB issue?
Since the OGB isn''t the copyright holder and doesn''t represent
the copyright
holder, I don''t think we can do any more than inform Sun & make a
recommendation that Sun attempt a peaceful release of code before
resorting to legal mechanisms.
> - Do you have an opinion on the patentability of software?  In particular,
>   what is the role of the OGB -- if any -- if Sun were to initiate legal
>   proceedings to protect a part of its software patent portfolio that
>   is represented in OpenSolaris?
The OGB''s role would be to advise Sun on what the probable fallout
would be
in relationships with the OpenSolaris community and other communities (which
for a lawsuit initiated by Sun, would probably be a very negative community
reaction in most cases).
> - When you give public presentations, do you run OpenSolaris on your
laptop?
>   Have you ever given a public demonstration of OpenSolaris technology?
I''ve given public presentations at the Desktop Developer''s
Conference [1] in
2005 & 2006, both given on Solaris Nevada, since it was required for the
demo
parts of my talks (the Xserver DTrace provider in 2005, Solaris Trusted
Extensions Xserver/desktop in 2006).
> - And an extra credit question:  Have you ever used DTrace?  When did you
>   most recently use it, and why?  The answers "just now" and
"to answer
>   this question" will be awarded no points. ;)
Yes - I created the Xserver DTrace provider [2] and integrated it to both
Solaris and the X.Org community repository.   My most recent use was when
a certain Mr. Cantrill e-mailed to ask why shift-click wasn''t working
in
nv_58 and dtrace was used to show the kernel level drivers were buffering it.
OpenSolaris bugid 6526932 [3] has a copy of the script (hopefully no points
are taken off for poor dtrace scripting style).

[1] http://www.desktopcon.org/
[2] http://people.freedesktop.org/~alanc/dtrace/
[3] http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6526932

[Wow.   That was an exhausting list of questions.   Hopefully enough of them
  answered general concerns that we don''t get many more lists that
long.]

-- 
	-Alan Coopersmith-           alan.coopersmith at sun.com
	 Sun Microsystems, Inc. - X Window System Engineering

[ resending to dtrace-discuss, as it was bounced back to me earlier
today due to dtrace-discuss being "reject" instead of
"moderate" for
unknown senders ]

Bryan Cantrill wrote:
> OGB Candidates,
>
> We in the DTrace Community have a couple of specific questions and 
> concerns that will affect how our Core Contributors vote in the upcoming
> OGB elections.  To see where you stand on the issues that are important
> to us, we would like you to answer the following questions before
> campaigning ends on Sunday night.  Please cc: dtrace-discuss at
opensolaris.org
> on your answers, and thanks in advance for helping us make an informed
> decision!
>
> 	- Bryan
>
> --------------------------------------------------------------------------
> Bryan Cantrill, Solaris Kernel Development.       http://blogs.sun.com/bmc
>
>
> Licensing Questions:
>
> - DTrace is one of only a small handful of OpenSolaris technologies that
>   has actually been incorporated into other operating systems.  Thus,
>   your position on dual-licensing is very important to us; what is your 
>   position on dual-licensing in general?
>   
I worry that dual-licensing will create schism in the community.  That
there will be created some "killer" feature that is _not_ dual
licensed,
but will be GPLv2 (or worse! GPLv3).  This ultimately would damage the
ability of others to make products based on OpenSolaris.... (including
Sun!)  Some things have good reasons for not being able to be open
source, even DRM.  (DRM is not _necessarily_ harmful in my opinion,
though it can be used for evil.)

There are some individuals (call them "zealots") that believe
licensing
is a tool for social activism.  I''m not one of those.  To me, the
choice
of software license should be made pragmatically.

But I also like to see software as widely reused as possible, and the
choice of license can help that.  That''s why my own ethernet drivers
are
licensed under a permissive BSD license.  I would not object to certain
pieces of OpenSolaris (such as DTrace) dual-licensing where that helps
other projects like Linux reuse those bits of code.  (In those cases,
perhaps a GPLv2 or BSD dual license would be nice.)  I''d look at those
on a case by case basis, I think.

I think one potential issue that could happen is if we believe that
there is likely to be major innovation in the technology (say major
enhancements to ZFS for example) driven in an external community, but
done under the terms of a secondary license.  It would be nice to be
able to make sure we could bring back in such enhancements into Solaris
under the terms of the CDDL.  I''m not sure how best to achieve that.

Some major critical components of Solaris (like the kernel itself)
should probably remain under CDDL, thereby avoiding the potential for a
major license-driven fork in Solaris itself.

I very much like the file-driven nature of CDDL, as it minimize the
viral influence of GPL.  (Some "overzealous" linux authors have taken,
IMO, the viral nature of the GPL (v2 in this case) a bit far ... Donald
Becker for example believes that it is illegal to create derived NIC
drivers from his Linux code unless they are going to be used in a
GPLv2''d OS.  Even if the drivers are separate loadable modules and
source is released.)
> - Do you agree with the conclusions and decrees of CAB/OGB Position 
>   Paper # 20070207?
>   
Generally, yes.  As noted, I can imagine dual licensing (or even
sublicensing) some components.
> - The OGB is responsible for the representation of OpenSolaris to 
>   third parties.  If a third party were to inquire about incorporating
>   DTrace into a GPL''d Program, what would be your response or
position?
>   
I think that would depend a lot on how the DTrace community felt.  If
the desire is to enable this, and the concern about being able to
reintegrate externally developed enhancements back under CDDL can be
met, then I see no problem with treating this as a special case, as
described above.

The details of how best to achieve that might require the consultation
services of a lawyer. :-)
> Constitutional Questions:
>
> - DTrace is currently a Community Group, but some could argue that it would
>   make more sense for DTrace to be a Project in (say) the Observability
>   Community Group.  In your mind, what is (or should be) the difference
>   between a Community Group and a Project -- and where should DTrace fall?
>   
I think nearly every involved with DTrace is also concerned with
Observability.  I think this probably represents a strict subset of
Observability, so it makes sense to bring it under the larger umbrella.
However, this is based on my understanding of the communities ... if the
experts in the community don''t believe there is a subset relationship,
then leave it the way it is.

> - The Draft Constitution says next-to-nothing about where the authority 
>   lies to make or accept changes to OpenSolaris -- only that Projects
>   operate at the behest of Community Groups, and that Community Groups
>   can be "terminated" by the OGB.  In your opinion, where does or
should
>   this authority lie?  And do you believe that the Constitution should
>   or should not make this explicit?  Finally, under what grounds do you
>   believe that a Community Group should be "terminated"?
>   
"Changes to Solaris" (meaning the source code) have historically been
approved by ARCs.  I think this has worked well for Solaris in the past,
and that somehow we need to find a way to allow external participants to
play with ARC (and even stand as ARC members), and figure out how to
define the relationship between OGB and ARCs.

As far as Community Groups goes, I would say that OGB should take a
laissez faire approach as much as possible.

We already have a process for creating them.

I believe that tearing them down should only be done with the unanimous
approval of either the identified leadership of the Community Group, or
the unanimous approval of OGB.   I''d support ammending the constitution
to this effect.

Note that the case of a reorg (like moving a group to a project) can be
handled the Group''s leadership.  The OGB should only intervene in cases
where the leadership disappears, or the group goes off in the weeds and
starts acting in a manner that is detrimental to OpenSolaris at large.
> - The Draft Constitution says that Community Groups (and in particular,
>   the Community Groups'' Facilitators) are responsible for
"communicating
>   the Community Group''s status to the OGB"; what does this
mean to you?
>   
I think that means several different things:

    - First, is the group alive and healthy? :-)  This helps OGB to
identify Groups that are no longer needed, perhaps because their goals
have been achieved and everyone else has moved on.

    - Second, status reports about the project help OGB know what is
going on with Solaris, so that this information can be forwarded to
relevant marketing groups, accumulated in an overall "State of
OpenSolaris" report, etc.

    - Third, if a Group needs something that they''re not getting (for
example, system resources, or marketing help, or recruiting help), OGB
can be called in to help them get what they need (or, in the event that
there is no solution, act as a final arbiter.)  OGB knowing what needs
are not being met can also help when trying to determine how to spend
budget, or when asking sponsors for help.

    - Fourth, if a conflict arises between groups or between a group and
a member, OGB stands as sort of the final arbiter.  Hopefully this will
rarely come into play.
> - According to the Draft Constitution, "nominations to the office of
>   Facilitator shall be made by the Core Contributors of the Community
>   Group, but the OGB shall not be limited in their appointment to those
>   nominated."  Under what conditions do you believe that the selection
of
>   a Facilitator would or could fall outside of the nominations made by
>   a Community Group''s Core Contributors?
>   
When no nominations are made, or the nominations are unsuitable for some
particular reason.  Its hard to imagine this occurring, but for example,
if the nominee doesn''t speak English, cannot interact due to timezone
differences (imagine all of the OGB members being in PST, but the
nominees being in China, for example) or has somehow violated acceptable
behaviors, or simply declines the nomination, then OGB could intervene.
I would consider this a last resort, however ... OGB should make every
effort to choose from one of the nominations from the Group.

I can see a situation arising where the OGB needs to work with someone
from a Community Group where there are legal considerations as well.
For example, it would be hard to deal with foreign nationals from
embargoed countries, and there could be issues in some cases where a
Facilitator might need to sign an NDA or something.  (That seems a bit
of stretch at the moment, but not entirely out of the realm of
possibilities...)

I think there is an unstated opportunity for the OGB to provide guidance
to the Community Groups about what kinds of nominations might not be
acceptable, before the nominations are actually made.
> - According to the Draft Constitution, "non-public discussion related
to
>   the Community Group, such as in-person meetings or private communication,
>   shall not be considered part of the Community Group activities unless or
>   until a record of such discussion is made available via the normal
meeting
>   mechanism."  In your opinion, in the context of a Community Group
like
>   DTrace -- where a majority of the Core Contributors spend eight to ten
>   hours together every work day -- what does this mean?  Specifically, what
>   does it mean to be (or not to be) "considered part of the Community
>   Group activities"?  And in your opinion, what role does the OGB have
in
>   auditing a Community Group''s activities?
>   
Heh.  This is going to happen a lot with Sun.  But not just at Sun.  For
example, at Tadpole, we are often working on stuff that may ultimately
be contributed back to OpenSolaris, but that doesn''t mean we do all of
our development in public.

The activities that matter in this respect are not code development.
Code development is going to take place offline naturally.  People will
continue to interact in whatever way is most natural to facilitate that.

The activities that do matter are conflict resolution, issues which
require a vote or approval of the leadership (such as nominating a
facilitator, identifying core contributors, etc.)  In the event that
there is disagreement about the direction (perhaps a technical
direction) that requires a vote to intervene, then the above public
mechanisms are needed.

One group that is interesting in this regard is the ARCs.  I''d like
more
of those discussions to be public, and involve more external community
representation.
> Potpourri:
>
> - Historically, binary compatibility has been very important to Solaris,
>   having been viewed as a constraint on the evolution of technology.
>   However, some believe that OpenSolaris should not have such constraints,
>   and should be free to disregard binary compatibility.  What is your
>   opinion?
>   
Binary compatibility is an important feature of Solaris.  I do think
there are cases to be made for letting some "ancient" bits fall away
though ... most of what is in /usr/ucb is a good example of what should
be allowed to go away ... nobody has released a product which relied on
that stuff for years.  (Or at least nobody _should_ have.)

> - If a third-party were to use and modify DTrace in a non-CDDL''d
system,
>   whose responsibility is it to assure that those modifications are
>   made public?  To put it bluntly:  is enforcing the CDDL an OGB issue?
>   
Initially, the responsibility is Sun''s, because Sun owns the copyright.
OGB cannot act on Sun''s behalf unless the copyright is assigned, or
some
other legal powers are granted to it.

I think it might be reasonable for OGB to request Sun to enforce the
CDDL in case of abuse where OGB learns of it.
> - Do you have an opinion on the patentability of software?  In particular,
>   what is the role of the OGB -- if any -- if Sun were to initiate legal
>   proceedings to protect a part of its software patent portfolio that
>   is represented in OpenSolaris?
>   
Software patents can be useful and meaningful.  One example I like to
use are the RSA patents.  These represent real innovation on the part of
the inventors, and granting them a time-limited monopoly in exchange for
making the information free later seems fair.

The times seem long, but in the case of the patents associated with RSA
and public key encryption in general, these have expired, and now the
entire world is the benefit.  I much prefer this to systems where the
details were protected as Trade Secrets and nobody benefits.

I''d like to see more careful review given to patents (not just
software,
but especially to them), and I''d like to see the time a software patent
is in force reduced.

In any case, OGB doesn''t have a role here, unless Sun goes after _it_,
which seems very unlikely.
> - When you give public presentations, do you run OpenSolaris on your
laptop?
>   Have you ever given a public demonstration of OpenSolaris technology?
>   
I rarely give demonstrations, but I run Solaris exclusively on my
computers.  The last time I presented anything, it was on a laptop
running Solaris 9.  At the moment I''m loading up b59 on my SPARC
laptop.  I have no x86 laptops of my own (that will soon change, and it
_will_ run OpenSolaris.)

I''ve run Solaris in some form or another on my desktop machines since
1995 (I started with Solaris 2.4).  Prior to that I used SunOS 4.x.
(I''m sorry, I just can''t bring myself to think of SunOS 4.1.3
as Solaris
1.0).

Most of my systems at home are loaded with Solaris 10u3.  The exceptions
are the wife''s computer (she has to use ActiveX for a website she uses
for here insurance agency... yech) and the kids'' computer (they like
Windows RealArcade games.)
> - And an extra credit question:  Have you ever used DTrace?  When did you
>   most recently use it, and why?  The answers "just now" and
"to answer
>   this question" will be awarded no points. ;)
>   
I''ve only played with it a tiny bit ... I''m still very heavily
an
mdb/kmdb user (and before that adb/kadb).  I did use it about a month
ago to do some call tracing in one of my ethernet drivers... and I was
impressed.  However, I often have to support pre-Soalris 10 (usually
Solaris 8), so I''ve not spent as much time working with it as I would
have liked.

    -- Garrett

Bryan Cantrill wrote On 03/08/07 10:01 AM,:
> OGB Candidates,
> 
> We in the DTrace Community have a couple of specific questions and 
> concerns that will affect how our Core Contributors vote in the upcoming
> OGB elections.  To see where you stand on the issues that are important
> to us, we would like you to answer the following questions before
> campaigning ends on Sunday night.  Please cc: dtrace-discuss at
opensolaris.org
> on your answers, and thanks in advance for helping us make an informed
> decision!
> 
> 	- Bryan
> 
> --------------------------------------------------------------------------
> Bryan Cantrill, Solaris Kernel Development.       http://blogs.sun.com/bmc
> 
> 
> Licensing Questions:
> 
> - DTrace is one of only a small handful of OpenSolaris technologies that
>   has actually been incorporated into other operating systems.  Thus,
>   your position on dual-licensing is very important to us; what is your 
>   position on dual-licensing in general?
I would advocate for sticking with the CDDL.
> 
> - Do you agree with the conclusions and decrees of CAB/OGB Position 
>   Paper # 20070207?
Yes.
> 
> - The OGB is responsible for the representation of OpenSolaris to 
>   third parties.  If a third party were to inquire about incorporating
>   DTrace into a GPL''d Program, what would be your response or
position?
I would get additional information about what exactly they want to do,
why, and how, and communicate it back to them so they know that I am
interested in helping them and to be sure I understand the inquiry.
Then, I''d probably agree to take the information they gave me back to
the appropriate leaders and possibly license legal reps. to find out
what issues there are because I know there are potential issues with
this, but I cannot detail those myself. Then, I''d put a response
together that describes whether this can be done and how and if not,
what are the issues and communicate that to all parties.
> 
> Constitutional Questions:
> 
> - DTrace is currently a Community Group, but some could argue that it would
>   make more sense for DTrace to be a Project in (say) the Observability
>   Community Group.  In your mind, what is (or should be) the difference
>   between a Community Group and a Project -- and where should DTrace fall?
I think Communities should be umbrella groups that make high-level
decisions about guidelines, best practices, the ''rules of
engagement''
for collaboration. Communities should also provide development tool
recommendations and requirements, if not tool chains themselves, to
enable projects to succeed by leveraging infrastructure and subject
matter expertise. For example, a community might provide help to a new
project to learn about some aspect of open work that they''ve got some
experience with already, to avoid duplication of efforts and create
relationships between new and experienced developers.

Projects should be able to go to Communities when they need something to
 be successful, when they need some advice to get through a blocking
issue, or when they are just getting started and need to learn how to
collaborate on OpenSolaris. Projects should have the source code and
make the day-to-day decisions about changes to the sources, validity
checking, putback shepherding, and gatekeeping. So, I think DTrace would
be a Project in the Observability community, if we''d had the Project
functionality from the start.
> 
> - The Draft Constitution says next-to-nothing about where the authority 
>   lies to make or accept changes to OpenSolaris -- only that Projects
>   operate at the behest of Community Groups, and that Community Groups
>   can be "terminated" by the OGB.  In your opinion, where does or
should
>   this authority lie?  And do you believe that the Constitution should
>   or should not make this explicit?  Finally, under what grounds do you
>   believe that a Community Group should be "terminated"?
I think the authority to make or accept changes is with the ARCs today
and should stay there. I don''t think the Constitution needs to make
this
explicit. I would only advocate for termination of a Community by the
OGB if there was overwhelming proof that the Community was destructive
and detrimental to the OpenSolaris project.
> 
> - The Draft Constitution says that Community Groups (and in particular,
>   the Community Groups'' Facilitators) are responsible for
"communicating
>   the Community Group''s status to the OGB"; what does this
mean to you?
I think this should just be a way to keep lines of communication open
between OGB and Communities. If an issue that crosses Communities or
effects multiple Communities emerges from this status reporting the OGB
might have something they need to do to help out.
> 
> - According to the Draft Constitution, "nominations to the office of
>   Facilitator shall be made by the Core Contributors of the Community
>   Group, but the OGB shall not be limited in their appointment to those
>   nominated."  Under what conditions do you believe that the selection
of
>   a Facilitator would or could fall outside of the nominations made by
>   a Community Group''s Core Contributors?
I would not advocate for appointment of a Community facilitator outside
of the nominations made by the Core Contributors of a
Community.
> 
> - According to the Draft Constitution, "non-public discussion related
to
>   the Community Group, such as in-person meetings or private communication,
>   shall not be considered part of the Community Group activities unless or
>   until a record of such discussion is made available via the normal
meeting
>   mechanism."  In your opinion, in the context of a Community Group
like
>   DTrace -- where a majority of the Core Contributors spend eight to ten
>   hours together every work day -- what does this mean?  Specifically, what
>   does it mean to be (or not to be) "considered part of the Community
>   Group activities"?  And in your opinion, what role does the OGB have
in
>   auditing a Community Group''s activities?
This seems like a clause to help folks who are outside of geographic
proximity from being forced into some decision made by those who are.
So, in your example of DTrace Core Contributors working all day
together, that is certainly Community Group activity, but if that
activity results in a decision or action that impacts the rest of your
Contributors, and you don''t communicate it publicly, then you
can''t
expect those Contributors to abide. I think the only auditing role OGB
has is the case where a problem or escalation comes to them for
resolution.
> 
> Potpourri:
> 
> - Historically, binary compatibility has been very important to Solaris,
>   having been viewed as a constraint on the evolution of technology.
>   However, some believe that OpenSolaris should not have such constraints,
>   and should be free to disregard binary compatibility.  What is your
>   opinion?
I would advocate for retaining binary compatibility
constraint.
> 
> - If a third-party were to use and modify DTrace in a non-CDDL''d
system,
>   whose responsibility is it to assure that those modifications are
>   made public?  To put it bluntly:  is enforcing the CDDL an OGB issue?
Enforcement of CDDL is not an OGB issue.
> 
> - Do you have an opinion on the patentability of software?  In particular,
>   what is the role of the OGB -- if any -- if Sun were to initiate legal
>   proceedings to protect a part of its software patent portfolio that
>   is represented in OpenSolaris?
I haven''t developed a mature opinion on this, but I don''t
think there is
a role for the whole OGB in legal proceedings initiated by Sun. There
might be for an individual who sits on the OGB and who has knowledge
pertinent to a case, but I don''t see a scenario where the whole OGB
would be involved in proceedings.
> 
> - When you give public presentations, do you run OpenSolaris on your
laptop?
Yes.
>   Have you ever given a public demonstration of OpenSolaris technology?
Yes, if the Starter Kit counts.
> 
> - And an extra credit question:  Have you ever used DTrace?  When did you
>   most recently use it, and why?  The answers "just now" and
"to answer
>   this question" will be awarded no points. ;)
Yes, I tried out some really simple stuff when I was working on the
Intro to Operating Systems document last year, to see if the exercise
steps made sense.
>

Hi ... some comments below:


Bryan Cantrill wrote On 03/09/07 03:01,:
> OGB Candidates,
> 
> We in the DTrace Community have a couple of specific questions and 
> concerns that will affect how our Core Contributors vote in the upcoming
> OGB elections.  To see where you stand on the issues that are important
> to us, we would like you to answer the following questions before
> campaigning ends on Sunday night.  Please cc: dtrace-discuss at
opensolaris.org
> on your answers, and thanks in advance for helping us make an informed
> decision!
> 
> 	- Bryan
> 
> --------------------------------------------------------------------------
> Bryan Cantrill, Solaris Kernel Development.       http://blogs.sun.com/bmc
> 
> 
> Licensing Questions:


I''m not an expert on licenses and wouldn''t argue strongly on
issues
outside my field. However, if and when presented with such issues, my 
position would be to try to make sure all avenues are explored so an 
intelligent decision can be implemented. As I develop some expertise in 
an area, I''ll assert my opinions more in that area.


> - DTrace is one of only a small handful of OpenSolaris technologies that
>   has actually been incorporated into other operating systems.  Thus,
>   your position on dual-licensing is very important to us; what is your 
>   position on dual-licensing in general?


If dual licensing is formally (and professionally) proposed, it would 
need to be considered carefully for no other reason than it would 
introduce significant complexity into the community. License issues are 
difficult enough when using just one, so I see no need for two unless 
there are overwhelming reasons to do so. It took Sun approximately a 
year to consider all the licensing options for OpenSolaris and to settle 
on CDDL. Similar thought processes need to go into any licensing change 
in the future.

I would certainly prefer that future licensing decisions be a community 
effort, but ultimately this is probably Sun''s decision. That''s
why the
OGB needs to establish a close and open relationship with the community 
/and/ with Sun executives so the decision-making process is clear. Also, 
everyone benefits if community members feel their opinions have been 
considered and Sun feels it can make decisions with community support. 
This is the essence of open communications and it''s required to 
establish trust -- something we currently do not have.


> - Do you agree with the conclusions and decrees of CAB/OGB Position 
>   Paper # 20070207?


No. And I said so here:

http://blogs.sun.com/jimgris/entry/the_ogb_s_decree
http://opensolaris.org/jive/message.jspa?messageID=92664&tstart=0

I seem to hold a minority opinion on this, but I actually feel more 
strongly about it now than I did last month.


> - The OGB is responsible for the representation of OpenSolaris to 
>   third parties.  If a third party were to inquire about incorporating
>   DTrace into a GPL''d Program, what would be your response or
position?


The licenses either mix or they do not. I''m not sure I see a formal 
administrative role here for the OGB other than to act as a 
communications and consensus building body in the case of a conflict 
that affects the OpenSolaris community.


> Constitutional Questions:
> 
> - DTrace is currently a Community Group, but some could argue that it would
>   make more sense for DTrace to be a Project in (say) the Observability
>   Community Group.  In your mind, what is (or should be) the difference
>   between a Community Group and a Project -- and where should DTrace fall?


I''m fine with the current notion that projects are spaces where 
development work gets done and communities are larger, social groups of 
people addressing meta issues within their community as well as the 
details of specific projects. I would like to see a more clear 
representation of that in the site structure, though, because this issue 
has been confusing to many people. And, obviously, we''ll need to do a 
bit of re-organizing of communities and projects as many have suggested. 
DTrace seems to me to be a project under this system, but this would 
have to be more fully discussed.


> - The Draft Constitution says next-to-nothing about where the authority 
>   lies to make or accept changes to OpenSolaris -- only that Projects
>   operate at the behest of Community Groups, and that Community Groups
>   can be "terminated" by the OGB.  In your opinion, where does or
should
>   this authority lie? 

I think the intent of the Constitution is to push development decisions 
to those who are closest to the code -- the projects and the communities 
managing or overseeing projects. I''m not going to take a strong
position
here since I don''t touch code or development processes. However, I 
certainly agree with this intent and would like to see this relationship 
more clearly articulated in the governance.


>  And do you believe that the Constitution should
>   or should not make this explicit?  Finally, under what grounds do you
>   believe that a Community Group should be "terminated"?

Obvious example such as abuse, lack of activity, lack of involvement or 
communication with the OGB, etc.

> - The Draft Constitution says that Community Groups (and in particular,
>   the Community Groups'' Facilitators) are responsible for
"communicating
>   the Community Group''s status to the OGB"; what does this
mean to you?


Each Community Group needs get involved in governance or they''ll 
abdicate their responsibility and eventually undermine their viability.


> - According to the Draft Constitution, "nominations to the office of
>   Facilitator shall be made by the Core Contributors of the Community
>   Group, but the OGB shall not be limited in their appointment to those
>   nominated."  Under what conditions do you believe that the selection
of
>   a Facilitator would or could fall outside of the nominations made by
>   a Community Group''s Core Contributors?


I''m not sure I agree with this provision. The Community Groups ought to
be able to select their own representatives. If they can''t do that, 
their viability may be questionable.


> - According to the Draft Constitution, "non-public discussion related
to
>   the Community Group, such as in-person meetings or private communication,
>   shall not be considered part of the Community Group activities unless or
>   until a record of such discussion is made available via the normal
meeting
>   mechanism."  In your opinion, in the context of a Community Group
like
>   DTrace -- where a majority of the Core Contributors spend eight to ten
>   hours together every work day -- what does this mean?  Specifically, what
>   does it mean to be (or not to be) "considered part of the Community
>   Group activities"?  And in your opinion, what role does the OGB have
in
>   auditing a Community Group''s activities?


If we are an open and global community then conversations, decisions, 
and development ought to be open as well. I realize that our "opening"
is taking some time, but I think the intent here is to simply promote 
the notion of open development and open communications and discourage 
groups from taking too many things private. I think we need to be 
reasonable here, though. There is value in proximity, but for a 
community to grow those groups should seek to be as open as possible. 
The OGB can''t audit this because the community is too big and too 
dynamic, and I don''t see any clear enforcement mechanism in place. 
However, the OGB probably could address complaints of developers who 
feel that too many decisions are made on closed lists or via private 
emails threads.


> Potpourri:
> 
> - Historically, binary compatibility has been very important to Solaris,
>   having been viewed as a constraint on the evolution of technology.
>   However, some believe that OpenSolaris should not have such constraints,
>   and should be free to disregard binary compatibility.  What is your
>   opinion?


Sun ships a tested and supported product and the engineering ethic that 
goes into the product is a distinguishing characteristic of the 
OpenSolaris Community. It seems to me that OpenSolaris offers project 
spaces for experimentation and before integration. We have mix of both.

I cast it that way because I generally believe that at its core the 
OpenSolaris Community /starts/ with Sun''s Solaris engineers and grows 
outward to embrace non-Sun developers. As that process occurs, the 
non-Sun developers learn the development process and the code, etc, and 
they certainly benefit as they begin to contribute. Actually, they''d be
helping to create some of these processes. Simultaneously, as new 
developers get involved, Sun''s Solaris engineers benefit from that 
external involvement and talent and the flow of new ideas. Somewhere in 
the middle a new community develops, one that is hopefully better than 
before and one that offers the highest quality engineering as well as 
spaces for free experimentation and faster movement. We have to find a 
balance between both, but both are important.


> - If a third-party were to use and modify DTrace in a non-CDDL''d
system,
>   whose responsibility is it to assure that those modifications are
>   made public? To put it bluntly:  is enforcing the CDDL an OGB issue?


If DTrace code is modified and distributed in a product without the 
source being published, Sun would have to enforce any license 
requirements when those requirements are triggered. The OGB could -- and 
probably should -- get involved to the extent that it can help 
communicate the community''s concerns, but I don''t see an
enforcement
mechanism. That seems to be Sun''s corporate responsibility.


> - Do you have an opinion on the patentability of software?  In particular,
>   what is the role of the OGB -- if any -- if Sun were to initiate legal
>   proceedings to protect a part of its software patent portfolio that
>   is represented in OpenSolaris?

It''s easy to say I don''t agree with patents on software in the
world of
open source, but I''d be hard pressed to argue the point without more 
direct experience. I simply don''t have expertise in this area.

> - When you give public presentations, do you run OpenSolaris on your
laptop?
>   Have you ever given a public demonstration of OpenSolaris technology?


I''ve given many public presentations, and I run Solaris for those 
presentations. I don''t do demos of specific OpenSolaris technologies, 
though, because I''m not a developer and I don''t mix down in
the code. I
could certainly be trained to give a demo or two here and there, but
I''d
shy away from this because my credibility would be, well, zero.


> - And an extra credit question:  Have you ever used DTrace?  When did you
>   most recently use it, and why?  The answers "just now" and
"to answer
>   this question" will be awarded no points. ;)

Never used it. But I''ve heard it''s quite good. :)


Jim

On Thursday 08 March 2007 10:01 am, Bryan Cantrill
wrote:
> - DTrace is one of only a small handful of OpenSolaris technologies that
>   has actually been incorporated into other operating systems.  Thus,
>   your position on dual-licensing is very important to us; what is your
>   position on dual-licensing in general?
Dual licensing is something that can''t be avoided in today''s
open source
climate. This is one of the reasons CDDL exists today.

However, it''s not up to the OGB to decide license issues, and I feel
the OGB
should offer advice where applicable. However, for proper legal advice in 
regard to licensing, if there are concerns, seek proper legal counsel to 
understand ramifications that may result in such dual license issues.

Guidance and advice should be given, but the OGB is not a legal counsel and 
cannot be held liable for any decisions in regard to such.

Each case can and will be different, and to give a general comment based on a 
broad question as above, is difficult at best to answer.
> - Do you agree with the conclusions and decrees of CAB/OGB Position
>   Paper # 20070207?
I definitely agree with most of it, and emphasis should be placed on the first 
point of their conclusion, that discussing the GPLv3 is pre-mature as the 
license does not exist at this time.

A heavy emphasis should also be placed on their last point, being that there 
are other higher priorities that could create more acceptance amongst 
developers.

Speaking for myself, I do not feel the license is an OGB issue, but educating 
and helping people to understand our license is more key, and Sun has done a 
poor job at doing that to date, IMO. The OGB can probably fill some of the 
void that has been left to date.
> - The OGB is responsible for the representation of OpenSolaris to
>   third parties.  If a third party were to inquire about incorporating
>   DTrace into a GPL''d Program, what would be your response or
position?
I would have to be honest and mention that there are incompatibilities between 
CDDL and GPL, and I view the GPL as being a CDDL incompatible license. This 
is a vauge question, which doesn''t state the version of GPL, which
could
change things in the future, but GPLv2 is your primary question I suspect.

This is a situation where I believe the OGB needs to point out that the CDDL 
is an OSI approved license, and point out some of the merit of why it was 
created to begin with, to protect them specific and not Sun.

I think it''s important to point out *ALWAYS* in these cases that the
OGB is
not a legal counsel in this regard, and while it can offer advice in regards 
to licensing, proper legal counsel should be sought out by them.
> - DTrace is currently a Community Group, but some could argue that it would
>   make more sense for DTrace to be a Project in (say) the Observability
>   Community Group.  In your mind, what is (or should be) the difference
>   between a Community Group and a Project -- and where should DTrace fall?
For me, communites are bodies of projects and interest based around a 
technology. Projects are more of a solution. In the case of DTrace, it is 
more than a solution, it is a technology which is multi-faceted. I would 
discourage putting this into another community.
> - The Draft Constitution says next-to-nothing about where the authority
>   lies to make or accept changes to OpenSolaris -- only that Projects
>   operate at the behest of Community Groups, and that Community Groups
>   can be "terminated" by the OGB.  In your opinion, where does or
should
>   this authority lie?  And do you believe that the Constitution should
>   or should not make this explicit?  Finally, under what grounds do you
>   believe that a Community Group should be "terminated"?
It is difficult to project which communities will flourish, and I believe 
OpenSolaris will need to experiment and find out by some trial and error 
which communities thrive and which don''t. It is not the intent for the
OGP to
use their power as a prevention of communities.

As an example, here''s a controversial point I will use to show this.
KDE is a
desktop project which is needed in our community. If OGB should sweep the KDE 
community under the rug, to prevent our community from developing it on 
OpenSolaris just because Sun advocates and develops on GNOME, that would be 
wrong. If the KDE community was created and no participation occured for an 
entire year, I would think the OGB might want to consider terminating that 
community. Again, this is difficult to base without having a real world 
example, and KDE is one of the better ones I can think of.
> - The Draft Constitution says that Community Groups (and in particular,
>   the Community Groups'' Facilitators) are responsible for
"communicating
>   the Community Group''s status to the OGB"; what does this
mean to you?
To me it means that the facilitators should be required for giving some type 
of activity update, or notification to the OGB on any activity in their 
community. This also includes any type of problems which might be existing, 
so that the OGB can possibly offer assistance in getting involvement to 
resolve them. The OGB can at least make a best effort to help in that regard.
> - According to the Draft Constitution, "nominations to the office of
>   Facilitator shall be made by the Core Contributors of the Community
>   Group, but the OGB shall not be limited in their appointment to those
>   nominated."  Under what conditions do you believe that the selection
of
>   a Facilitator would or could fall outside of the nominations made by
>   a Community Group''s Core Contributors?
When the OGB feels a better Facilitator exists but didn''t get
nominated. In
this regard, the OBG could nominate the person they best feel is suited. This 
might make it fit in the guidelines better, but the above provides for that.
> - According to the Draft Constitution, "non-public discussion related
to
>   the Community Group, such as in-person meetings or private communication,
>   shall not be considered part of the Community Group activities unless or
>   until a record of such discussion is made available via the normal
> meeting mechanism."  In your opinion, in the context of a Community
Group
> like DTrace -- where a majority of the Core Contributors spend eight to ten
> hours together every work day -- what does this mean?  Specifically, what
> does it mean to be (or not to be) "considered part of the Community
Group
> activities"?  And in your opinion, what role does the OGB have in
auditing
> a Community Group''s activities?
This doesn''t seem to be a problem when the contributers are spending
that much
time and interacting, that is the intention as I see it, to prevent a 
minority in a community from discussing and/or doing something behind the 
scene, without the majority knowing. It just seems like good protocol to me. 
If the DTrace contributers are spending that much time in their community, 
that is good and should be used as an example of collaborating within a given 
community.

The role could be imporant for the OGB to determine if a project is active or 
not.
> - Historically, binary compatibility has been very important to Solaris,
>   having been viewed as a constraint on the evolution of technology.
>   However, some believe that OpenSolaris should not have such constraints,
>   and should be free to disregard binary compatibility.  What is your
>   opinion?
The compatibility has been what allows a device driver that was built on 
Solaris 8, to run on Solaris 10 today. This is GOOD!

I have also heard a compelling argument for lightening up the requirements for 
userland software that would be layered on top of ON, and that might be 
something I could be receptive to, I''m not sure. Without specifics,
it''s also
hard to comment on. Each situation might have to be looked at on a case by 
case basis. In general there should not be any slack of testing and/or binary 
compatibility that would sacrifice the integrity of OpenSolaris. The 
community has some responsibility as the caretakers of the sources, and since 
Sun will need to pre-qualify OpenSolaris packages for thier own Solaris 
distribution, the community does have some onus in regards to continuing the 
standards that have made Solaris what it is to date.

I know your question was in regard to binary compatibility, and not 
requirements, but there is some grey overlapping area in regards to testing. 
I am involved in community device driver testing, and making the testing less 
for some drivers has been mentioned. Unfortunately I have grown more on the 
side that we must ensure the quality, and especially for ON, it is 
imperative.

From an ON perspective, which is OpenSolaris as we know it today, this is 
extremely important. I emphasize this because it is important to allow the 
community to have software repositories outside of ON which can be layered on 
top of ON to form an OpenSolaris distribution, just as Sun will do.
> - If a third-party were to use and modify DTrace in a non-CDDL''d
system,
>   whose responsibility is it to assure that those modifications are
>   made public?  To put it bluntly:  is enforcing the CDDL an OGB issue?
I believe it is Sun''s responibility, and why they have a legal staff on
board,
to resolve such issues.

However, the OGB should be responsible for monitoring/following the open 
source world to notify Sun''s legal to handle.
> - Do you have an opinion on the patentability of software?  In particular,
>   what is the role of the OGB -- if any -- if Sun were to initiate legal
>   proceedings to protect a part of its software patent portfolio that
>   is represented in OpenSolaris?
Personally, patents appear to be a double-edge sword. Unfortunately, the way 
our system works we often need them to protect ourselves.

There is a clause in the CDDL for this, but I can''t site it off the top
of my
head. At least I recall there is and I would be astonished if Sun legal would 
have overlooked something like that when they created CDDL.
> - When you give public presentations, do you run OpenSolaris on your
> laptop?
I run it for the majority of what I do. I have a MacBook I dual boot, which 
runs Solaris Express on the metal and which I use for presentations also. I 
like take a dual pack to presentations, an x86 laptop running Nexenta, and a 
MacBook Pro running Solaris Express.
> Have you ever given a public demonstration of OpenSolaris
> technology?
I think I''ve done about 2 dozen. I''ve given presentations at
LISA, USENIX,
OSCON, Tech Day(s), JavaOne(s), BayLISA, LinuxWorld(s), etc...

Trivia/Fact: Linux can''t deny DTrace, yet it can deny TNS Probes
> - And an extra credit question:  Have you ever used DTrace?  When did you
>   most recently use it, and why?  The answers "just now" and
"to answer
>   this question" will be awarded no points. ;)
+1 :-/

I have actually been using it a bit lately, and the toolkit has a bunch of 
great stuff in it to anal-ize. Brendan has been helpful in getting me started 
in D script-fu. I have pondered on wether a device provider would be useful 
or not, but scripting the device on the available providers does seem to work 
as it is. I haven''t hung out in your community though, even though you
didn''t
ask that...;-)

Solaris Performance & Tools book is nice to have much of it in nicely laid
out
text. Sun should pelt more of those books out at events.

Also, as OGB, I will encourage DTrace to be a topic of the Silicon Valley User 
Group <g>. I notice we haven''t had DTrace as a topic there. We
have quite a
few DTrace users...

Pardon my typos if any, I don''t use spell checking...

-- 

Alan DuBoff - Solaris x86 Engineering - IHV/OEM Group
Advocate of insourcing at Sun - hire people that care about our company!

Licensing Questions:
>
> - DTrace is one of only a small handful of OpenSolaris technologies that
>   has actually been incorporated into other operating systems.  Thus,
>   your position on dual-licensing is very important to us; what is your
>   position on dual-licensing in general?

The OGB does not have the authority to change the license governing the
OpenSolaris source. If Sun were to ask me to provide input into the
licensing decision, I would represent the views and goals of the community
as a whole.


- Do you agree with the conclusions and decrees of CAB/OGB
Position
>   Paper # 20070207?

I can appreciate why the CAB/OGB felt the need to publish a statement about
such a contentious question. Having said that, I don''t think publishing
a
position paper is the approach I would have taken.

The OGB has no control of the future license of OpenSolaris. The OGB should
represent the community''s opinion to Sun. In my opinion, the best way
to do
that is by chartering a working group to analyze licenses and the impact of
changing the OpenSolaris license. When GPLv3 is finalized, that license
should be included in the analysis. The OGB should then provide Sun with a
report of its recommendation and the potential impact.


- The OGB is responsible for the representation of OpenSolaris
to
>   third parties.  If a third party were to inquire about incorporating
>   DTrace into a GPL''d Program, what would be your response or
position?

Like others, I would suggest that the third party seek legal counsel to
ensure they do not violate any of the licenses. I would also encourage them
to participate in the DTrace Community.


Constitutional Questions:
>
> - DTrace is currently a Community Group, but some could argue that it
> would
>   make more sense for DTrace to be a Project in (say) the Observability
>   Community Group.  In your mind, what is (or should be) the difference
>   between a Community Group and a Project -- and where should DTrace fall?

The Community Group should be something which is broad enough to have
multiple, related projects associated with it. Once elected, the new OGB to
work with the community to clearly define these relationships, create
criteria for evaluating the existing Community Groups and projects, and then
classify the existing Community Groups and projects appropriately.

Until the criteria are defined, I won''t offer an opinion about any
Community
Group.


- The Draft Constitution says next-to-nothing about where the
authority
>   lies to make or accept changes to OpenSolaris -- only that Projects
>   operate at the behest of Community Groups, and that Community Groups
>   can be "terminated" by the OGB.  In your opinion, where does or
should
>   this authority lie?  And do you believe that the Constitution should
>   or should not make this explicit?  Finally, under what grounds do you
>   believe that a Community Group should be "terminated"?

A Community Group needs to define the process development process: there is
already a draft process document that needs to be ratified by the Community
Groups.

The Constitution itself should not be explicit about how to resolve
technical issues: the process and resolution mechanism could change as the
community evolves.

Community Groups might be terminated because they are inactive or in
violation of the principles of the community.


- The Draft Constitution says that Community Groups (and in
particular,
>   the Community Groups'' Facilitators) are responsible for
"communicating
>   the Community Group''s status to the OGB"; what does this
mean to you?

I think the OGB should do periodic reviews of the Community Groups in order
to ensure that they are still active. The Community Group Facilitators are
the ones responsible for communicating this status. The OGB might also
review a Community Group if a dispute with the Community Group has been sent
to the OGB for resolution.


- According to the Draft Constitution, "nominations to the office
of
>   Facilitator shall be made by the Core Contributors of the Community
>   Group, but the OGB shall not be limited in their appointment to those
>   nominated."  Under what conditions do you believe that the selection
of
>   a Facilitator would or could fall outside of the nominations made by
>   a Community Group''s Core Contributors?

I can''t envision such a scenario.


- According to the Draft Constitution, "non-public discussion related
to
>   the Community Group, such as in-person meetings or private
> communication,
>   shall not be considered part of the Community Group activities unless or
>   until a record of such discussion is made available via the normal
> meeting
>   mechanism."  In your opinion, in the context of a Community Group
like
>   DTrace -- where a majority of the Core Contributors spend eight to ten
>   hours together every work day -- what does this mean?  Specifically,
> what
>   does it mean to be (or not to be) "considered part of the Community
>   Group activities"?  And in your opinion, what role does the OGB have
in
>   auditing a Community Group''s activities?

Working together in person is obviously a great way to make progress on a
project, but communicating those discussions to the community is the best
way to get more people participating and interested in the project. I think
the Draft Constitution is trying to encourage this broader participation and
provide a mechanism for a member to appeal to the Community Group and OGB if
they feel there are insufficient opportunities to participate.

I already mentioned that I think there should be a periodic review of the
groups to ensure that they are still active. Otherwise, the OGB should
resolve disputes with Community Groups when members bring it to the
OGB''s
attention.


Potpourri:
>
> - Historically, binary compatibility has been very important to Solaris,
>   having been viewed as a constraint on the evolution of technology.
>   However, some believe that OpenSolaris should not have such constraints,
>   and should be free to disregard binary compatibility.  What is your
>   opinion?

This is something for the community to decide. The OGB should help help
enforce what the Community Groups decide.

My personal opinion is that it is easier to achieve binary compatibility if
you enforce it at the source. If an individual has a change that breaks
binary compatibility, they could choose to make that change in their
distribution.


- If a third-party were to use and modify DTrace in a non-CDDL''d
system,
>   whose responsibility is it to assure that those modifications are
>   made public?  To put it bluntly:  is enforcing the CDDL an OGB issue?

No, enforcing CDDL isn''t an OGB issue. The OGB is not a legal entity
and
does currently doesn''t have the legal resources to pursue license
violators.



- Do you have an opinion on the patentability of software?  In
particular,
>   what is the role of the OGB -- if any -- if Sun were to initiate legal
>   proceedings to protect a part of its software patent portfolio that
>   is represented in OpenSolaris?

The only role I could see for the OGB is to act as a liaison between the
community and Sun: providing advice and information proactively and
responding to questions from Sun.


- When you give public presentations, do you run OpenSolaris on your
laptop?
>   Have you ever given a public demonstration of OpenSolaris technology?

I have given OpenSolaris presentations in the past, but did not specifically
demonstrate the OpenSolaris technology except by running it on my laptop
during the presentation.


- And an extra credit question:  Have you ever used DTrace?  When did
you
>   most recently use it, and why?  The answers "just now" and
"to answer
>   this question" will be awarded no points. ;)
>
> No, I haven''t used DTrace.
- Karyn
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://mail.opensolaris.org/pipermail/dtrace-discuss/attachments/20070309/04f5aae2/attachment.html>

>Licensing Questions:
>
>- DTrace is one of only a small handful of OpenSolaris technologies that
>  has actually been incorporated into other operating systems.  Thus,
>  your position on dual-licensing is very important to us; what is your 
>  position on dual-licensing in general?
Dual licensing creates many issues with little or no actual
benefit and an extremely high potential cost of a license based
fork which can never be reconciled.  Regardless of which license is
used.
>- Do you agree with the conclusions and decrees of CAB/OGB Position 
>  Paper # 20070207?
I endorsed the paper as an OGB member.
>- The OGB is responsible for the representation of OpenSolaris to 
>  third parties.  If a third party were to inquire about incorporating
>  DTrace into a GPL''d Program, what would be your response or
position?
That they should consult a lawyer about the licensing issues which might
arise.
>Constitutional Questions:
>
>- DTrace is currently a Community Group, but some could argue that it would 
>  make more sense for DTrace to be a Project in (say) the Observability
>  Community Group.  In your mind, what is (or should be) the difference
>  between a Community Group and a Project -- and where should DTrace fall?
The between projects and communities are somewhat blurry.  Being a project
imples (to me) active development and a community does not carry
that implication.  A project draws people working to build something;
a community feels more geared toward users.
>- The Draft Constitution says next-to-nothing about where the authority 
>  lies to make or accept changes to OpenSolaris -- only that Projects
>  operate at the behest of Community Groups, and that Community Groups
>  can be "terminated" by the OGB.  In your opinion, where does or
should
>  this authority lie?  And do you believe that the Constitution should
>  or should not make this explicit?  Finally, under what grounds do you
>  believe that a Community Group should be "terminated"?
The authority to make changes derives from the development process.
When changes follow due process and are approved by the appropriate
ARC, pass codereview and C-team review, then they are free to integrate.

When conflicts arise, I expect to OGB''s role to assess whether due
process was followed and not override decisions.  From my experience
at Sun I know those cases where management decided to override the
ARCs, large, predictable, problems came back to hound us.

In the cases where the OGB find grounds for appeal, my preferrence would
be to defer the case back and have it heard again by different members
of the particular group.

Communities would be terminated if they are inactive (no mailing
list traffic).  I think terminating an active community is something
I would not conside.
>- The Draft Constitution says that Community Groups (and in particular,
>  the Community Groups'' Facilitators) are responsible for
"communicating
>  the Community Group''s status to the OGB"; what does this
mean to you?
Clearly, an inactive community does not communicate so it''s not about
communicating a level of activity; the communication would primarily
about the status of the community''s members.
>- According to the Draft Constitution, "nominations to the office of
>  Facilitator shall be made by the Core Contributors of the Community
>  Group, but the OGB shall not be limited in their appointment to those
>  nominated."  Under what conditions do you believe that the selection
of
>  a Facilitator would or could fall outside of the nominations made by
>  a Community Group''s Core Contributors?
When a community does not nominate one.
>- According to the Draft Constitution, "non-public discussion related
to
>  the Community Group, such as in-person meetings or private communication,
>  shall not be considered part of the Community Group activities unless or
>  until a record of such discussion is made available via the normal meeting
>  mechanism."  In your opinion, in the context of a Community Group
like
>  DTrace -- where a majority of the Core Contributors spend eight to ten
>  hours together every work day -- what does this mean?  Specifically, what
>  does it mean to be (or not to be) "considered part of the Community
>  Group activities"?  And in your opinion, what role does the OGB have
in
>  auditing a Community Group''s activities?
"If a tree fell in the forrest, and nobody blogged about it; did it
happen?"

I don''t believe the OGB''s role is to actively audit
communities; only
those events which are reflected on the OpenSolaris website are considered
to have happened.  If the DTrace core contributors work 40 hours
a week together but none of that is visible on OpenSolaris, then I would
say nothing is contributed to the community, in the open.
>Potpourri:
>
>- Historically, binary compatibility has been very important to Solaris,
>  having been viewed as a constraint on the evolution of technology.
>  However, some believe that OpenSolaris should not have such constraints,
>  and should be free to disregard binary compatibility.  What is your
>  opinion?
I feel very strongly about binary compatibility.

Both for drivers and for binaries; in a number of cases I have personally
been involved in trying to shoehorn new functionality in old binary
shoes and I have found that it does work in many cases.

And I''ve learned from this that future binary compatibility is a design
consideration for new interfaces also.
>- If a third-party were to use and modify DTrace in a non-CDDL''d
system,
>  whose responsibility is it to assure that those modifications are
>  made public?  To put it bluntly:  is enforcing the CDDL an OGB issue?
The OGB has no money and no budget for lawyers; the OGB is not the
copyright holder and so is no party in a conflict between the copyright
holder and the violator of the CDDL.
>- Do you have an opinion on the patentability of software?  In particular,
>  what is the role of the OGB -- if any -- if Sun were to initiate legal
>  proceedings to protect a part of its software patent portfolio that
>  is represented in OpenSolaris?
Since software can be patented, it is only prudent for companies to
patent sofware if only for defensive reasons.
The majority of patent infringement lawsuits, however, appear to be about
trivial inventions used by "IP companies" to extort huge amounts of
money from other companies.

I support patent litigation only when it is used defensively.

In an ideal world, software patents would not exist because the invention
cost is small, the reproduction cost is nihil.  Certainly a validity
of more than 5 years is preposterous.
>- When you give public presentations, do you run OpenSolaris on your laptop?
>  Have you ever given a public demonstration of OpenSolaris technology?
Technically, one cannot run "OpenSolaris" on one''s laptop;
only an
OpenSolaris based distribution.  My systems generally run the latest
build of Sun''s OpenSolaris distribution, Solaris Nevada but they may be
a few builds out of date in case there are showstopper bugs.

I have demontrated OpenSolaris technolgies on several occassions.
>- And an extra credit question:  Have you ever used DTrace?  When did you
>  most recently use it, and why?  The answers "just now" and
"to answer
>  this question" will be awarded no points. ;)
Yes.  The most recent use was an attempt at looking what went wrong
with bcmndis in 64 bit mode on the Ferrari 3400 which unfortunately made
it work (which points to a timing issue).

Casper

Bryan Cantrill <bmc at eng.sun.com> wrote:
> Licensing Questions:
>
> - DTrace is one of only a small handful of OpenSolaris technologies that
>   has actually been incorporated into other operating systems.  Thus,
>   your position on dual-licensing is very important to us; what is your 
>   position on dual-licensing in general?
These technologies did make it into other OS while not being dual licensed.
I see this as a proof that there is no need for dual licensing.

Dual licensing does not solve problems (if it would, there may be a need
for many more simultaneous licenses), dual licensing however may create 
problems that are not present with a single license. 

I believe that the intention for opensourcing Solaris is to allow others
to use the technology (Sun did make it''s biggest success from open
technologies
like NFS that have been developped by Sun) and to let Solaris benefit from
changes made by others. Dual licensing creates a hard to understand 
situation and may prevent a return of such code.
>From a long face to face discussion, I had with Andy Tucker in September
2004
I know that we did agree on that there is no real way to prevent specific
other projects to use OpenSolaris technology by using a speciafic license.
We did also agree that this cannot/should not even be the intention. Dual 
licensing may create the absurd situation that a code flow back to OpenSolaris
is harder than the code flow from OpenSolaris to other projects.

> - Do you agree with the conclusions and decrees of CAB/OGB Position 
>   Paper # 20070207?
I agree that it is counter productive to have a discussion on GPLv3 before it
is even finished. 

> - The OGB is responsible for the representation of OpenSolaris to 
>   third parties.  If a third party were to inquire about incorporating
>   DTrace into a GPL''d Program, what would be your response or
position?
I would say:

The CDDL is a free license that itself does not create any barriers to mix
with other software. Depending on the way you plan to integrate the code into
your
program, the GPL may prevent this code mix. Please ask a specialized lawyer 
whether your specific attempt for a mix may create problems before you publish
it.

The OGB is neither a legal facility nor the Copyright holder for the code.
If you have legal questions that cannot be answered by your lawyer, please
contact
Sun''s Legal department.

> Constitutional Questions:
>
> - DTrace is currently a Community Group, but some could argue that it would
>   make more sense for DTrace to be a Project in (say) the Observability
>   Community Group.  In your mind, what is (or should be) the difference
>   between a Community Group and a Project -- and where should DTrace fall?
>From my understanding, DTrace is a project. How it should be represented in
a
community depends on the wishes/needs of the people interested in a related 
community.

My private opinion:

A project is something that creates code or white papers.
Projects that create code (except when they are created only for integrating 
code) should not be short lived. Other OSS projects show that there is always
something to improve.

A community is a groupg of people who are interested in a topic.

----->
Note that this is a private view. If I was elected, I would refer you to the
OGB and please you to officially ask for a reply. The OGB is a 
committee/council and you may only get an official answer after there has 
been an official decision by the OGB. Single persons cannot speak for the OGB.
<----

> - The Draft Constitution says next-to-nothing about where the authority 
>   lies to make or accept changes to OpenSolaris -- only that Projects
>   operate at the behest of Community Groups, and that Community Groups
>   can be "terminated" by the OGB.  In your opinion, where does or
should
>   this authority lie?  And do you believe that the Constitution should
>   or should not make this explicit?  Finally, under what grounds do you
>   believe that a Community Group should be "terminated"?
My private opinion is that a community is a group of people interested
in a topic. For this reason, my private opinion is that the OGB should
not terminate a Community. It may be that a specific person needs to be 
excluded from a community if this person is identified as a troublemaker.

> - The Draft Constitution says that Community Groups (and in particular,
>   the Community Groups'' Facilitators) are responsible for
"communicating
>   the Community Group''s status to the OGB"; what does this
mean to you?
I understand this as a hint that it is not the the duty of the OGB
to spend it''s time in getting an overview of the current community
activities
but that the Facilitators should help the OGB to keep track with the activities.

> - According to the Draft Constitution, "nominations to the office of
>   Facilitator shall be made by the Core Contributors of the Community
>   Group, but the OGB shall not be limited in their appointment to those
>   nominated."  Under what conditions do you believe that the selection
of
>   a Facilitator would or could fall outside of the nominations made by
>   a Community Group''s Core Contributors?
If a community definitely agrees on the Facilitator, I belive that the OGB
should accept it, as this is some democratic basics.

> - According to the Draft Constitution, "non-public discussion related
to
>   the Community Group, such as in-person meetings or private communication,
>   shall not be considered part of the Community Group activities unless or
>   until a record of such discussion is made available via the normal
meeting
>   mechanism."  In your opinion, in the context of a Community Group
like
>   DTrace -- where a majority of the Core Contributors spend eight to ten
>   hours together every work day -- what does this mean?  Specifically, what
>   does it mean to be (or not to be) "considered part of the Community
>   Group activities"?  And in your opinion, what role does the OGB have
in
>   auditing a Community Group''s activities?
If a project is a project in the open, it should discuss important things 
in the open. It is not the duty of the OGB to investigate about the way
communication is done. If however a community member complains about decisions
that have been based on closed information, it may be that the OGB needs to act
as an arbitrator. I hope that the communities work in a way that makes this a
rare case.

> Potpourri:
>
> - Historically, binary compatibility has been very important to Solaris,
>   having been viewed as a constraint on the evolution of technology.
>   However, some believe that OpenSolaris should not have such constraints,
>   and should be free to disregard binary compatibility.  What is your
>   opinion?
OpenSolaris is not a distribution and the OGB cannot influence distributions.
"OpenSolaris" as the code base should stay with binary compatibility
as this
is one of the main reasons for the Solaris success.

> - If a third-party were to use and modify DTrace in a non-CDDL''d
system,
>   whose responsibility is it to assure that those modifications are
>   made public?  To put it bluntly:  is enforcing the CDDL an OGB issue?
The OGB is no legal board that has the authority to start legal cases.

I would like to see that Sun is doing the legal actions in such a case.
I would like to see the OGB to encourage Sun to help but the OGB is a 
committee/council and this is only my private opinion.
> - Do you have an opinion on the patentability of software?  In particular,
>   what is the role of the OGB -- if any -- if Sun were to initiate legal
>   proceedings to protect a part of its software patent portfolio that
>   is represented in OpenSolaris?
Software cannot be patented in Europe but this is ignored even by the 
European patent office. The USA allow to patent software even though
patenting software is against the original idea of patents from 200 years ago.

The CDDL allows to use patents (related tp OpenSolaris) for free in case that
the patents are controlled by contributors. Patents owned by Sun do not seem 
to be a threat to OpenSolaris and it''s users.

> - When you give public presentations, do you run OpenSolaris on your
laptop?
OpenSolaris is not a distribution, but I of course run an Opensolaris 
installation on my laptop.

>   Have you ever given a public demonstration of OpenSolaris technology?
Yes, e.g. on the Linux days in Chemnitz ;-)
> - And an extra credit question:  Have you ever used DTrace?  When did you
>   most recently use it, and why?  The answers "just now" and
"to answer
>   this question" will be awarded no points. ;)
I did of course use DTrace to help my driver development. 

The last time I did use it was when I was enhancing the "hsfs" code
from
OpenSolaris (e.g. to verify correct file system type discrimination
after adding Joliet/ISO-9660-v2 support and SDT probes to hsfs).

I would need to take some time to learn more on it and to use it more fequently.

J?rg

-- 
 EMail:joerg at schily.isdn.cs.tu-berlin.de (home) J?rg Schilling D-13353 Berlin
       js at cs.tu-berlin.de                (uni)  
       schilling at fokus.fraunhofer.de     (work) Blog:
http://schily.blogspot.com/
 URL:  http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily

Casper.Dik at Sun.COM wrote:
>> Licensing Questions:
>>
>> - DTrace is one of only a small handful of OpenSolaris technologies
that
>>  has actually been incorporated into other operating systems.  Thus,
>>  your position on dual-licensing is very important to us; what is your 
>>  position on dual-licensing in general?
> 
> Dual licensing creates many issues with little or no actual
> benefit and an extremely high potential cost of a license based
> fork which can never be reconciled.  Regardless of which license is
> used.
> 
>> - Do you agree with the conclusions and decrees of CAB/OGB Position 
>>  Paper # 20070207?
> 
> I endorsed the paper as an OGB member.
> 
>> - The OGB is responsible for the representation of OpenSolaris to 
>>  third parties.  If a third party were to inquire about incorporating
>>  DTrace into a GPL''d Program, what would be your response or
position?
> 
> That they should consult a lawyer about the licensing issues which might
> arise.
> 
>> Constitutional Questions:
>>
>> - DTrace is currently a Community Group, but some could argue that it
would
>>  make more sense for DTrace to be a Project in (say) the Observability
>>  Community Group.  In your mind, what is (or should be) the difference
>>  between a Community Group and a Project -- and where should DTrace
fall?
> 
> The between projects and communities are somewhat blurry.  Being a project
> imples (to me) active development and a community does not carry
> that implication.  A project draws people working to build something;
> a community feels more geared toward users.
> 
>> - The Draft Constitution says next-to-nothing about where the authority
>>  lies to make or accept changes to OpenSolaris -- only that Projects
>>  operate at the behest of Community Groups, and that Community Groups
>>  can be "terminated" by the OGB.  In your opinion, where does
or should
>>  this authority lie?  And do you believe that the Constitution should
>>  or should not make this explicit?  Finally, under what grounds do you
>>  believe that a Community Group should be "terminated"?
> 
> The authority to make changes derives from the development process.
> When changes follow due process and are approved by the appropriate
> ARC, pass codereview and C-team review, then they are free to integrate.
> 
And what would be your (the general "you", not just Casper) plans to
help
make the ARC and especially the C-team more practically part of OpenSolaris 
process, rather than a part of Sun process we''re exposed to from one
side,
but not, so far, fully involved with?

-- Rich

Hey,

Richard Lowe wrote:
> Casper.Dik at Sun.COM wrote:
>> The authority to make changes derives from the development process.
>> When changes follow due process and are approved by the appropriate
>> ARC, pass codereview and C-team review, then they are free to
integrate.
>>
> 
> And what would be your (the general "you", not just Casper) plans
to
> help make the ARC and especially the C-team more practically part of
> OpenSolaris process, rather than a part of Sun process we''re
exposed to
> from one side, but not, so far, fully involved with?
This is another really hard question, and again, I''m going to reply
with ''I
don''t know''. There are a *lot* of groups involved in Solaris
somewhere along the
line that aren''t in touch with OpenSolaris, or worse choose not to be
in touch
with OpenSolaris. I suspect it''s going to be a slow process seeing
everyone
begrudgingly come out of the wood-work. At this stage, I''m a little
unsure of
whether it''s poor communication, poor information, or something wider.

Assuming the first couple, there are a couple of things that I''d like
to do
while on the OGB that may help this. Get a small charged and clueful team around
the various engineering offices spreading the message about what OpenSolaris is,
and how things are currently done. The message is coming down from the top, and
coming up from the bottom, and it''s just getting real hazy somewhere in
the
middle. We need to change that. Another plan is to organize an OpenSolaris
conference, where I''d push to host it in a place that could maximize
the most
Solaris-based engineers possible, and try and magic up a budget to fly our other
awesome hackers over for it too.


Glynn

Bryan Cantrill wrote:
> --------------------------------------------------------------------------
> Licensing Questions:
> 
> - DTrace is one of only a small handful of OpenSolaris technologies that
>   has actually been incorporated into other operating systems.  Thus,
>   your position on dual-licensing is very important to us; what is your 
>   position on dual-licensing in general?

Generally, I am against multiple-licensing. I think it creates more
problems than it solves - unless none of the licenses are GPLv2 in
which case I am prepared to consider such a proposal carefully.

I do not have a problem with forking, but I am concerned that the
effect of licensing OpenSolaris under CDDL as well as another license
will dilute or otherwise render void the benefits which I see the CDDL
providing to the OpenSolaris community and others which with to make
use of the CDDL.
> - Do you agree with the conclusions and decrees of CAB/OGB Position 
>   Paper # 20070207?
Given the way which that particular discussion was progressing, I was
glad to see that the OGB published that document. It addressed my
concerns with that debate, and I agree with the conclusions therein.
I am quite happy to consider GPLv3 as a potential license for OpenSolaris
subject to my concerns above, however until GPLv3 has been formally
released and discussion concerning it has calmed down, I see no reason
to give it any consideration.

On the flip side, I was disappointed that the OGB felt that there was
a need to issue such a position paper.

> - The OGB is responsible for the representation of OpenSolaris to 
>   third parties.  If a third party were to inquire about incorporating
>   DTrace into a GPL''d Program, what would be your response or
position?
As long as the third party was planning to re-implement DTrace from
theoretical and algorithmic first principles rather than copy code,
I would not have a problem with it.

> Constitutional Questions:
> 
> - DTrace is currently a Community Group, but some could argue that it would
>   make more sense for DTrace to be a Project in (say) the Observability
>   Community Group.  In your mind, what is (or should be) the difference
>   between a Community Group and a Project -- and where should DTrace fall?
I''ll answer the second question first - I think DTrace should be a
fully
fledged Community, since its facility is so far reaching.

Overall I think the current structures have worked reasonably well as
a first approximation. Like any approximation, these need to be refined
and since we''ve now got ~2 years worth of experience with the evolution
of our OpenSolaris community I think the new OGB should take this as an
item requiring their consideration.

To me, a Community is a group of people who have a collection of goals,
some of which can be expressed as or worked towards as Projects.

> - The Draft Constitution says next-to-nothing about where the authority 
>   lies to make or accept changes to OpenSolaris -- only that Projects
>   operate at the behest of Community Groups, and that Community Groups
>   can be "terminated" by the OGB.  In your opinion, where does or
should
>   this authority lie? 
I find this question to be vague. Are you (on behalf of the DTrace community)
asking about changes to the software which forms OpenSolaris, or are you
asking about changes to the way our community is constructed?

 From a software perspective, if a Project can convince the OGB that
their entity has sufficient value to be included or integrated into
the OpenSolaris source tree, then I think that opinion should carry
some weight or influence to the relevant ARC. Until or unless the ARC
process changes, that is.

 From the community perspective, I do not think that termination of a
Community would necessarily be a good thing, unless there was serious
conflict within that Community which prevented it from achieving its
stated goals. I would want to see all members of the OGB engage with
a Community which was believed to be in such a state, and try to give
it a kickstart as a first response.

>   And do you believe that the Constitution should
>   or should not make this explicit? 
I think it should be made explicit, to remove doubt.
>   Finally, under what grounds do you
>   believe that a Community Group should be "terminated"?
(a) if the Community Group is unable to be kickstarted back to life
(b) if the Community Group is riven by the inability to work cooperatively
in pursuit of its stated aims

Following on from this, if a Community group has achieved its stated
goals, I think it preferable that rather than terminating the group
it was given the opportunity to re-form into something more appropriate
which reflects the effect which it has had on the wider OpenSolaris
community.
> - The Draft Constitution says that Community Groups (and in particular,
>   the Community Groups'' Facilitators) are responsible for
"communicating
>   the Community Group''s status to the OGB"; what does this
mean to you?
(a) On an as-required basis, the OGB may request that a Community''s
leadership provide an indication of the health of the group. This
might include an appraisal of whether the Community''s stated goals
are being addressed.

(b) "alive" or "dormant" or "dead" :)

> - According to the Draft Constitution, "nominations to the office of
>   Facilitator shall be made by the Core Contributors of the Community
>   Group, but the OGB shall not be limited in their appointment to those
>   nominated."  Under what conditions do you believe that the selection
of
>   a Facilitator would or could fall outside of the nominations made by
>   a Community Group''s Core Contributors?
Assuming that there are sufficient experienced people who join the
Community, this should be a very rare occurrence. I define
"experience"
in case as "people who have worked on software engineering projects" -
not a very tight definition, but one which should allow people who are
new to get a foothold. Rather than appoint a Facilitator in such a
circumstance, I would suggest that the Community accept a mentor.


> - According to the Draft Constitution, "non-public discussion related
to
>   the Community Group, such as in-person meetings or private communication,
>   shall not be considered part of the Community Group activities unless or
>   until a record of such discussion is made available via the normal
meeting
>   mechanism."  In your opinion, in the context of a Community Group
like
>   DTrace -- where a majority of the Core Contributors spend eight to ten
>   hours together every work day -- what does this mean?  Specifically, what
>   does it mean to be (or not to be) "considered part of the Community
>   Group activities"? 
Exactly what it says. Now, in the context of a tightly knit group
of Core Contributors such as DTrace or FMA, I have absolutely no
objection to continuing on as you have done in the past, however if
you want something you decide to be part of the Community direction
then you''ll have to publicise it and get the Community members to
agree.

>   And in your opinion, what role does the OGB have in
>   auditing a Community Group''s activities?
Minimal, unless the Community Group requests it.

> Potpourri:
> 
> - Historically, binary compatibility has been very important to Solaris,
>   having been viewed as a constraint on the evolution of technology.
>   However, some believe that OpenSolaris should not have such constraints,
>   and should be free to disregard binary compatibility.  What is your
>   opinion?
Binary compatibility must stay. It is one of the most powerful and
tractable selling points for OpenSolaris. I understand the desire to
move away from this guarantee, but I think such desires are based on
a misunderstanding of the benefit which binary compatibility brings.

> - If a third-party were to use and modify DTrace in a non-CDDL''d
system,
>   whose responsibility is it to assure that those modifications are
>   made public?  To put it bluntly:  is enforcing the CDDL an OGB issue?
In the first instance, I would expect that the Initial Contributor
would have the responsibility of enforcing the CDDL conditions. I
also expect that the OGB could be used as a path for providing public
support for such effort, but since the OGB has no ownership of copyrights
or IP I do not expect that the OGB would have any other standing.

> - Do you have an opinion on the patentability of software?  In particular,
>   what is the role of the OGB -- if any -- if Sun were to initiate legal
>   proceedings to protect a part of its software patent portfolio that
>   is represented in OpenSolaris?
I accept that patents on software exist, and there are justifiable reasons
why they exist, but I am not in favour of patents on software.

Unless a member of the OGB was requested to provide expert testimony in
patent litigation over OpenSolaris started by Sun, I do not expect that
the OGB would have anything to do except keep up with the current status
of the case. Any serious or in-depth comment would have to be referred
to Sun.
> - When you give public presentations, do you run OpenSolaris on your
laptop?
>   Have you ever given a public demonstration of OpenSolaris technology?
I run Solaris Nevada on my laptop. I have done so since build 5. Prior to
that I ran pre-FCS builds of Solaris 10.

I have presented/talked about ZFS and the Leadville stack, but I have not
demonstrated their features per se. If I could just find a pc-card fibre
channel hba .....

> - And an extra credit question:  Have you ever used DTrace?  When did you
>   most recently use it, and why?  The answers "just now" and
"to answer
>   this question" will be awarded no points. ;)
(a) yes. I even got the DTrace Iron Chefs to autograph my bound copy
of the manual while at SunEngCon 2005.

(b) yesterday, to help a new engineer provide more data for his
performance bug so that I could triage it appropriately.

(c) I''m certainly no expert at using DTrace and hit the manual a lot,
but there''s no way I want to go back to a pre-DTrace universe.



I''ll be placing the above answers on my blog within the next day.


cheers,
James C. McPherson
--
Solaris kernel software engineer, system admin and troubleshooter
               http://www.jmcp.homeunix.com/blog
Find me on LinkedIn @ http://www.linkedin.com/in/jamescmcpherson

Glynn Foster wrote:
...
> Assuming the first couple, there are a couple of things that I''d
like to do
> while on the OGB that may help this. Get a small charged and clueful team
around
> the various engineering offices spreading the message about what
OpenSolaris is,
> and how things are currently done. The message is coming down from the top,
and
> coming up from the bottom, and it''s just getting real hazy
somewhere in the
> middle. We need to change that. Another plan is to organize an OpenSolaris
> conference, where I''d push to host it in a place that could
maximize the most
> Solaris-based engineers possible, and try and magic up a budget to fly our
other
> awesome hackers over for it too.
I am violently in favour of having an OpenSolaris conference,
with attendance from as many community groups as possible. I
would want such an event to have both organisational discussions
and engineering / "hey isn''t this neat" presentations. And I
definitely want to go!

Logistics-wise, I would like to see annual conferences like
the Gnome community has, held on a different continent each
year if at all possible. We''re are a global community, so we
should make our presence felt all over the world.


cheers,
James C. McPherson
--
Solaris kernel software engineer
Sun Microsystems

On Sat, Mar 10, 2007 at 08:29:18PM -0500, Richard Lowe wrote:
> And what would be your (the general "you", not just Casper) plans
to help
> make the ARC and especially the C-team more practically part of OpenSolaris
> process, rather than a part of Sun process we''re exposed to from
one side,
> but not, so far, fully involved with?
Of the two, the ARC is much more difficult to rationalise; I''ll
explain why below.  As for the C-teams and the more general problem of
consolidation management, I''ll let this text from my position paper[0]
do the answering:

    One of the OGB''s most important tasks will be to rationalise the
    Community Group structure into one which will allow meaningful
    self-government. The centerpiece of my plan for doing this is
    construction of Consolidation-Sponsoring Community Groups
    (CSCGs). Each of these groups will be given control over an
    existing consolidation. This structure is not unlike that which
    exists today in the misnamed Nevada Community, representing
    ON. But that Community does not govern openly, and other
    consolidations are entirely missing structure under which they can
    be governed legitimately. Since the Constitution provides for the
    Community Group as the unit of independent government, each
    consolidation requires one to oversee its progress. The CSCGs will
    be responsible both for controlling the content of their codebases
    and for providing guidance and leadership to project teams
    desirous of integration. They will be required to adopt a set of
    rules (harmonised but not necessarily identical across all CSCGs)
    for integration and apply fairly these rules.

The challenge associated with the ARC (or ARCs) is that it maps poorly
onto the Community Group structure.  It makes little sense to me that
an Architecture Community Group would sit alongside, say, an
Observability Community Group.  Observability incorporates a number of
subsystems in the OS which in turn need to be properly integrated into
each project.  So would Reliability, or Virtualization.  Architecture
is not another such feature set but rather the way in which all those
features, along with the new ones offered by the project, fit together
and expose themselves to other consumers.  That is, Architecture is
both a superset of and yet entirely disjoint from all other CGs'' areas
of interest.  The practical effect is substantial overlap: we would
expect each CG to offer project teams advice concerning how best to
integrate their work with existing features (and, for projects
directly related to the CG''s area of expertise, what features it
should offer to others).  In some ways, however, this directly
conflicts with the mandate of an Architecture CG, which is to provide
architecture guidance to all project teams.  In the current system, an
observability expert cannot override the ARC''s decisions with respect
to a proposed observability project.  Yet under the Constitution, the
Observability CG is supposed to be self-governing.  The defining
question is what exactly the latter CG is expected to govern, and by
what mechanism - the very question the Constitution so conspicuously
fails to answer.

It''s easy enough in my CSCG model to simply require that all CSCGs
adopt rules requiring architectural review by a particular CG just as
they should require other CGs with expertise in relevant areas to
review and perhaps approve each project prior to integration.  Indeed,
this is not unlike the system that exists today.  The CSCGs do indeed
have complete control over their areas of responsibility, namely, the
existing consolidations.  But this leaves all other CGs less equal,
their endorsements subject to veto and without any code of their own
to govern.  A logical conclusion one could reach on this line of
thinking is that CSCGs and perhaps the ARC should be the *only* CGs.
The reality on the ground thus maps poorly to the Constitution we''ve
been given, suggesting that the Framers either did not consider this
matter in sufficient detail or intended much more radical changes in
either the structure of consolidations, global review processes, or
both.  Mr. Fielding in fact hinted at just such an intent[1]:

    We don''t need to enshrine one committee''s view of how
C-Teams
    operate in an organization-wide constitution because C-Teams
    simply aren''t relevant to *every* activity at OpenSolaris, and the
    vast majority of comments we have received so far clearly indicate
    that the existing consolidation boundaries are arbitrary AND
    dysfunctional. Personally, I am hoping that the communities feel
    empowered to change the things that are obviously causing them
    harm right now, and let the consensus process ensure that the
    traditions are adequately promoted and maintained over time.

Presumably Mr. Fielding and perhaps others have some grand detailed
view of how all these things should be made to fit together in the
rather obvious presence of existing bodies of code with no associated
governing units and vice versa.  Unfortunately, they''ve not seen fit
to share that view nor to stand for election themselves.  If consensus
does not emerge within a few months as to an appropriate way to map
the (possibly modified) existing practical devices of government onto
the new constitutional structures, I''ll probably favour amending the
constitution rather than spinning our wheels forever trying to
shoehorn OpenSolaris into a framework that may well be inappropriate
to our broader goals.

At some point I''d like to hear Mr. Plocher and others more intimitely
involved with the operation of the ARC Community express their views
on how that Community could be made to fit into the new Constitutional
world of governing CGs.  Their testimony will be needed before the OGB
as it considers how best to restructure the Communities into
meaningfully self-governing units.

[0] http://blogs.sun.com/wesolows/entry/ogb_election
[1] http://www.opensolaris.org/jive/message.jspa?messageID=99494#99494

-- 
Keith M Wesolowski		"Sir, we''re surrounded!" 
FishWorks			"Excellent; we can attack in any direction!"

On 09/03/07, Bryan Cantrill <bmc at eng.sun.com>
wrote:
> Licensing Questions:
>
> - DTrace is one of only a small handful of OpenSolaris technologies that
>   has actually been incorporated into other operating systems.  Thus,
>   your position on dual-licensing is very important to us; what is your
>   position on dual-licensing in general?
In my view, generally speaking, dual-licensing can be frustrating and
confusing for development communities. I think that dual-licensing is
something that should only be done whenever it provides a clear and
tangible benefit for a given development community. Part of the reason
I believe this is because all of the problems I have seen that have
occurred whenever a community chooses to dual-license their particular
development product. I also think that dual or tri licensing tends to
encourage license proliferation.
> - Do you agree with the conclusions and decrees of CAB/OGB Position
>   Paper # 20070207?
Partially; I was happy to see someone step in and point out that now
probably wasn''t an appropriate time for a lengthy, involved license
discussion. In part because our community has so many other challenges
that it needs to meet that are more important to the long-term growth
of our community. In addition, license discussions are unfortunately
rarely as productive as they should be and tend to stir up strong
emotional reactions among communities.
> - The OGB is responsible for the representation of OpenSolaris to
>   third parties.  If a third party were to inquire about incorporating
>   DTrace into a GPL''d Program, what would be your response or
position?
My response would be that although the CDDL allows combinations of
software with many different licenses, the GPL likely prohibits this
particular combination. I would also tell them that I would like to
see OpenSolaris technology (such as DTrace) adopted by as many people
as possible. However, they should seek legal counsel in their
respective country to determine whether they can legally redistribute
the derivative work in question.
> Constitutional Questions:
>
> - DTrace is currently a Community Group, but some could argue that it would
>   make more sense for DTrace to be a Project in (say) the Observability
>   Community Group.  In your mind, what is (or should be) the difference
>   between a Community Group and a Project -- and where should DTrace fall?
I''m not certain. My initial reaction is that DTrace seems to belong as
a project within the Observability community given its use. However,
it seems as though communities and projects in general are in need of
a bit of redefinition or strategic reorganisation at the moment.
> - The Draft Constitution says next-to-nothing about where the authority
>   lies to make or accept changes to OpenSolaris -- only that Projects
>   operate at the behest of Community Groups, and that Community Groups
>   can be "terminated" by the OGB.  In your opinion, where does or
should
>   this authority lie?
The authority to accept changes to OpenSolaris should lie with the
community that is responsible for that particular part of OpenSolaris.
The OGB should only interfere if the community itself is somehow
failing to perform their responsibilities to ensure the positive
growth and development of their respective community, and then only as
a last resort when arbitration has completely failed.
>  And do you believe that the Constitution should
>   or should not make this explicit?
I''m a bit torn as to whether or not the constitution should make this
explicit; part of me believes that some individuals would feel "more
comfortable" it were an explicit part, and another part of me says
that its easier to keep things simple and, for the most part, trust
individuals to the right thing. As it is, our constitution is fairly
lengthy, and it may "put off" some individuals from contributing
because they think that the rules are more important than the
community.
>  Finally, under what grounds do you
>   believe that a Community Group should be "terminated"?
To me, termination implies dissolving a community instead of combining
it with another or performing reorganisation of said community. With
that in mind, I believe that termination should only happen when:

* it has become apparent that the existence of the allocation of
resources to a given community has become a poor investment (no
benefit to the community)
* no activity has ever occurred after the initiation of the community
* the community in question has continuously shown a lack of
willingness to contribute in a positive way to the community whether
through failure or otherwise
* termination should be an absolute last resort after all arbitration
processes have failed
> - The Draft Constitution says that Community Groups (and in particular,
>   the Community Groups'' Facilitators) are responsible for
"communicating
>   the Community Group''s status to the OGB"; what does this
mean to you?
It means that the community''s facilitator should  communicate on a
regular basis about the activity of their community, its growth, and
what positive contribution they feel they are making to the
OpenSolaris community to the OGB.
> - According to the Draft Constitution, "nominations to the office of
>   Facilitator shall be made by the Core Contributors of the Community
>   Group, but the OGB shall not be limited in their appointment to those
>   nominated."  Under what conditions do you believe that the selection
of
>   a Facilitator would or could fall outside of the nominations made by
>   a Community Group''s Core Contributors?
At the moment, I cannot fathom under what circumstances this would
specifically happen. However, I think this would only happen as a
result of a choice made by that community that the OGB knows cannot be
accepted (legal reasons or issues regarding said individual that would
cause negative repercussions to the community). It should definitely
be a last resort.
> - According to the Draft Constitution, "non-public discussion related
to
>   the Community Group, such as in-person meetings or private communication,
>   shall not be considered part of the Community Group activities unless or
>   until a record of such discussion is made available via the normal
meeting
>   mechanism."  In your opinion, in the context of a Community Group
like
>   DTrace -- where a majority of the Core Contributors spend eight to ten
>   hours together every work day -- what does this mean?  Specifically, what
>   does it mean to be (or not to be) "considered part of the Community
>   Group activities"?  And in your opinion, what role does the OGB have
in
>   auditing a Community Group''s activities?
I think the intent is that important communications and discussions
regarding the future of the community should happen in an open manner,
at least within that community, so that communities can feel positive
about the decisions that are made regarding their future.

The OGB''s roles in auditing a community group''s activities
should be
limited to ensuring that a community is contributing in a positive
manner, and to ensure efficient allocation of resources and
organisation of the community. Wherever possible, the OGB should be
assisting a community in a way that will help that community encourage
growth and development.
> Potpourri:
>
> - Historically, binary compatibility has been very important to Solaris,
>   having been viewed as a constraint on the evolution of technology.
>   However, some believe that OpenSolaris should not have such constraints,
>   and should be free to disregard binary compatibility.  What is your
>   opinion?
I think that backwards compatibility is perhaps one of the most
important, defining characteristics of Solaris/OpenSolaris technology.
I do not feel as though this focus has overly constrained the
development of innovative and important technology as proven by
DTrace, ZFS, SMF, Containers, and others.

With that said, I think that a "experimental" development project
would be appropriate for those that like "living on the edge" and want
to "try new things." The Kitchen Sink Language Project for Java is a
great example of how this can have a positive effect on a community''s
development.
> - If a third-party were to use and modify DTrace in a non-CDDL''d
system,
>   whose responsibility is it to assure that those modifications are
>   made public?  To put it bluntly:  is enforcing the CDDL an OGB issue?
I not believe enforcing the CDDL is an OGB issue. I think that the
community whose code was used improperly should ask those improperly
using it to stop first *politely as possible*. Failing that, the
community in question should then notify the OGB, who in turn can
discuss the issue with Sun''s legal counsel who can discuss the
enforcement of it.
> - Do you have an opinion on the patentability of software?  In particular,
>   what is the role of the OGB -- if any -- if Sun were to initiate legal
>   proceedings to protect a part of its software patent portfolio that
>   is represented in OpenSolaris?
In short, unless Sun were going against the terms of the CDDL, I don''t
think that the OGB has any role. If for some reason (which I don''t
believe would ever happen) Sun were to initiate legal patent
proceedings against a CDDL users, I believe the would be obligated to
inquire why and attempt to discern the nature of the situation.
> - When you give public presentations, do you run OpenSolaris on your
laptop?
>   Have you ever given a public demonstration of OpenSolaris technology?
I have never given a presentation of OpenSolaris, though I would
certainly try my best to do so if requested. I run an
OpenSolaris-based distribution on my desktop and workstation
frequently (I''m typing this on my SXCR b56 laptop right now).
> - And an extra credit question:  Have you ever used DTrace?  When did you
>   most recently use it, and why?  The answers "just now" and
"to answer
>   this question" will be awarded no points. ;)
Yes, I recently used DTrace to attempt to discern a problem I was
having with being unable to unmount a ZFS share. However, I am not an
expert in using it yet as I''m still more comfortable with truss, mdb,
and gdb.

-- 
"Less is only more where more is no good." --Frank Lloyd Wright

Shawn Walker, Software and Systems Analyst
binarycrusader at gmail.com - http://binarycrusader.blogspot.com/

On 09/03/07, Bryan Cantrill <bmc at eng.sun.com>
wrote:
> - Do you have an opinion on the patentability of software?
Unfortunately my opinion can only be based on that of the US patent
system as I am unfamiliar with any other. With that in mind, I think
that *true* innovation should be patentable and rewarded (the original
intent of the US patent system). However, I feel that the patent
system today has been horribly abused, and because of that, patents
are an unfortunate, necessary part of any organisation or company that
provides innovative solutions to their customers or community.

-- 
"Less is only more where more is no good." --Frank Lloyd Wright

Shawn Walker, Software and Systems Analyst
binarycrusader at gmail.com - http://binarycrusader.blogspot.com/

Richard Lowe wrote:
> And what would be your (the general "you", not just Casper) plans
to
> help make the ARC and especially the C-team more practically part of 
> OpenSolaris process, rather than a part of Sun process we''re
exposed to
> from one side, but not, so far, fully involved with?
I''ve been thinking about the ARC side of this for a while, both as an
ARC
member/case sponsor, and as the person who gets poked most often on
#opensolaris IRC to go remind other case sponsors to open their cases to
the community.

I think the current approach of trying to bolt the community onto the side of
the current ARC process without disrupting it too much is not going well, and
is definitely confusing case sponsors who don''t know the current state
of
things and leaving far too many cases closed which should be open.

I''d actually like to propose that the solution be disruptive (and what
follows
is definitely a rough first draft to be fleshed out with the community) and
that it should be so obvious how to file an open case that no one ends up in
the closed bucket by mistake.

First, I''d dissolve PSARC & LSARC - it makes no sense that the only
consolidation with externally committable source code has closed ARC
reviews, just because desktop has historically gone to LSARC, but only
PSARC has been opened up.

I''d form 3 new ARC''s out of the membership of those two, and
realize there''s
probably going to be some overlap of the membership lists:

1) OpenSolaris ARC (OSARC) - covers all open consolidations, including
    Desktop, will include external members - all mail & case logs will be
    open from the beginning.
2) Closed Platform ARC (CPARC) - covers the closed portions of Solaris,
    both encumbered components of open consolidations (like the various IHV
    drivers in ON) and consolidations not being opened (like CDE).
3) Layered Product ARC (LPARC) or System Management ARC (SMARC) - the various
    system management and other tools currently reviewed by LSARC.

(I''m not sure if developer tools like Sun Studio would go to 2 or 3. 
LSARC
  has the current problem of being such a dumping ground for anything that
  doesn''t fit into the other ARC''s that even after three
years, I can''t begin
  to really figure out what some of these things do - there''s just too
many
  different technologies to understand - to review LSARC cases you have to
  be familiar with desktop toolkits, accessibility issues, SNMP, NetBeans,
  MBeans, MIB''s, WSDL, patch distribution tools, databases, and a whole
lot
  more.)

For the OpenSolaris ARC, I''d set up an externally hosted Mercurial
repository
to contain the case materials, and put an externally runnable version of
the nextcase tool in there to create new cases.   The coordination with the
internal tools would simply be a child of that repository checked out on the
internal ARC infrastructure machine.   (If you want to avoid overlap of case
numbers between OSARC and other ARC''s, just make OSARC case numbers
start at
5000 or some other suitably high number that the other ARC''s
won''t hit in a
year.)

I''d also give up on the overcomplicated overloading of mailing lists as
membership rosters for the opensolaris ARC and just have OSARC at
opensolaris.org
containing members, interns, and interested parties to save having to wait
another 6 months for the opensolaris.org staff to rewrite the mailman/jive
tools to work with the unusual mailing list structure used by current
ARC''s.

When gathering requirements for the external ARC tool base, anything whose
only goal was "compatibility with the internal ARC infrastructure"
would be
rated only "nice to have" and not must have, and lower priority than
getting
the ARC up and running fully externally in the next few months.

-- 
	-Alan Coopersmith-           alan.coopersmith at sun.com
	 Sun Microsystems, Inc. - X Window System Engineering

[ Roaming really far afield from DTrace here, but ... ]

 > I''d form 3 new ARC''s out of the membership of those two,
and realize there''s
 > probably going to be some overlap of the membership lists:
 > 
 > 1) OpenSolaris ARC (OSARC) - covers all open consolidations, including
 >     Desktop, will include external members - all mail & case logs will
be
 >     open from the beginning.
 > 2) Closed Platform ARC (CPARC) - covers the closed portions of Solaris,
 >     both encumbered components of open consolidations (like the various
IHV
 >     drivers in ON) and consolidations not being opened (like CDE).

This split seems challenging to me, since the legal encumbrances tied to
our source do not align themselves on architectural boundaries.  For
instance, if the ARC is reviewing a change that will have a widespread
impact on GLDv3 network drivers, then all network drivers need to be
considered at once, not just those that are open or closed source.

Or would a case like that end up going to CPARC in your proposed world?

 > 3) Layered Product ARC (LPARC) or System Management ARC (SMARC) - the
various
 >     system management and other tools currently reviewed by LSARC.

-- 
meem

> > The authority to make changes derives from the development process.
> > When changes follow due process and are approved by the appropriate
> > ARC, pass codereview and C-team review, then they are free to
integrate.
> >
>
> And what would be your (the general "you", not just Casper) plans
to help
> make the ARC and especially the C-team more practically part of
> OpenSolaris
> process, rather than a part of Sun process we''re exposed to from
one side,
> but not, so far, fully involved with?

As others have pointed out, there are many issues rapped up together with
making the ARC and the C-Team part of OpenSolaris process:

  * Source code management system: Without the live source being managed
outside the firewall, there are some things that the community will not be
able to do. I think the updates happen quickly enough to make this workable,
but (for example) there can not be a non-Sun gatekeeper today.

  * Bug tracking software: Universal access to the same bug tracking
software and database is critical to having a truly interactive process.

  * General process openness for ARC and C-Team: I agree with others that
simply opening the doors to these teams is not going to work, the process
needs to be re-worked to include the community.

I don''t have any bright ideas for how to improve things moving forward,
but
the community and OGB need to work with Sun to drive these critical issues.
In many cases, this is going to mean not relying entirely on Sun, so the OGB
also needs to work on finding a way to get people in the broader community
to help work on the solution to these issues where and when they can.

- Karyn
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://mail.opensolaris.org/pipermail/dtrace-discuss/attachments/20070312/5047d348/attachment.html>

Karyn Ritter writes:
>   * General process openness for ARC and C-Team: I agree with others that
> simply opening the doors to these teams is not going to work, the process
> needs to be re-worked to include the community.
Not sure about the C-team issue, but I think the ARC one is something
that we can get started on right away.

Let''s have non-Sun folks volunteer as interns on the ARC.  In all of
the ARC improvements I''ve seen, the changes have been driven by
someone who has an interest in the outcome.  I think that having
external folks volunteering to be on the ARC will do more than
anything else we can dream up for moving things forward.

For those who might be listening in, but aren''t familiar with how the
ARC functions, here''s a brief background on internship.  Interns have
fast-track privileges, and are expected to attend ARC meetings
regularly.  They volunteer to write opinions for full cases, and
usually direct the reviews of those cases.  Periodically, the full
members vote to promote interns to membership status, but there''s no
set timeline for this, as every intern is different.  What we look for
is technical breadth and skill.  In other words, ask good questions
and have insightful comments on a wide variety of cases (not just one
narrow area).

Anyone interested in internship (the qualification bar is
intentionally low; showing up is most of it) should contact the chair
of the ARC or opensolaris-arc at opensolaris.org.  (Yes, I know the
infrastructure isn''t quite there yet.  Nothing like lighting a fire.)

As for the open/closed issue, I don''t quite agree with Alan
Coopersmith.  The current division of labor in the ARC is based on
technical areas that (mostly) hang together due to interrelationships.
I suspect that dividing the ARC cases by other (conflicting) criteria
will just produce confusion, much as it does with the org chart.
Instead, I''d rather see the ARC forced to deal with *everything* as
open by default.  Closed cases should be extraordinarily rare and
should require substantial justification to launch.  They should be
hard to do, because we really do want everything to be open, and
because we want people considering closed work to reconsider.

Creating a closed ARC and a special ARC for open software just makes
it too easy.

-- 
James Carlson, Solaris Networking              <james.d.carlson at
sun.com>
Sun Microsystems / 1 Network Drive         71.232W   Vox +1 781 442 2084
MS UBUR02-212 / Burlington MA 01803-2757   42.496N   Fax +1 781 442 1677

James Carlson wrote:
> As for the open/closed issue, I don''t quite agree with Alan
> Coopersmith.  The current division of labor in the ARC is based on
> technical areas that (mostly) hang together due to interrelationships.
> I suspect that dividing the ARC cases by other (conflicting) criteria
> will just produce confusion, much as it does with the org chart.
> Instead, I''d rather see the ARC forced to deal with *everything*
as
> open by default.  Closed cases should be extraordinarily rare and
> should require substantial justification to launch.  They should be
> hard to do, because we really do want everything to be open, and
> because we want people considering closed work to reconsider.
> 
> Creating a closed ARC and a special ARC for open software just makes
> it too easy.
The point wasn''t having just an Open ARC, but an OpenSolaris ARC, i.e.
one tied to this community, and not OpenSPARC or OpenJava.   How does
a multi-project ARC fit into the governance structure of multiple communities?
If OpenJava chooses a governance model similar to OpenSolaris, will PSARC
members all instantly become core contributors with voting rights there
as they have here?

I admit to being biased by spending most of my time in LSARC, where maybe
10% of cases are relevant to OpenSolaris, so soliciting for OpenSolaris
community members to join wouldn''t work well when they''d be
excluded from
90% of meetings/cases.   And I''ll admit we could solve that by just
declaring the desktop part of the core platform and moving JDS to PSARC.

But that still doesn''t solve the problem of getting people to change
their
ways, and that was the larger problem I was proposing to solve.

-- 
	-Alan Coopersmith-           alan.coopersmith at sun.com
	 Sun Microsystems, Inc. - X Window System Engineering

Peter Memishian wrote:
> [ Roaming really far afield from DTrace here, but ... ]
> 
>  > I''d form 3 new ARC''s out of the membership of those
two, and realize there''s
>  > probably going to be some overlap of the membership lists:
>  > 
>  > 1) OpenSolaris ARC (OSARC) - covers all open consolidations,
including
>  >     Desktop, will include external members - all mail & case logs
will be
>  >     open from the beginning.
>  > 2) Closed Platform ARC (CPARC) - covers the closed portions of
Solaris,
>  >     both encumbered components of open consolidations (like the
various IHV
>  >     drivers in ON) and consolidations not being opened (like CDE).
> 
> This split seems challenging to me, since the legal encumbrances tied to
> our source do not align themselves on architectural boundaries.  For
> instance, if the ARC is reviewing a change that will have a widespread
> impact on GLDv3 network drivers, then all network drivers need to be
> considered at once, not just those that are open or closed source.
> 
> Or would a case like that end up going to CPARC in your proposed world?
No, it would to to OSARC, and changes to closed source drivers either 
described in terms that don''t violate their NDA''s, or held to
a separate case.
(And re-aligning the ARC''s doesn''t cause this problem - you
have the same
  problem with the current PSARC in choosing whether to file an open or closed
  case.)

OpenSolaris defined open source and open development as the development model.
Those following it should not be penalized and kept from discussions relevant
to them just because not everybody else has.   In fact, one could argue that
once the constitution is approved, closed ARC cases are no longer allowed for
any project integrating to any open gate (Section 7.10 of the draft).

Making life harder for closed components is not just an option, it should be a
design goal.   (When talking about designing our processes that is - clearly
the choices of CDDL and strong binary compatibility for things like the DDI
show support for closed components at the licensing & software levels.)

-- 
	-Alan Coopersmith-           alan.coopersmith at sun.com
	 Sun Microsystems, Inc. - X Window System Engineering

Alan Coopersmith writes:
> > Creating a closed ARC and a special ARC for open software just makes
> > it too easy.
> 
> The point wasn''t having just an Open ARC, but an OpenSolaris ARC,
i.e.
> one tied to this community, and not OpenSPARC or OpenJava.
Yes, I know.  Unfortunately, I think "OpenSolaris" is overly broad for
a single ARC, as it sounds like you''re suggesting.

Perhaps if platform software is one "project" within an ARC
"community," then it works.
>   How does
> a multi-project ARC fit into the governance structure of multiple
communities?
I think it fits in as well as C-teams and the other bits of process
fit in -- by which I mean that if you think of ''Communities''
as
strictly non-overlapping and completely independent entities with
essentially equivalent capacities, then it doesn''t fit it well at all.
If instead you allow different forms of communities that represent
different reference points in the development process, then things are
easier.

As for the follow-up, no, I don''t think the distinctions need to be
enshrined in the constitution.
> If OpenJava chooses a governance model similar to OpenSolaris, will PSARC
> members all instantly become core contributors with voting rights there
> as they have here?
I strongly prefer the "one ARC" model.  I think that full members are
mostly smart enough to figure out when they have no business voting on
a case outside their ken, if that''s what your concern is.  (If
that''s
not your concern, please elaborate.)
> I admit to being biased by spending most of my time in LSARC, where maybe
> 10% of cases are relevant to OpenSolaris, so soliciting for OpenSolaris
> community members to join wouldn''t work well when they''d
be excluded from
> 90% of meetings/cases.   And I''ll admit we could solve that by
just
> declaring the desktop part of the core platform and moving JDS to PSARC.
Eek!  ;-}

So what do we do about load balancing?
> But that still doesn''t solve the problem of getting people to
change their
> ways, and that was the larger problem I was proposing to solve.
I was addressing the "how do we make it more open" problem from a
different angle.  I think soliciting interns from non-sun.com
addresses as soon as possible would help immensely in building the
community, even if we don''t quite have the operational model set in
stone.  "Now" would be good.

-- 
James Carlson, Solaris Networking              <james.d.carlson at
sun.com>
Sun Microsystems / 1 Network Drive         71.232W   Vox +1 781 442 2084
MS UBUR02-212 / Burlington MA 01803-2757   42.496N   Fax +1 781 442 1677

Alan Coopersmith wrote:
> James Carlson wrote:
>> As for the open/closed issue, I don''t quite agree with Alan
>> Coopersmith.  The current division of labor in the ARC is based on
>> technical areas that (mostly) hang together due to interrelationships.
>> I suspect that dividing the ARC cases by other (conflicting) criteria
>> will just produce confusion, much as it does with the org chart.
>> Instead, I''d rather see the ARC forced to deal with
*everything* as
>> open by default.  Closed cases should be extraordinarily rare and
>> should require substantial justification to launch.  They should be
>> hard to do, because we really do want everything to be open, and
>> because we want people considering closed work to reconsider.
>>
>> Creating a closed ARC and a special ARC for open software just makes
>> it too easy.
>
> The point wasn''t having just an Open ARC, but an OpenSolaris ARC,
i.e.
> one tied to this community, and not OpenSPARC or OpenJava.   How does
> a multi-project ARC fit into the governance structure of multiple
> communities?
> If OpenJava chooses a governance model similar to OpenSolaris, will PSARC
> members all instantly become core contributors with voting rights there
> as they have here?
>
> I admit to being biased by spending most of my time in LSARC, where maybe
> 10% of cases are relevant to OpenSolaris, so soliciting for OpenSolaris
> community members to join wouldn''t work well when they''d
be excluded from
> 90% of meetings/cases.   And I''ll admit we could solve that by
just
> declaring the desktop part of the core platform and moving JDS to PSARC.
As I see it, there are a few "different" technology groups that need
ARC
oversight:

- Core OS (Solaris), including desktop, kernel, driver, network
protocols, etc.  These should be handled by PSARC, IMO.

- Firmware products.  OBP, NIC firmware, etc.  I think FWARC already
handles these.

- Java related stuff.  (Isn''t there already a JCP for this?)  Is there
an ARC, or is this part of what LSARC handles?

- Hardware products (e.g. SPARC servers, Opteron workstations, Storage
products, etc.).  Its not clear to me that these products have ever
gotten ARC review.

>
> But that still doesn''t solve the problem of getting people to
change
> their
> ways, and that was the larger problem I was proposing to solve.
>
Getting a full community membership onto PSARC will help.  Heck, I''m
willing to volunteer to intern for PSARC, although my value as an
"external community member" may soon be vanishing ....

    -- Garrett

Garrett D''Amore writes:
> - Core OS (Solaris), including desktop, kernel, driver, network
> protocols, etc.  These should be handled by PSARC, IMO.
This is actually broken up a little differently.

Yes, the "Core OS" does go to PSARC.  However, this doesn''t
generally
(and notably) include the desktop or most system management bits,
which instead go to LSARC.

The main issue here is workload.  If we''re going to start joining ARCs
back together, we need to look at this very closely.

Note that the boundaries are not quite as crisp as you''re suggesting
for at least two (similar) reasons:

  - interrelationships: some projects may be assigned to an ARC
    because that ARC has reviewed other cases in a related area, and
    thus has the context to handle the new project, even if it''s
    outside of the charter.

  - legacy: some things go to a particular ARC because all things of
    that ilk have always gone there, such as printing in LSARC.

The way this is handled is with a triage team consisting of the ARC
chairs.  They review the inbound cases each week and assign them to an
ARC.  This process doesn''t have to be 100% perfect, because project
teams and ARC members are free to object and suggest a better
assignment.
> - Firmware products.  OBP, NIC firmware, etc.  I think FWARC already
> handles these.
Yes.  There was also a "SNARC" that worked on such things.
> - Java related stuff.  (Isn''t there already a JCP for this?)  Is
there
> an ARC, or is this part of what LSARC handles?
JCP handles it.
> - Hardware products (e.g. SPARC servers, Opteron workstations, Storage
> products, etc.).  Its not clear to me that these products have ever
> gotten ARC review.
SHARC, SNARC, FWARC, PSARC, and other groups have handled some of
these.  You''re right, though, that not all hardware goes through the
review process in the same way that software does, though I''d
certainly argue that it ought.
> >
> > But that still doesn''t solve the problem of getting people to
change
> > their
> > ways, and that was the larger problem I was proposing to solve.
> >
> 
> Getting a full community membership onto PSARC will help.  Heck,
I''m
> willing to volunteer to intern for PSARC, although my value as an
> "external community member" may soon be vanishing ....
Cool.  There''s one.  ;-}

-- 
James Carlson, Solaris Networking              <james.d.carlson at
sun.com>
Sun Microsystems / 1 Network Drive         71.232W   Vox +1 781 442 2084
MS UBUR02-212 / Burlington MA 01803-2757   42.496N   Fax +1 781 442 1677

On Mon, Mar 12, 2007 at 09:21:29AM -0700, Alan Coopersmith wrote:
> Making life harder for closed components is not just an option, it should 
> be a
> design goal.   (When talking about designing our processes that is -
clearly
> the choices of CDDL and strong binary compatibility for things like the DDI
> show support for closed components at the licensing & software levels.)
While I agree with this, I don''t see why the existing layering of ARCs
needs to change, nor do I understand why it is necessary to create
separate closed ARCs (or, more properly, why this is a matter for any
OpenSolaris CG or the OGB).  Your solution also oversimplifies matters
(the assumption that there is basically a one-to-one mapping between
consolidation and relevant ARC is deeply flawed) and fails to address
questions such as how an open FWARC case works.

I would prefer to see all of the existing ARCs opened (disruptively if
necessary) and proprietary cases handled by the subset of members of
the appropriate ARC employed by the company/ies to whom the
information is proprietary.  For all practical purposes, that means
nothing, since the closed parts of the consolidations are invisible
and irrelevant to OpenSolaris anyway, and the broader community has no
knowledge of which companies have their own closed-source add-ons.
Arguments for or against aspects of open cases which rely on details
of how specific proprietary code works will not be considered,
although as you say the general need to allow properly-architected
closed code to continue working is relevant and needs to be
considered.  One possibility is requiring closed projects to obtain
contracts when consuming Consolidation Private interfaces.

-- 
Keith M Wesolowski		"Sir, we''re surrounded!" 
FishWorks			"Excellent; we can attack in any direction!"

On 11/03/07, Richard Lowe <richlowe at richlowe.net>
wrote:
> And what would be your (the general "you", not just Casper) plans
to help
> make the ARC and especially the C-team more practically part of OpenSolaris
> process, rather than a part of Sun process we''re exposed to from
one side,
> but not, so far, fully involved with?
I unfortunately do not have enough experience with ARC to give a
qualified answer. However, I would definitely encourage more external
members to become involved and push for even more external
communication where possible.

-- 
"Less is only more where more is no good." --Frank Lloyd Wright

Shawn Walker, Software and Systems Analyst
binarycrusader at gmail.com - http://binarycrusader.blogspot.com/

On Monday 12 March 2007 07:14 am, Karyn Ritter wrote:
> I don''t have any bright ideas for how to improve things moving
forward, but
> the community and OGB need to work with Sun to drive these critical issues.
This will be one of the bigger issues for the OGB to hash out, IMO.

OpenSolaris has made an incredible amount of progress, even though some of the 
community members commented otherwise. This will be a pivotal year as it will 
push the code outside Sun''s firewall and the community will be getting
it''s
hands around more of the process.

-- 

Alan DuBoff - Solaris x86 Engineering - IHV/OEM Group
Advocate of insourcing at Sun - hire people that care about our company!