pkg kde talk - Jun 2008 - The next thrilling episode of "KDE version in Lenny: KDE 3.5.9 vs KDE 4.1"

{Backstory: I''ve been away from every sort of hacking for a few months
now,
but very soon will be able to devote some real time again.  Meanwhile,
I''ve
tried (not always successfully) to keep quiet about concerns I wasn''t 
actually helping to solve.}

While there are valid concerns with matters of missing features or 
obsolescence, I see the two greatest issues in this decision being 
maintainability and (wo)manpower.  While there are certainly other reasons to 
release with 4.1, those are the two issues I see as being critical, since 
they ultimately decide the viability of having KDE in Debian.

By maintainability, I''m primarily referring to the lack of upstream 
maintenance of the KDE 3 branch.  For example, there are libraries used by 
KDE 3 that have changed API, but the KDE 3 code has not been updated to build 
with the newer version.  Should there be another unexpected breakage, such as 
the one caused by Flash Player, upstream will be entirely uninterested in 
resolving it.  Any bugs in KDE 3 not related to packaging will be largely 
unfixable.

The deeper issue, though, is one of developer time and effort.  By choosing to 
include KDE3 as the primary desktop in Debian, the Debian maintainers are 
effectively the new upstream.  Considering the existing scarcity of developer 
time, in addition to post-release work on KDE 4, there is simply not enough 
time for the Debian KDE team to provide the quality of maintenance that 
people expect from a Debian release.  On a related note, since there is no 
doubt whatsoever that lenny+1 will use some form of KDE 4, that work will 
have to be done regardless of what is released with lenny.  The amount of 
additional work needed to also revive KDE 3 serves as an unnecessary burden 
on an already busy team.

The packaging of KDE 4.1 is already mostly complete and has received about as 
much testing as anything in experimental is going to get.  Other 
distributions, such as Fedora, OpenSuse, and Ubuntu, have all accomplished to 
release with a KDE 4 desktop while still providing the KDE 3 applications 
that do not yet have a KDE 4 equivalent.  Considering how for we have already 
come, I see little reason why that would not be possible for Debian.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part.
Url :
http://lists.alioth.debian.org/pipermail/pkg-kde-talk/attachments/20080609/f8b66303/attachment.pgp

On Tuesday, 10. June 2008, Matthew Rosewarne wrote: 
[...]
> While there are valid concerns with matters of missing features or 
> obsolescence, I see the two greatest issues in this decision being 
> maintainability and (wo)manpower.  While there are certainly other reasons
to
> release with 4.1, those are the two issues I see as being critical, since 
> they ultimately decide the viability of having KDE in Debian.
>
> By maintainability, I''m primarily referring to the lack of
upstream
> maintenance of the KDE 3 branch.  For example, there are libraries used by
[..cut..]

I dare to disagree ;) IMHO the KDE3 lack of maintainabilty and required
manpower is FUD ;) Lemme try to prove it:

When the debian release is out, ''all'' you still get then for
lenny is:

	o security fixes
	o fixes for data loss bugs

Where do you expect to see more of them:

	o in the solid KDE 3.5.9
	o in KDE 4.1.0 with much new code

The KDE3 svn branch sees roughly one bugfix per day.  SuSE, Redhat Enterprise,
etc, will have to fix security/dataloss bugs for 3.5.9 for more year than lenny
will be maintained.

Where do you expect KDE SVN merges of security and dataloss fixes to be easier:

	o feature frozen 3.5 BRANCH aka 3.5.9 svn -> 3.5.9 as in lenny
or
        o from 4.5.* 4.4.* 4.3.* 4.2.* -> a (theoretical) 4.1 in lenny

So my obvious(?) conclusion:  maintainability and manpower for 3.5.9 in
lenny is much lower than for a 4.1 in lenny.  IMHO: Q.E.D


Of course all ''serious'' distro will have to have 4.* pkgs. 
But the advantage
of not having 4.1 in lenny itself but in another repo is that you can
concentrate on 4.1+n as soon as it''s released and
''forget'' about 4.1.  And
that''s what the I-wants-KDE4-user-base wants to have: the latest 4.x.y.


Achim

-- 
  To me vi is Zen.  To use vi is to practice zen. Every command is
  a koan. Profound to the user, unintelligible to the uninitiated.
  You discover truth everytime you use it.
                                      -- reddy at lion.austin.ibm.com