Nouveau - Mar 2019 - [Piglit] X.Org GSoC 2019 - Student Application Period

On Tue 2019-03-26 @ 10:40:49 AM, Ilia Mirkin wrote:
> Just looked over the projects... they all seem valid
Thank you for taking the time to have a look and provide feedback!
> but are there
> people who could realistically mentor a GSoC student for these? IMHO
> unless mentors can be identified, these should all be archived.
Attracting mentors is probably the hardest part of running GSoC. Perhaps if a
student shows interest in a given topic, someone might be inspired to sign up
as a mentor?

Keeping the list of ideas up-to-date and finding mentors are the two things
I'm hoping to make a noticeable improvement to this year. It also ties in
strongly with:

	https://gitlab.freedesktop.org/freedesktop/freedesktop/issues/76

Best regards,
	Trevor

On Tue, Mar 26, 2019 at 10:56 AM Trevor Woerner <twoerner at gmail.com>
wrote:
>
> On Tue 2019-03-26 @ 10:40:49 AM, Ilia Mirkin wrote:
> > Just looked over the projects... they all seem valid
>
> Thank you for taking the time to have a look and provide feedback!
>
> > but are there
> > people who could realistically mentor a GSoC student for these? IMHO
> > unless mentors can be identified, these should all be archived.
>
> Attracting mentors is probably the hardest part of running GSoC. Perhaps if
a
> student shows interest in a given topic, someone might be inspired to sign
up
> as a mentor?
>
> Keeping the list of ideas up-to-date and finding mentors are the two things
> I'm hoping to make a noticeable improvement to this year. It also ties
in
> strongly with:
>
>         https://gitlab.freedesktop.org/freedesktop/freedesktop/issues/76
>
> Best regards,
>         Trevor
BTW - I meant my comment to only apply to nouveau-related projects.
Both in terms of validity, as well as mentorship-unavailability.

It sucks for a student to come in, think that some project is
available to work on, only to find that there's no one who can mentor
them. I don't know how often that happens, but I bet it's
"sometimes".
Better to get them to dedicate themselves to projects that do have
mentors available.

[And yeah, I know these are meant as "ideas", but I think in practice
it's "here's a list of projects for you to select from". Only
a very
rare student will come up with their own idea.]

  -ilia

Thanks for reaching out on the matter Ilia.
I am not to be relied on. If you're lucky, I might be able to take some 
questions about past work, but cannot make a solid time commitment that 
warrants being a mentor.
Of course, the mentor does not necessarily have to have the deepest 
understanding of the subject matter. *At the end of the GSOC* a good 
student should be trusted with having more knowledge on the specific 
topic than the mentor. That being said, DVFS is fidgetty and the outcome 
is binary (it either works or crashes your machine spectacularly), so 
it's difficult to get someone started and keep 'em motivated if as a 
mentor you don't have experience on the topic. As such, I'd consider it 
a high risk topic to promote unless one of the relative veteran 
developers (Ben, Martin, Karol, or Lyude in the case of clock/power 
gating) is keen to mentor.

Roy


On 26/03/2019 15:07, Ilia Mirkin wrote:
> On Tue, Mar 26, 2019 at 10:56 AM Trevor Woerner <twoerner at
gmail.com> wrote:
>>
>> On Tue 2019-03-26 @ 10:40:49 AM, Ilia Mirkin wrote:
>>> Just looked over the projects... they all seem valid
>>
>> Thank you for taking the time to have a look and provide feedback!
>>
>>> but are there
>>> people who could realistically mentor a GSoC student for these?
IMHO
>>> unless mentors can be identified, these should all be archived.
>>
>> Attracting mentors is probably the hardest part of running GSoC.
Perhaps if a
>> student shows interest in a given topic, someone might be inspired to
sign up
>> as a mentor?
>>
>> Keeping the list of ideas up-to-date and finding mentors are the two
things
>> I'm hoping to make a noticeable improvement to this year. It also
ties in
>> strongly with:
>>
>>         
https://gitlab.freedesktop.org/freedesktop/freedesktop/issues/76
>>
>> Best regards,
>>          Trevor
> 
> BTW - I meant my comment to only apply to nouveau-related projects.
> Both in terms of validity, as well as mentorship-unavailability.
> 
> It sucks for a student to come in, think that some project is
> available to work on, only to find that there's no one who can mentor
> them. I don't know how often that happens, but I bet it's
"sometimes".
> Better to get them to dedicate themselves to projects that do have
> mentors available.
> 
> [And yeah, I know these are meant as "ideas", but I think in
practice
> it's "here's a list of projects for you to select from".
Only a very
> rare student will come up with their own idea.]
> 
>    -ilia
> _______________________________________________
> Nouveau mailing list
> Nouveau at lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/nouveau
>