Re: Google Summer of Code participation

[Carl Fredrik Hammar]

Hi,

<olafBuddenhagen@gmx.net> writes:

>> > Especially the project list needs work. Some of the project
>> > descriptions are only skeletons; I lack the knowledge to fill in the
>> > details. But I know that some of you have much better understanding
>> > about some of these projects.
>> 
>> I noticed ``Translator stacking mechanism'' in there.  I'm planning on
>> getting some serious work done on this soon, as I'm just getting out
>> of a particularly stressful quarter (just one exam left now :-), and
>> I'm entering a particularly calm quarter and I will have plenty of
>> spare time on my hands.
>
> Oh, I didn't know that. Good to hear, though :-)
>
> In this case, maybe it's better to take it off the ideas list
> alltogether...

I took it down.  Since I'm most suited for the project (and obviously
already familiar with it), and since I'd rather do it individually
then mentor someone else doing it.  (Bringing a student up to speed
seems almost as much work as just doing it myself.  :-)

> It's not like there are no other project ideas :-)

Indeed, the list is actually quite impressive.  If nothing else we
should be accepted as an organization simply because there's so much
to do.  :-)

>> So I'm not sure how much will be left for a student to do.  But then
>> again life happens, and should I fail to get started with it, it would
>> be a nice project.
>
> Well, we could still propose the task again for next year's SoC, if it
> turns out you can't do it after all. By no means I want to take this
> project away from you :-)

This is reasonable, I wouldn't consider it a high priority project
that needs to be completed any time soon.

>> Of course, I'm more than willing to sign up for another GSoC to get
>> this done if we have a shortage of volunteers.  ;-)
>
> I was actually partially thinking of that when putting it in there.
>
> There is no guarantee however that the project would actually get
> accepted. I very much doubt that we could find someone better for this
> task than you, considering the knowledge you already have about it, and
> the good work you did previously... But then, we do not know yet how
> many brilliant students will sign up for the other tasks :-) The limited
> number of available slots always makes it a tough choice...

The best outcome of GSoC would be new long term contributers, so I
think it's best for me to step aside and let someone else have the
slot.

I would only be comfortable with taking a slot if we were given
unexpectedly many slots, or there were only a few or iffy
applications.  But this seems unlikely, so I don't think I'll waste
time writing an application /just in case/.

>> Not sure if I'm experienced enough to be an actual mentor, but I'm
>> more than willing help out when it comes to fitting the ``Sound
>> support'' and ``TCP/IP stack'' projects into the translator stacking
>> framework.
>
> Well, to fit something into it, first the framework needs to exist :-)

I would be developing it in parallel, of course.  :-)

> Unlike some others, I do not believe that libchannel is a prerequisite
> to implement these things. On the contrary, I think it makes more sense
> to implement them as standalone translators first, and only think of
> optimized stacking mechanisms later on.

Agreed.

> In fact, I think that it would be better even to have the actual
> applications there before writing the stacking framework -- much better
> to test it in the actual use cases, rather than designing it on a purely
> abstract basis, and trying to fit the applications into that later on...

This is closer to what I meant to say, perhaps I should have said
``fit them together'', i.e. making changes to both.

> You would soon have to decide whether you want to apply as student
> again, or mentor the project. (Or just do it independently as you
> planned...) It's not possible to do both :-)

But I could mentor it and be a student for another project.  ;-)

Regards,
  Fredrik