GSOC: Deadline soon!

[Andrew Godwin]

Hi everyone,

It's great to see all the GSOC proposals on the list and the feedback being useful, but I feel I should remind you all that the deadline is this Friday, May 3rd, at 7pm UTC (so 11am if you're on the west coast of the US, for example).

Melange, the GSOC software, can be a little hard to understand sometimes so please try and get an inital copy of your application in with plenty of time to spare - you can always save new versions right up to the deadline.

Andrew

[Satinderpal Singh]

Hi Andrew,
I wish to be a part of GSoC, and i have a plan to improve some
features of forms of the Django, like adding different formats such as
tables with different layouts. Also some improvements in the default
form settings in the Django, like adding the features of dynamic form
to add rows and columns in the forms.

> --
> You received this message because you are subscribed to the Google Groups
> "Django developers" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to django-develop...@googlegroups.com.
> To post to this group, send email to django-d...@googlegroups.com.
> Visit this group at http://groups.google.com/group/django-developers?hl=en.
> For more options, visit https://groups.google.com/groups/opt_out.
>
>



--
Satinderpal Singh
http://devplace.in/~satinder/wordpress/
http://satindergoraya.blogspot.in/
http://satindergoraya91.blogspot.in/

[Jacob Kaplan-Moss]

On Wed, May 1, 2013 at 7:38 AM, Satinderpal Singh
<satinder...@gmail.com> wrote:
> I wish to be a part of GSoC, and i have a plan to improve some
> features of forms of the Django, like adding different formats such as
> tables with different layouts. Also some improvements in the default
> form settings in the Django, like adding the features of dynamic form
> to add rows and columns in the forms.

Well, you've got about 48 hours to get your proposal in. It's not a
lot of time, especially because you need to post it *here* before
putting into Melange (the GSoC system), but if you get started this
moment you may still have a shot.

If you'd like a serious chance at getting accepted, please:

* Read our student guidlines, at
https://code.djangoproject.com/wiki/SummerOfCode2013#Students.

* Write a proposal that follows those guidelines. If you want to see
examples of some good proposals, look at the threads here started by
Michal Petrucha
(https://groups.google.com/d/msg/django-developers/e0-rOIkrXaQ/w5aiW_R6aFYJ)
and Christopher Medrela
(https://groups.google.com/d/msg/django-developers/e0-rOIkrXaQ/w5aiW_R6aFYJ);
both are well-written, solid proposals that show what we expect to see
from students.

* Show that you understand the problem space. For your particular
project, there's quite a history here. Forms were the subject of a
previous Summer of Code project which failed. You need to demonstrate
that you understand why the past project failed and show that you have
a plan to succeed this time. There's also quite a few third-party
forms libraries that address the concerns you have; you should show
that you are aware of these projects or at least plan to spend some of
your project time researching.

* Submit your proposal here. Once you have a proposal, you need to
post it here for comment before submitting it to Melange.
Realistically, to stand a chance of being accepted you need to have
your proposal posted here in the next 24 hours. If you get your
proposal posted on this list within 24 hours I'll commit to giving you
feedback in time to revise your proposal before the submission
deadline.

* Get your revised proposal in before the deadline.

Good luck,

Jacob

[Nicolas Bouliane]

I have done as suggested and submitted my proposal in the mailing list. I heard about the Google Summer of Code very recently, but hopefully I'll be able to submit an accurate and sensible proposal on time!

Coincidentally, I am also proposing changes to the Forms API, and I must say it is very hard to find more information about the 2011 GSoC project and the reason of its failure. Is there a better source than the commit log to learn more about it?

[Carl Meyer]

Hi Nicolas,

On 05/01/2013 10:52 AM, Nicolas Bouliane wrote:
> I have done as suggested and submitted my proposal in the mailing list.

> I heard about the Google Summer of Code /very/ recently, but hopefully

> I'll be able to submit an accurate and sensible proposal on time!
>
> Coincidentally, I am also proposing changes to the Forms API, and I must
> say it is very hard to find more information about the 2011 GSoC project
> and the reason of its failure. Is there a better source than the commit
> log to learn more about it?

Probably searching the archives of this mailing list during that summer
would be your best option.

I was the mentor for that GSoC project, and the reason for its failure
is that the Django template language proved too slow for the
template-based system we proposed for rendering form widgets and form
layouts; form-rendering speed suffered by an order of magnitude in some
cases. There was another GSoC project that summer to speed up the
template language and for a time we'd hoped that would provide a
solution, but that project also failed in the end.

Carl

[Nicolas Bouliane]

That's what I gathered from the various tidbits of information, but I will try looking at the archives.

It's a shame, because it was a fantastic idea. Even to this day, it's pretty hard to properly customize form elements.

Carl

--
You received this message because you are subscribed to a topic in the Google Groups "Django developers" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/django-developers/xrP6-Fx0Jbs/unsubscribe?hl=en.
To unsubscribe from this group and all its topics, send an email to django-develop...@googlegroups.com.

To post to this group, send email to django-d...@googlegroups.com.
Visit this group at http://groups.google.com/group/django-developers?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.

--
Nicolas Bouliane
(450) 577-1663 · nicolasbouliane.com

[Carl Meyer]

On 05/01/2013 11:04 AM, Carl Meyer wrote:
> I was the mentor for that GSoC project, and the reason for its failure
> is that the Django template language proved too slow for the
> template-based system we proposed for rendering form widgets and form
> layouts; form-rendering speed suffered by an order of magnitude in some
> cases. There was another GSoC project that summer to speed up the
> template language and for a time we'd hoped that would provide a
> solution, but that project also failed in the end.

I should mention that the project was only a partial failure -- thanks
to the efforts of Jannis Leidel, Bruno Reni�, and Gregor Muellegger (the
GSoC student), his work was later merged into the django-floppyforms[1]
project. So it's available for anyone who wants to use it, we just
weren't able to make it the default in Django core.

Carl

[1] http://django-floppyforms.readthedocs.org/en/latest/

[Nicolas Bouliane]

Reading the description of django-floppyforms, I realize most of these features are overlapping with those in my proposal.

What do you think would be advisable from that point? 

On Wed, May 1, 2013 at 1:11 PM, Carl Meyer <ca...@oddbird.net> wrote:

On 05/01/2013 11:04 AM, Carl Meyer wrote:
> I was the mentor for that GSoC project, and the reason for its failure
> is that the Django template language proved too slow for the
> template-based system we proposed for rendering form widgets and form
> layouts; form-rendering speed suffered by an order of magnitude in some
> cases. There was another GSoC project that summer to speed up the
> template language and for a time we'd hoped that would provide a
> solution, but that project also failed in the end.

I should mention that the project was only a partial failure -- thanks

to the efforts of Jannis Leidel, Bruno Renié, and Gregor Muellegger (the

GSoC student), his work was later merged into the django-floppyforms[1]
project. So it's available for anyone who wants to use it, we just
weren't able to make it the default in Django core.

Carl

[1] http://django-floppyforms.readthedocs.org/en/latest/

--
You received this message because you are subscribed to a topic in the Google Groups "Django developers" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/django-developers/xrP6-Fx0Jbs/unsubscribe?hl=en.
To unsubscribe from this group and all its topics, send an email to django-develop...@googlegroups.com.
To post to this group, send email to django-d...@googlegroups.com.
Visit this group at http://groups.google.com/group/django-developers?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.

[Satinderpal Singh]

Thanks for the reply, i will be here with a proposal in a short time.

> Good luck,
>
> Jacob

[Carl Meyer]

On 05/01/2013 11:18 AM, Nicolas Bouliane wrote:
> Reading the description of django-floppyforms, I realize most of these
> features are overlapping with those in my proposal.
>
> What do you think would be advisable from that point?

In general, I think your proposal would benefit from additional research
and specifics, including a detailed schedule estimate (successful
proposals usually include a schedule with at minimum weekly granularity).

I believe every item in your proposal (with the exception of
form.as_label()) already has a ticket in Trac with an extensive history
of discussion and patches. You mention only one of those existing
tickets (#9321), and even there you don't discuss how you propose to
find a path forward that resolves the concerns that have been raised on
that ticket over the years.

I also am personally inclined against the form.as_label() and fieldsets
proposals. I think that form HTML presentation (including fieldsets)
properly belongs in templates, not in Python code, so I dislike adding
more as_* methods or more Python-level layout definitions.

I don't think overlap with django-floppyforms (or any other third-party
project) is itself problematic for a GSoC proposal, but the proposal
certainly needs to discuss the overlap, understand why the functionality
is currently in a third-party project, make a good case for why it
should be in core instead, and be clear about where existing code will
be reused or new code written.

Hope this is helpful - good luck with your proposal!

Carl

[Nicolas Bouliane]

It seems that I have underestimated the barrier of entry to the Google Summer of Code. With so many hoops to jump through and very little time left, I will not play against the odds.

I still look forward to contributing to Django, but not as a GSoC student.

Carl

--
You received this message because you are subscribed to a topic in the Google Groups "Django developers" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/django-developers/xrP6-Fx0Jbs/unsubscribe?hl=en.
To unsubscribe from this group and all its topics, send an email to django-develop...@googlegroups.com.
To post to this group, send email to django-d...@googlegroups.com.
Visit this group at http://groups.google.com/group/django-developers?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.

[Carl Meyer]

On 05/01/2013 01:38 PM, Nicolas Bouliane wrote:
> It seems that I have underestimated the barrier of entry to the Google
> Summer of Code. With so many hoops to jump through and very little time
> left, I will not play against the odds.

I understand; it is late in the process to be starting on a proposal.

> I still look forward to contributing to Django, but not as a GSoC student.

Glad to hear it - look forward to working with you as a contributor!

Carl

[Nicolas Bouliane]

Yes indeed! I heard about it around the April 27. With a few more days left, I might have been able to review my proposal, but I will simply concentrate on finding work this summer. I would still love to have my proposal chosen, but I cannot put all of my eggs in that basket.

Carl

--
You received this message because you are subscribed to a topic in the Google Groups "Django developers" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/django-developers/xrP6-Fx0Jbs/unsubscribe?hl=en.
To unsubscribe from this group and all its topics, send an email to django-develop...@googlegroups.com.
To post to this group, send email to django-d...@googlegroups.com.
Visit this group at http://groups.google.com/group/django-developers?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.

[Russell Keith-Magee]

Hi Nicholas,

I agree that there's a lot of overlap between your proposal and django-floppyforms -- but I still think there's some scope for a very interesting GSoC project. There may be only 2 days left, but that's still time for a couple more revisions before the deadline.

I'll leave some comments on the thread where you provided your proposal.

Yours,

Russ Magee %-)

You received this message because you are subscribed to the Google Groups "Django developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to django-develop...@googlegroups.com.

[Nicolas Bouliane]

Thank you for your kind words! I will try to find a few hours to correctly edit it.