[Bf-committers] GSoC ideas page
rogerwickes at yahoo.com
Sat Mar 20 17:20:28 CET 2010
you're arguing over the relevance and importance of domain knowledge, btw.
you all agree that domain knowledge is important, and that the more the better.
however, is it relevant to accepting a GSOC proposal? Obviously, less domain
knowledge will require the student to spend time acquiring that knowledge.
Since time is fixed, the question is how much of it will be spend acquiring
that knowledge versus implementing that knowledge into code?
That answer varies based on the student submitting the proposal,
and cannot be resolved in the general context. hence endless posturing.
So the question is, is the proposal acceptable based on the student's
domain knowledge, with the domains required are:
- functionality desired; theoretical knowledge
- blender codebase (C++/Python, depending on proposal)
- coding in general in the target language and environment (SVN, etc)
Check out my website at www.rogerwickes.com for a good deal on my book and
training course, as well as information about my latest activities. Use coupon
Papasmurf for $15 off!
From: Matt Ebb <matt at mke3.net>
To: bf-blender developers <bf-committers at blender.org>
Sent: Fri, March 19, 2010 6:32:02 PM
Subject: Re: [Bf-committers] GSoC ideas page
On 20/03/2010, at 08:12 , Tom M wrote:
>> - We need students with proven competence in the area they want to
>> contribute to
> That I disagree with. A coder should be proven competent by dint of
> previous coding experience in something. I'd bet 99% of kernel and
> compiler coders could jump in to Blender without difficulty for
> instance. Most of our successful project participants did not have
> previous coding experience in the area they wanted to contribute to.
No, I think Brecht is absolutely right. There is not a 'hierarchy of
coding talent' as if it's a generic commodity, that starts with kernel
programmers or whatever at the top. Different people have different
areas of experience and ability. I most probably would certainly not
want a compiler programmer working on anything that involves user
interaction or workflow, if they aren't already quite experienced and
familiar with Blender and how 3D artists work.
The same thing goes for plenty of other areas, rendering for example
demands specific knowledge and techniques that not a lot of coders in
general are familiar with. Without a decent level of experience and
awareness of the domain, people will make the same old rookie
mistakes, and generally won't know how to actually make something
useful in the context of 3D production. Again in the rendering
example, I'd wager than even most specialist academics/researchers in
the rendering field would need to learn a lot in order to make a
useful contribution to Blender, and tools that are useful in practical
Bf-committers mailing list
Bf-committers at blender.org
More information about the Bf-committers