Re: ObjC 3D engine
Re: ObjC 3D engine
- Subject: Re: ObjC 3D engine
- From: Jonathan deWerd <email@hidden>
- Date: Tue, 26 Jun 2007 10:47:18 -0600
On Jun 26, 2007, at 9:51 AM, Ilan Volow wrote:
I think what you are doing is a great idea. People new to Cocoa
need an easy way to create 3D environments. I started working on a
fork of Paolo's stuff a while back with this goal, but I never
finished work on it.
My random advice (in order of your questions)
1. Yes. Turn it into an open source project. there's definitely a
need. More and more, I'm thinking that mac developers have to be
independent from Apple and not peg our hopes and dreams and future
on what they *might* realease that *might* work for us.
I have signed up for a sourceforge project and am awaiting
confirmation. MIT license :)
2. Sign me up, sir.
Sure thing! What's your SF account?
3. I've always felt that "will people think my code is an awful
mess" fear as well. For what it's worth, all the "mature" c++
toolkits are such a mess and so inaccessible to people starting
out, I wouldn't worry about it. If you let other people look at
your stuff early enough into your project, they have some early
input that could help you avoid architectural headaches later on. I
think for a project such as yours, cleanness of code is probably
less important than excellent, high-quality tutorials for beginners
using your API.
Ok. Again, I want to make sure you don't have any misconceptions:
this is not a fully working API yet. But I am writing an orientation
to my coding style, layout, etc.
4. It's your project. But I think SVN is a better choice, because
it is a standard that people already know. You want to lower the
barrier to entry on your project.
x2. I have decided to bite the bullet and go with SVN. I will still
be keeping my private git repository as a sort of private branch,
though. I don't think I will bother with giving SVN the complete
history of my git repository since SVN is so slooowwww.
5. Figure out who your core audience is. If it really is people
starting out 3D programming for the very first time, speed power
and optimization won't really be very important whereas making an
easy to use and learn, instant gratification environment is.
I really need to make myself trust in optimization only when
necessary. Maybe being in a group will help :)
You could probably write the entire thing in Objective-C/Cocoa and
it wouldn't matter. What will really be a killer is if you
introduce a c++ toolkit that requires crazy exotic build
dependencies with their own crazy exotic build tool--it'll quickly
shoot down a novice programmer if they have to hunt down the
dependencies and build and compile themselves. Again, instant
gratification over optimization.
Right now I *think* I have it building with no non-included
dependencies. I don't know about the legalities concerning the Cg
framework, I'll have to check that.
-- Ilan
On Jun 23, 2007, at 8:51 PM, Jonathan deWerd wrote:
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
About a year ago I was looking for a good open-source objc/cocoa
3D engine. After several days of searching, the best I could find
were a few outdated and/or abandoned frameworks (note: I could
just suck at searching, so if you are working on one don't take
offense). I decided that it would be a huge learning experience to
write my own, so I did. Or, to be more accurate, I hacked together
a bunch of classes which suited my needs. In any case, I now have
about 15k lines of cocoa 3D code that I want to put to some use. I
will say again that I *do not* have anything resembling any of the
mature C++ engines out there. I just have a start, if that. But I
would like to get some of your thoughts.
1) Is it worth turning this into an open source project at all
(who would actually like a cocoa 3D engine, or does core animation
do everything you want)?
2) Who would be willing to work on it with me?
3) If it is worth turning into a project, how much should I clean
my codebase before opening it up? Going through and presenting a
unified interface, documentation, and convention set would be
essential, but a lot of work. Should I do this before opening the
project to avoid scaring devs away or would it be best to "harness
the power of open source" and do it after? For that matter, should
I get it to a working "alpha" stage on my own?
4) What should I go with as far as SCM/forums/website are
concerned? Personally I really, really love git for SCM
(seriously, check it out. It blows svn out of the water in just
about every area, and it's really easy to compile. Just one
dependancy, which is itself a clean build on a default dev install
of OSX). Should I give up all the git goodness to conform? Should
I use sourceforge, host the dev stuff myself (then look for a
better webserver when approaching production), or do something else?
And I think that's all I can think of for now. Please tell me what
you think :)
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.1 (Darwin)
iD8DBQFGfamdJFLIvUXTZHsRAhz7AJ92ZQ4wBAQxJXQFiAXtkkafCLShhgCfcdKy
C09h/YZdcwVVgb7MQdJMDZg=
=WE+s
-----END PGP SIGNATURE-----
_______________________________________________
Cocoa-dev mailing list (email@hidden)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden
Ilan Volow
"Implicit code is inherently evil, and here's the reason why:"
_______________________________________________
Cocoa-dev mailing list (email@hidden)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden