A question for the community
Hello again, FOSS Gaming Community!
I don't know about anybody else, but I for one have old code lying around from projects that never quite came to fruition. For instance, I have, right now, sitting on this very hard drive, some code for a 2D platformer (highly optimized -- 400 enemies on screen at once back when my machine only ran at 400mhz), some code for a top-down 2D RPG engine based on the aforementioned platformer code, and finally some code for the beginnings of a 3D space strategy game (check out a brief video of it here).
Here's my question: Is there anywhere out there that serves as a searchable dump for old code? Mind you, I'm not talking about project management sites like SourceForge and its ilk, but rather a place I could drop of my old code without the expectation of anyone maintaining it, so that maybe someone in the future could come along and use it for their own nefarious purposes. In particular, this site would need be searchable, much like OGA is right now -- there's not much point in dumping my code somewhere if no one will ever find it anyway. There needs to be a way to disseminate it to interested parties.
So, is there a site out there that does this? If not, is this something that people would be interested in? Do you have old code that other people might be interested in? Would you be interested in my old code (or someone else's)? Please stop by and leave your comments. I'd really like to hear what people think about this.
Peace,
Bart
- bart's blog
- Log in or register to post comments
Comments
Interesting idea. I have some unused sources sitting there too. The feature wouldn't hurt(well, the game programming is art too, isn't it?) but I don't know if anybody could find use for it. For learning purposes, tutorial are here. For some functionality programmer is looking for, libraries are here.
I find there are often a lack of examples how to do some specific thing using a particular API or framework. A well indexed and user-taggable code store could solve this, especially if it is made very easy to contribute code. This is good for two categories of code:
'deep magic' and 'straightforward but non-obvious'
and also for analyzing the usefulness of varying ways to model a given problem.
For example, metaclasses in Python (straightforward but non-obvious) really couldn't possibly have too many examples available.
What I really want myself is a unified code search engine.. google sort of does this, but snippets and tagging are essential IMO.
I think a code dump is a great idea, especially if people care to describe the code, tag it, and do stuff that makes the code reusable or good for learning special purposes. Unfortunately I think most people won't.
But the main problem to me is probably that it is a too big issue to make it as a spinoff of OGA or include it in OGA. I think it is healthier for OGA just to search for a partner site we can collaborate with ... maybe we can even convince someone from the freegamedev forum to start such a site for game code. IMO such a site needs much more attention and work than we imagine it at the beginning to work good.
Well, the upside to doing it here is that we already have the bulk of the infrastructure in place. Licensing, tags, etc.
I think that is actually more of a burden, as code needs a completely different approach and infrastructure... so only small bits could be reused and it would be easier and better IMO to start from scratch (or from drupal) again
Sometimes code can just be so specialized that it's hard for a code base to be of use as a library or API. It takes quite a lot of effort for code to be portable or reusable, and if the code is abandon-ware, and not designed for general usage in the first place, then it's extremely difficult to adapt it to something else.
Codebases are good for demonstrating design patterns, but I think a UML diagram would be easier to generate and be more concise.
On forums or self-help blog tutorials, code is always designed to be generic, descriptive and as a framework. I'm sorry but I just don't see this working out.