I see everywhere people looking to create a team that will make games and proprietary engines from scratch. This is great, it means that the community hobbyists are interested in real game development. The problem I see is that everyone wants to make a seperate team, all by itself. I'm sure that each person, or small group looking to get larger has a few good sized modules of code that are efficient and could see a future in the game engine market. I fear that as people stress on for years through the team development phase, that thousands of lines of code and effective modules will get lost in the shuffle.

I suggest that someone organize a pool of people looking to create an engine from scratch. We build one team at a time, seperate into specific projects that are similar where people can make some compromises in order to get some finished product created. I want to make a game engine from scratch, sure, but every bit of it doesn't have to be my own code. It would drive me crazy trying to perfect every single module, and it has been driving me crazy.

I work on graphics implementations. Mesh code, textures, systems that handle this data. If I could focus on that specifically, I could get alot of work done. Unfortunatly I have to split my time across all the fields, like script managment, camera viewpoints, game logic, and other misc topics. If I could get on board with a good structured team that was willing to make some comprimises I coudl see a real future doing this. I just hope that people can get past thier egos in order for the greater picture to come into focus.

This is my coding horror. I hope you like it. :)

It's about programming. I could go into details about the code, like the old .obj file loading implementation that found got working, the emitter and particle code that I wrote that allows for volume emitters and field constraints, the mesh and texture implementations I wrote from scratch, the fact that I refuse to use anything but openGL 1 until I have mastered it thouroughly, the windows programming that I fear is gracously out dated, the colission and detection code I'm working on, and everything else in between, but it's about coding horrors and I figured the last thing you all would want to see is more programming examples in a forum called coding horrors. I just fear that this, and alot of other indie projects end up on the cutting room floor because they can't keep up with the fast paced industry, and I wanted that fact to be known. I care about my fellow upstart programmers and I'd like to see something done about it.

Move it wherever you want, wherever you think that people will see it. :)

Is this a coding horror?

When has cooperative coding among unpaid amateurs not resulted in a coding horror?

@OP: it's been tried more than a few times. But without a seasoned software architect and a strong technical lead, I doubt you'll get very far.

Languages, coding styles, the idea guy , bad management, programmers' skill, no incentive .... these doom a lot of "free labor" projects to fail .

When did this turn from being something fun and explorative like games are supposed to be into "free labor"? What, are we all just slaves doomed to repeat the process of work unfufilling jobs and pay bills? With a system that relies on the bottom line as it's sole progressive incentive, it's easy to see why there are so many games out there with little to no innovation or soul. It's sad to see so much talent get wiped out by a bunch of money hungry, back-breaking, slave laborers. This is my dream. This is my nightmare.

I would think a project produces good results if all people involved share the same goal and someone is watching for code quality strictly. If instead everyone just wants their pet feature included and refactoring gets postponed it will probably generate a mess over time. Though that is not really related to these people being paid or not or the code being open source or not.

If you force people to cooperate on your dream engine they will just put in what they need themself using shortcuts that mess up the design. Then they will jump off, because its a mess. Then you have to clean up yourself.

There it may be easier to collect usable code yourself.

yeah.

my usual experience with open-source has been scavenging whatever parts I need, and very rarely contributing back to the main projects (many/most FOSS projects IME basically tend to operate where only a small inner circle has direct access to the repository, and everyone else has to send in diff patches, but usually it is enough of a hassle that it isn't particularly worthwhile, especially if it is for something "weird" that probably no one else will really find all that useful).

sometimes ugly hacks are needed to get things to build or work, often in areas where the wider functioning or relation to other parts of the code is not clearly understood.

...

sometimes I am left to think that, preferable to using "the one true source code" for a given project, it would be nicer if all of the data/formats/... in use were adequately documented, such that it would be possible for multiple implementations of everything to exist and to cooperate in terms of shared data and file-formats.

for example, I am more in favor of open standards, rather than treating particular libraries or implementations as canonical.

now, can this be applied to games?:

I think it would be nice if people could better nail down sets of file-formats and maybe work towards a set of inter-engine de-facto standards;

ideally they should be descriptive (describing various common use-cases and file-formats and encouraging reuse when applicable, ...), rather than prescriptive ("for use-case X, use format Y");

...

things that could be nicer if more standardized:

representations of textures and texture metadata;

video formats and codecs which are "actually useful" to games development (such as having an alpha channel, ...);

audio formats useful for real-time mixing;

3D model formats (characters, props, ...);

world map data;

various engine conventions (coordinate spaces and units);

...

(elaboration removed for length reasons...).

but, yeah, the general idea here is that if common file-formats catch on, then hopefully the current mess that is engines and tools would begin to clear itself up, with hopefully much more interchangeable engines and tools and making it easier to move content from one game to another, ...

then it wont matter as much who's code is being used, ...

like, take for example, the relative standardization of graphics formats:

with this, it doesn't really matter as much which applications people use to create and edit the images, since the applications are largely interchangeable.