How to save modern gaming

To solve or mitigate many of the problems recently developed games have had, I suggest a new philosophy of game development.

The developer creates a functional but generic 3D and client-server engine for a game, for example an FPS, RTS, racing game or MMORPG, taking care to document all code well. They can create a short single-player and a good multiplayer mode for it. Since the game is quite short and the quality of the graphics is not state-of-the-art, the budgets are not even close to normal games.

The client build's source code is posted under a free software license. The server code is only on run corporate, in-house servers with large bandwidth and is never distributed. The compiled game is posted as a free download on several PC and console software distribution clients.

To run the game, it is necessary to subscribe on a monthly basis to the developer. This should be made as painless as possible using the distribution client's usual paying service.

As soon as the money begins to come in, the developers work on very large custom content for their engine and release new maps, player models, weapons, game modes constantly and on a regular schedule (ideally weekly).

Note it will be impossible to pirate because traffic must pass through a company server and no two identical keys can use the system at once (and keys regularly expire.) Continuous free trial oppotunities that give temporarily valid keys should be given to encourage downloads, in addition the short single player component serves as a free demo. New content is only distributed to those with valid keys (and with the constant release of new content, pirated content will be worthless).

The free software nature of the client along with the good documentation will encourage player mods. Developers should make it easy to submit new content (and then offer it in the weekly updates) and even pay small, fixed sums for major mods like gameplay modes or single-player campaigns. Larger bounties or lifetime subscriptions could be offered for any users that could code a better, faster or more flexible core component, like for example improving the graphics engine - again, less money from the developer is needed.

The problem of developing games for a multitude of platforms and system requirements will be solved because the client can be ported by users to any architecture at no cost to the developer. The developer can assist with porting too. Since the source code can be modified anywhere, it can be developed on a Windows/Mac/Linux computer and downloaded to consoles too.

Revenue will be large (due to the much larger audience who will try your game) and constant (subscription-based), while initial development costs will be small compared to many epic FPSs. There will be incentive to keep developers on the project, and the constant release of tons of gameplay content, often developed for free by interested users, will increase the replay value. Balance tweaking and bugfixes will also be constant leading to a very stable and professional game that will even improve as technology improves - the cumulative effect of the community plus professionals will give a sort of hombrew-Crysis level of polish eventually. The potential for innovation will be high due to the large number of "potential" developers for the game.

I think it is the perfect solution to all of the problems modern gaming faces. If I had 5 million dollars to create a basic engine...

they could just use Valve's publishing software and add their games to Steam and have access to 15+ million users at once.

So you're just describing if Valve made an MMO?

I'd like to be able to play my games long after their lifecycle has ended so this idea fails.

Interesting idea. There are just a few points I would like to make/discuss:

1.  It is my understanding that the entire client engine (both gameplay and distribution) is the part which becomes open source. I see three initial issues with this:

• It is my understanding that the main way DRM works is through obscurity. Having the client open sourced removes part of the obscurity. I'm not necesarilly referring to directly being able to pirate games, but you need to be able to stop cheating and other exploits too. This isn't an area I'm entirely up to speed on, so perhaps someone can explain if this is actually feasible.
• Having a fully open source client means a 3rd party only needs to develop a server backend to distribute their own content (or they can outright lift the engine and make their own standalone game). While you might not see any problem with this and invite the competition, I can see investors being upset when their money is going towards developing software for the competition.
• If someone can gain access to the original content of another game (since you need a copy locally), what is to stop them connecting to a rogue server with a modified client? This basically stems from the previous two points.

2. You suggest that single player content serves as a free demo. As someone who dislikes online gaming, will this system have anything to benifit me when I'm after a solid single player experience?

3. Is the license GPL or BSD based? I'm assuming GPL as you have mentioned others submitting back to it. How do you deal with the issue of multiple people modifying the same parts to handle things differently?

Also forgot to add-- it isn't impossible to pirate. Someone just has to setup a dummy server that the game (or portal) thinks is the server. Granted they won't be able to play with other people unless they are using the dummy server too but it still isn't impossible to pirate.

I'm also not saying that would be easy and it would take some time but it is something that can and has been done before.

Yes, you can always cheat, but having proper server-side protection can and does work. The server would simply need to check for any "illegal" movements made by the client: the client will just act as a terminal for issuing commands rather than processing the actual change in location. The danger ought not to be any more than closed-source clients if done properly, especially if te incentive to cheat is lessened by not having virtual currency or valuables.

I agree that new a server backend could be made, but per above most of the work in processing movement would be done on the server, so most of the game engine would have to be rewritten. It just woldn't be worth it. The third-party would also find it hard to repackage and redistribute the content and graphics upgrades (illegally) to keep parity with the main game.

Yes, if you want to heavily modify the client and recreate a server, it is possible, but to be honest it would be easier to create your own game. Real players are likely to stick to the official edition: look at Red Hat and CentOS.

To be continued...

I wouldn't put it past people, if there is demand it will be done. Just of the top of my head, bnetd (Battle.net clone) is a decent example, and that was created through reverse engineering. If you are supplying them full documentation, this makes building a clone significantly easier. Many MMOs also have private servers available, too. If that can be done, I can't see why the client would need to be modified either, other than to redirect it to the new server.

I can't see repackaging and redistribution of upgrade to be a problem. Updates for games available through Steam are often cracked and repackaged as an update. TF2 often has its updates cracked and pirated and are repackaged just as the update itself. Portal was available as only the additional content (not the whole source engine) to pirates as well.