Will We Succeed? We Think So!

Lots of projects have sprung up with similar goals to WorldForge. Graphical online roleplaying games are a dime a dozen. What makes WorldForge special, how can it possibly succeed where so many have failed? The following is from a letter Bryce wrote for a journalist who wrote an article for Salon Magazine about WorldForge back in 1998.

One needs to dig into the mail archives and the CVS repository to see the true wealth of work that's going on; much of it hasn't made it to the website yet. The mailing list has been quite impressive; Jack (the mailing list administrator) says that the within the first month of the project alone, the archive comprised over 6 MB. Even more impressive is that nearly all of this mass is composed of productive contributions. The mailing list has a higher signal to noise ratio than any I've worked on.

[Since this was written we've managed to get a good deal of this initial brainstorming work fleshed out and moved into the website. As you'll see, things are still a tad disorganized, but getting better! The size of the mail archive is growing steadily, and is measured in the tens of thousands of messages.]

The CVS repository is also revealing. Nearly a dozen developers have been continuously adding and updating source code for review and testing. It's too early to expect this code to do anything useful, as much of it is still rather conceptual in nature. The amount and quality of code on the site is impressive given that the project has only been underway for a matter of weeks.

[Indeed, the CVS repository continues to grow. Code alone takes up several megs of space, art and music take up another 1GB beyond that. And today not only does the code actually do stuff, but we're finally well on our way towards achieving a playable game system.]

The WorldForge team has an excellent group of core developers with a great deal more programming experience than one would expect. This point was driven home to me when one of the developers commented to me, "Heh, I do multi-threaded, interprocess communications for breakfast here where I work."

Why is this project of interest to so many people?

The idea of making an online roleplaying game has been an everpresent part of the Internet since its earliest days. The desire to capture the comradery and pleasure of the interactive roleplaying found in paper-and-pencil games like Runequest or GURPS is quite strong in the Internet gaming community. In fact, Multi-User Dungeons (MUDs) have been a fad practically since the birth of the Internet. The current interest in free software is nothing new to the MUD community, as it has long been the tradition to share MUD codes to allow continual development and refinement. People with access to a server will set up and run one of these codes, often modifying its rules to suit their own desires. Some MUDs emphasize combat, others are more focused on roleplaying; one allows players to hunt and kill one another, others enact restrictions against such behavior. This is a boon to the player community because they have a choice of many styles of play to choose from, and if the rules (or theme, or server performance, or whatever) on one MUD are not to their liking, they have many more to choose from -- or can even set up their own!

[Playerkilling is a frequently asked question, and bears addressing here as well as elsewhere: Just as different flavors of MUDs allow or disallow things like playerkilling, so too do we expect different flavors of WorldForge servers to spring up with various controversial rules enabled or disabled. Remember, the key to success is variety and choice!]

One of the several long time "holy grails" of the MUD community has been the desire for graphics. Nearly all MUDs are text based, "Zork-style". Many open source efforts to create "graphical MUDs" have been undertaken, but to date few or none have succeeded. The problem is that the underlying MUD engine, developed long ago when computer processing power was much more limited than it is today, is not sophisticated enough to support graphical interfaces.

Commercial companies have attempted to provide this Holy Grail to the gaming community. There is certainly a financial benefit to be had. Because of the addictiveness of these games, players often spend many hours each day playing, and by charging the players by the minute, a gaming company can make a great deal of money. This promise of wealth encourages them to invest money into developing a very attractive graphical game. And since the MUD community hadn't been able to provide this capability, it was felt that maybe commercial investment could.

Yet experience has shown that while there are significant benefits to the commercial game development approach, much is lost. The codebase is closed, thus preventing people from creating alternate worlds focused on a particular genre or style of play, and preventing people from eliminating the bugs that trouble so many. The game provider is torn between allocating developers to add new features or to fix bugs. All too often it seems like commercial marketing needs drive this choice more to the former than the latter. Another troubling result is that when the company ceases to provide the gaming service, the source code to the game is effectively lost to the community, thus preventing the continuity of gaming that has been such a feature in the MUD community.

[As well, players on these commercial servers expect that, since they are paying good money, the company must be responsive to player demands, leaving no room for consistency of vision on the part of the server admin. These companies are also forced to adopt a "one-world-to-please-all" attitude which can leave the less numerous (or simply less vocal) players out in the cold.]

WorldForge, I believe, is an attempt to merge the technical triumphs of the recent commercial online games with the spirit and freedom of the MUD community. Its goals are to provide a modern core engine for building online gaming communities to be shared freely. People who strongly desire Hack and Slash (HAS) gaming can adjust the system to suit them, while those desiring a game focused more on roleplaying, world building, or community building can adjust the system to meet their needs as well.

This "scratches a big itch." Lots of people are hooked on online roleplaying games yet are very disappointed with the offerings of commercial companies. They are the unserved minorities, the frustrated majorities, and those whose hands itch to tinker under the hood. They want something better, and they're enthusiastic about WorldForge and willing to lend a hand.

Don't all Net Projects Fail?

Many of us have been involved in lots of other net projects, nearly all of which fail. We know the signs: Endless noisy banter, flamewars, silly arguments, committee-like management-hell, difficulty communicating and producing documentation, political power grabbing, growing bitterness, and finally decreasing participation. The best developers depart and the signal to noise ratio falls to zero; finally everyone else departs. We've see this happen so often with RPG and computer game related projects (both of which are popular among the young and inexperienced) that many of us reflexivly think, "Oh, that'll fail" whenever we see such projects.

But many of the things that cause similar projects to fail are not present in the WorldForge project. A surprisingly large percentage of the developers are experienced, mature programmers, artists, and writers; resources are not a problem, as mailing lists, news groups, CVS archives, web sites, and irc channels sprung up left and right; and (most of all) the morale of the group is high. Ideas are fresh yet quite sophisticated, and it's clear that many of the participants have been thinking about this kind of a game for a very long time.

The Benefits of Linux

It is also interesting how strongly pro-Linux the group is. There have always been strong unix contingents in any net project, yet I'm beginning to see that Linux is much more important to WorldForge than being just an operating system to run on. Unlike Windows, it has all of the development tools that a person needs: gcc, cvs, make, archiving, debugger, good text editors, and robustness. Many of those who were using legacy operating systems at the start of the project have installed Linux in order to participate. I have to smile each time someone declares on the IRC channel, "Okay, I'm installing Linux now so that I can help in the development, now how do I..." In past efforts, division over whether to develop for unix or Windows has been a major obstacle. With Linux available at little cost to every participant, this hasn't been an issue at all.

This is not to say we leave Windows out in the cold. We've got enough Windows using developers that providing Windows binaries is a definite must. And many developers take on the challenge of porting things to alternate platforms - Macintosh and BeOS for instance.

Too Much to Do?

One might think that with all the editors, clients, server-pieces, utilities, game world, and all the rest, that perhaps there's too much work required. In actuality, open source projects seem to do best when there are lots of little tasks that people can focus on. If a single person can focus on a single module, it's more likely that they'll be able to finish it on their own and sidestep the bickering that can occur when lots of people collaborate on an interdependent item.

People also note that there is a lot of "redundant" parallel efforts going on, and wonder if it wouldn't be more time efficient to combine forces on a single product. Actually, we've found the reverse. Having two or three alternate contenders for a given task area brings healthy competition, and since the competitors are actively encouraged to steal from one another, design ideas and algorithms flow between them quickly. Alternate ideas are explored and, if found insufficient, must be discarded in order to "keep up".

Won't It Take Forever Before People Can Use It?

Unlike commercial game systems, WorldForge does not have to wait until it's 100% done before people start playing it. There are so few good games for Linux that any Linux game will draw players. This will hopefully create good positive feedback. Players enjoying the game will serve as ego-boost for the developers, and those players who want more from the game might BECOME developers themselves.

On the other hand, by necessity, open source projects become visible in the public eye a lot earlier in their life cycle than their commercial counterparts. The consequence is the appearance that the project keeps churning its wheels without producing anything but slow and buggy code. Have patience, voice your support and encouragement, and lend a hand in helping if you find an area that needs your talents.

Why WorldForge Will Be a Success

Three reasons why WorldForge will be a success: momentum, a high signal-to-noise ratio, and timing. The quality of developers, the dedication, the support and interest of the gaming community, and infrastructure have combined to give it the mass and velocity to overcome many obstacles. Why the mailing lists, web site, and irc channels are so on-topic and productive is anyone's guess. Spam just magically hasn't been a problem. We've gotten most of our arguing out of the way long ago and have found a common purpose to pursue. Our early anarchical organization has been replaced by a stable hierarchy of coordinators; we can scale up to double the current size of the project without bursting at the seams.

More importantly, we've got working code, tons of art, a good start on music, and are attracting a few new developers each week. The website is full of an overwhelming amount of documentation. And game companies, research institutes, and university lecturers have taken an interest in our game design concepts, and expressed interest in collaboration.

Regarding timing, consider the following:

  • Commercial game companies have only just begun to see Linux as a viable platform on which to make money.
  • MS has been slowly encouraging game developers to use closed, proprietary APIs and libraries like DirectX, which are not officially available on Linux, thus adding some technical hurdles to porting Windows games to Linux.
  • Multiplayer network games are very popular these days
  • Many people have systems on the internet capable of running MUD-style games.
  • UO was very popular, but player sentiment of it is very low. Lots of expectations were ruined. The same is being seen with EQ. There were high hopes for NWN, but given limited availability of it's Toolset (no Linux or BSD Toolset) and only being able to use the D&D ruleset we find this woefully inadequate.
  • The Linux market is growing by leaps and bounds
  • Many Linux users are also die-hard gamers and want good games for their OS. Lack of games is an oft-spoken disadvantage of Linux.
  • Most Linux users are technically savvy (by necessity)
  • Most Linux distributions come with all the tools needed to work on an Open Source project: compilers, editors, revision control, etc.
  • "Open Source" has gotten a lot of attention lately, giving net projects added credibility, and making people realize that there is fame, fortune, and good feelings to be gained from participation.
  • Slashdot, freshmeat, LinuxGames, and other web sites are available, providing free reporting of progress and access to thousands of potential developers.

Taking all of the above into account, it just feels like this is an ideal time for a free, open source game to be developed.