prog: (Volity)
[personal profile] prog
Or rather, go ahead, but I'll just tell you the same thing I tell everyone else, when they say "Hey, you should make Volity for $SOME_NEW_THING" (most recently, this has been Google Wave). "No," says I, "you should make it. It's an open protocol. Knock yourself out."

My attitude towards Volity today is something like borderline hostility. I consider it an aspect of what William Gibson has called "the great clomping foot of nerdism", the kind that is always more interested in taking things apart and exhaustively cataloguing the components than it is in creating wonderful new stuff. Obsessed with categorization and taxonomies, of finding the common root to all things, and then trying to capture that in code, or at least in sprawling wikis. And then, when it's "done", wondering why nobody except for one's fellow robed adepts show any interest at all in it.

It's the video-game equivalent of spending more time writing and trimming an enormous, detailed "world bible" than in creating any stories set in that world. Or of tabletop-game "systems" like Icehouse or Piecepack, which despite their aspirations never sold to anyone other than hardcore game geeks (hi).

If you want to make a video game, go make it. The tools, community and resources to help you do so are all there for you. And yet, if you're a certain kind of geek, the temptation will exist to instead treat your game idea as the top level of a stack: the real prize, you're sure, lay in generalizing all the lower levels, paring and refactoring them into some sort of Ur-Game technology that will solve gaming, somehow, and lead inevitably to lifelong fortune and glory.

My advice is: don't go there, because I know you have great ideas and you're a ninja and everything, but that is folly. Please just make your game instead. I guarantee that you'll be happier with it, and you'll make more fans that way, too. If you're new to making games, the fans might not come, but you'll be so thrilled at what you made - even though it sucks - that you'll do it again, and again, and it will keep getting better. And eventually you'll really be onto something.



I started writing this post with the intent that it'd accompany a release of Webgamut source code to Volity's Sourceforge account. I had a burst of energy to do so earlier today, but it didn't take long to peter out. I am loath to put context-free, commented-but-otherwise-undocumented code out there, because that sounds worse that nothing. I'd instead want to spend a day or so writing some nice farewell documentation for it, first. And I just can't muster the energy right now to re-learn how to get this 18-month-old glop of Perl, Mason and Javascript to run on my laptop.

What do you think? Would it be useful to you or anyone you know were I to just paste a couple of my hard disk's directories into Sourceforge and just put a "Here, you figure it out" README next to them? I don't know, I'm asking. Would the fact that the target for this maneuver would intentionally be obsessed game geeks make it OK?



This post also briefly had a concluding thought along the lines of "I wish someone told me all this six years ago, alas," but that's just dumb, and I apologize for the five minutes of wrong-idea-giving it gave.

I don't regret my work on Volity, nor the work that others have put in, and certainly not any interest that others still have in the project. I think that's great, and I wouldn't even be asking about Sourceforge if y'all didn't exist. I just wanted to put my own current attitude about Volity into words. I'm proud of what we did manage to build, and I am wiser - the real kind, not the cynical kind - for the experience.

Date: 2009-10-09 05:15 pm (UTC)
From: [identity profile] dariusk.livejournal.com
Reminds me of checker/blow's "Make a [fucking] game!" advice. (curse word added for emphasis by me)

http://www.chrishecker.com/New_Year's_Resolutions_for_Game_Industry_Newbies

Date: 2009-10-09 05:16 pm (UTC)
From: [identity profile] dariusk.livejournal.com
Oh, and yes, your post is useful stuff.

Date: 2009-10-09 05:50 pm (UTC)
From: [identity profile] taskboy3000.livejournal.com
If there was ever a post that summed up why I don't start open source projects, it's this one. I love to build, hate to maintain. I get paid to maintain code, but I'll build stuff a day long.

Date: 2009-10-09 11:35 pm (UTC)
From: [identity profile] dougo.livejournal.com
I don't see how releasing code is worse than nothing. The only downside (besides the time it takes to publish) seems to be that people might point and laugh because your code sucks, but you can ignore that. You could GPL it now, and then later write documentation when and if you get the motivation. Or maybe someone else will get to it first.

You're right that people could and should just write their own Volity clients. But if the work you did would save them some time, even if it just serves as an example of what not to do, it seems like a net win to publish.

Date: 2009-10-10 12:21 am (UTC)
From: [identity profile] mmcirvin.livejournal.com
I think that when it comes to writing software, I have an almost complete lack of this architecture-astronaut urge to generalize everything into a maximally extensible framework. I just want to make my little project work. I like fixing bizarre malfunctions, which is something most developers hate.

It's helped me in my career, except that I'm getting older now and into the phase where employers kind of want you to be more of a big-picture guy (even, perhaps especially, if you're not inclined toward management). So I actually have to push myself a little to think this way.

Date: 2009-10-10 12:42 am (UTC)
From: [identity profile] radtea.livejournal.com
Yeah, the biggest problem with working on code generators is that sometimes I find myself using them to write actual applications. It took a long time to get over that, and accept that no matter how much fun the framework is, what matters is the concrete, shippable result.

Date: 2009-10-10 12:53 am (UTC)
From: [identity profile] katre50.livejournal.com
I'll vote for "release now, document later". Sure, the chances are low that in the interim anyone will get hugely interested in writing a web-based volity client, but it could happen. What's to lose?

Date: 2009-10-10 01:17 am (UTC)
From: [identity profile] mr-choronzon.livejournal.com
the experience of having done this thing is valuable regardless of the outcome of the thing itself. you're a far wiser man for it.

Date: 2009-10-11 12:09 am (UTC)
ext_2472: (Default)
From: [identity profile] radiotelescope.livejournal.com
Speaking as one of the other Volity guys (but not the one who banged on WebGamut), I vote for posting the source code. It's better than nothing (not worse). If somebody comes across it and is confused, you might get a slew of annoying questions -- *but* an indication of interest might get your energy level back up too. It's not purely a time sink.

August 2022

S M T W T F S
 123456
78910111213
14151617181920
21222324252627
28 293031   

Most Popular Tags

Style Credit

Expand Cut Tags

No cut tags
Page generated Jul. 31st, 2025 04:45 am
Powered by Dreamwidth Studios