[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: failure and determination

Mark 'Kamikaze' Hughes wrote:

Or any bucks. Web advertising sometimes works, but requires initial
investments that are often difficult to get, and uncertain to even break
even. Puzzle Pirates is the only web-distributed game I know of with
even modest success.


What about Spiderweb Software? I don't know if they're entirely web distributed, but they do seem to be financially healthy. They've got a niche: old school Nethack style games with slightly-but-not-much-better graphics.

Just a beach bum conference with no program tracks wouldn't work,
either. No company would allow the employees to go on company time, and
the indies can't afford it without some guaranteed return in the form of
information.


I think there's a demographic that would go for it. What corporations want is of no interest.



Most people have to work for one,

I take issue with "have to." We can certainly agree on "choose to." Also, not everybody's cut out to be an indie, so there is a statistical reality of "are going to."

and convincing them that a
conference is serious work is the only way you get PTO to go, otherwise
you have to use up vacation days.

"Most people" aren't relevant. The question is whether there's enough people in the indie demographic who can see a value proposition in a "party" conference. I think it should be a given that people are paying for the conference themselves, and that the conference has to have value to them in terms of their own hard-earned dollars. Heck, if someone else were footing my bill, I'd have no problem at all in the GDC! That's the whole cynical basis of the conference circuit: they're paid vacations, an employee perk. But nobody gets to openly acknowledge that. Everyone has to pretend they're learning something, even though for the most part they aren't.

I think, for a person spending their own money to be attracted, the price of the conference itself would have to be modest. The location should be good for tourism, i.e. a beach, a forest, an interesting archaeological site, a cave network. Why not an Oregon Caves conference? It should also be easy to get to, i.e. this might inevitably be a local conference idea, rather than a pull-people-across-the-world idea.

Maybe it would work as an Open Source conference. Sort of a mentality that we really shouldn't be paying all this money for all this crap. A certain amount of volunteerism and crowd ownership of the conference might occur.



 I'm of two minds on this.  It'd absolutely be a great way of drawing
people, and with the right design and project management it *should* be
possible to make an Open Source game that's still commercially viable.
There's certainly many discussion topics related to it, which should
make for a lively conference.

But in my experience, Open Source doesn't work at all for games, and
it's misleading to build a conference around it. Even aside from the
commercial aspect, it's one of the applications where the design vision
is so important it requires a hard dictatorship to get anything done,
and OS developers are going to cry "fascist!" and fork the code the
second they encounter resistance.

I'm not sure I understand your objection here. I take it as a given that Open Source is unreliable, that one is personally supplying labor to make up for capital that one doesn't have. I suppose my vision is not an Open Source conference but rather an "Open Source As Applicable To The Indie Game Developer" conference. Presentations should embrace business models and what actually works. Things people have gotten results with. This means I'm not going to even consider holding such a conference until I've proven the $$$$$$$ validity of my own open source toolchains. *If* I'm still using such toolchains. :-)

Plus, you risk inviting the lunatic
fringe Free Software dweebs into the conversation, and nobody wants
that.


True, that's a risk. But one could stack the deck with MIT/BSD speakers and technologies. Invite the corporate open source types, i.e. Eclipse.

like the only thing holding me back are my lousy tools. Some would say that's the whole rub. Maybe most would. But most never ship anything of consequence, so maybe it's all a matter of how "high road" you want to be about anything.


But that's the problem: you're working on your tools, but publishing
nothing. Is "nothing" all that good?

What's the comparo? Have you gotten independently wealthy from pursuing an alternate strategy yet? I think persistence in the face of adversity is probably worthwhile in the long haul. I have yet to find or devise the programming technology that "fits my brain." I think I'm getting closer though.



No wealth, but I've got a lot of pleasure from people enjoying my
games. Money is insignificant to me, it's just a tool I use to get free
time to work on games, and if I had enough I could buy art and music
resources to make prettier games that even more people would enjoy. On
the scale that I measure things, I'd say I'm modestly successful.


Yeah, but that's a personal scale, not how the world keeps score. My personal scales are "aesthetic satisfaction with what I'm doing" and "doing things on my own terms, according to my own decisions." Compromosing on technical aesthetics in order to "get things done,"past a certain point, makes me deeply unhappy, So, I really am happier with "nothing" to show for my efforts than with having something half-assed. In terms of personal scorekeeping, we're each doing fine. We just have different scorecards.

The world only respects money. That's the only external, objective scorecard that matters. All other scorecards are internal, so they can be whatever you like. You just have to be honest with yourself about how happy or miserable you are about what you're doing, and why. I figured out about 2 years ago that I'm not actually a bad programmer. I just hate the vast majority of industrial coding that other people try to ask me to do. I'm definitely a "bad industrial programmer." I might even be unemployable in that capacity, but I'm not quite certain of that yet. I definitely want industrial technologies to solve my own problems, not just those of other people, and I haven't come up with that two-birds-one-stone solution yet. I look for every excuse I can to avoid "bending over" for other people's coding problems. I'm happy to say that signature gathering just got more lucrative, and I'm probably off the hook for another 6 weeks yet.

 Last month, I had 1173 downloads of various computer games, and--hard
to count exactly, since these are visits to various web pages instead of
zip files--maybe 2000 tabletop games.  A lot of tabletop RPG publishers
would kill their grandmothers to have my numbers, even if they are free
(and that situation's going to change very soon).

I haven't pushed GameScroll at anyone but Linux users yet, because I
want to get the graphical editor done before Mac and Windoze users see
it; Linux users are smart enough to edit XML files by hand. If I don't
get another 2000 downloads a month from that with minimal free
advertising, I'll be shocked. At its new release peak, Hephaestus was
getting 3000 downloads a month, and the shareware fees paid for a year
of lattes and a new computer.


Hopefully you are poised for bigger $$$$$$ things.

If something sucks beyond your willingness to bend, I think you need to be honest with yourself about it.



That's true. Have you looked at doing console development? The
business environment sucks beyond belief, but they need C/ASM guys
desperately. You won't get to do game design for them, but at least
it's your kind of coding work.


Consoles are a difficult thing to "look at" though. The getups are all proprietary, require NDAs, big wads of cash for dev environments, etc. The historical architectures have also been nothing to get excited about. Recall, I cut teeth on the DEC Alpha in 1996. I quit my job in 1998 and I'm still waiting for the rest of the industry to catch up to what we were doing in 64-bit computing back then. Now, I'll admit, if I could find an entry level job doing ASM code for the IBM Cell, I'd probably be happy. But they'd have to pay me well, and not work me to death. Also they'd have to let me learn the skills on the job. There's no way I'm going to fork over time and money for other people's technology wonderdogs anymore. When I take on those burdens myself for no pay, they have to precisely solve my problems. And right now, nitty gritty console code doesn't solve any of my problems.

Actually what I probably want, is for IBM to ship some workstations based on Cell, and go back to doing high end CPU optimization. But it's early days for the Cell yet. Also I have considerations like, not wanting to leave Seattle. All these things are only a day job, not my career.

reuse. By the time I understand the design requirements, I may end up rewriting all the code 6 times. That's what happened to me last time around, at any rate. Takes me about 6 tries to get something right.



Well, as I said a bit ago, it's been my experience that evaluating the
merits of systems that don't exist is *really* *hard*.

It's not that hard. Much of strategic feasibility can be designed on paper. Quite often as one pursues the pencil-and-paper part of the design process, one realizes that one's approaches aren't as great or workable as one first thought. For instance, stories about Ocean Mars or The Game Of Mallor. Can't seem to get the stories out, and words can certainly implement prototypical stories, so maybe these concepts aren't working so well? Maybe I don't have a grand enough story to keep my attention?

Another example: AI architectures. Just because you're working on a 2D array of squares, doesn't mean you can just brute force whip through that 2D array with bitmap operations to implement path search problems and the like. I liked the sound of the "bitmap ops" paradigm, and spent a lot of time figuring out how it would work, but it doesn't really work. Another example: scripting AI operations. If you're trying to implement a set of high-level AI path operators, you should be able to pseudocode the kinds of semantics you're interested in. Then design either a library or a Domain Specific Language around them. I haven't adequately chugged through that one yet. I know what kinds of AI problems I need to solve, but not yet a better way to linguistically express them. All that's really stopping me is financial pressure, spending all my time building broken open source tools, etc.

There's always seemingly 10 things I need to do at any given time. Progress gets made, but there's never really a critical path. I think I last put down the AI problems because I was sick of paperizing about them. I wanted to implement, as I had enough to start implementing. But this was decidedly not easy to do, as I still don't have a production Bigloo Scheme programming environment. I don't have an OpenGL binding; creating one is a much more tedious task than banging out my AI code. I almost started coding stuff in C just to avoid these problems, but upon further paper design I realized how quickly I'd run out of flexibility in C. The algorithms might demand a more Schemeish approach. And so it goes. I started worrying about tools and build environments about 2 weeks ago and I'm still worrying about them. At least I'm no longer "burnt out" on AI theory problems though!

Anyways, these sorts of issues are why I say that good tools are essential to the progress of IF. When you're always being distracted by really basic issues, like getting your tools to work, you get pulled away from delivering any content.

Throwing
something over the wall and seeing what people say is often the best way
to learn whether it's good or bad and what you need to work on


"Tossing over the wall" often means posting a fairly cooked idea to something like gamedesign-l and seeing what other people think of it. Certainly this works for High Concept stuff. I can't really do this for technical stuff though. Nobody really wants to talk about detailed technical designs, least of all me. For technical stuff, I just keep bugging people, "does this basic tool work?" Sure are a helluva lot of yes/no answers I feel a need to collect up, since the answers in open source land are usually "no!"


Cheers,                     www.indiegamedesign.com
Brandon Van Every           Seattle, WA

"The pioneer is the one with the arrows in his back."
                         - anonymous entrepreneur