1   It is abundantly, unarguably clear that programmers can work together to produce programs of the highest quality in collaborations that in some of the largest cases number in the thousands of programmers. It's not always the best choice and there are strengths and weaknesses, but it can and is done.

Such evidence is lacking in almost (but not quite) every other domain, and in particular the arts. We basically have the technology to create a distributed/open source anime production; all the music technology is easily available, all the voice recording technology is easily available, all the animation technology is easily available; the missing pieces are small and would quickly emerge if there was demand. Yet no such beast has been made.

At some point, and we are well past that point, you have to stop arguing about what could be and look at what is. It is clear the arts aren't going to have the same open source revolution that software did. Programmer experiences don't apply to the arts, or it would be obvious by now that they do; we'd be discussing it in past tense, not nebulous future tense.

I think anything resembling open source animation is going to be confined to things that two or three people can produce. (And all putative exceptions that leap to mind are either clearly not of commercial quality, or are indeed produced by a very small team that isn't going to compete with commercial quality and output rates anytime soon. e.g., I can think of some awesome animations that some individuals have made, but which clearly took many, many months or years for a handful of minutes.)

2 This guy is an excellent example of that. His flash animations are of a quality comparable to any commercial product -- but he's using grants to finance a lot of it, and his output is measured in minutes per year, if not even less than that.

3

Actually, it has happened in some of the arts. Writing, for instance. Isn't the blogging revolution effectively "open source writing"?

And like software development, in some cases pros are being paid to blog, because it's commercially valuable to their employers.

There is a degree of that in the visual arts, too. That's what drives sites like "Deviant Art" and Hentai-Foundry. (Not to mention 4-chan.)

Doujinsoft is a kind of hybrid.

It even happens in animation, with people creating flash animations and distributing them for free. And various kinds of video production; YouTube is used for a lot besides pirating segments of TV shows.

But it hasn't happened for this kind of animation, and you're right that we'd be seeing at least a nascent form of it by now if there were any chance of it becoming big eventually.

I have seen exactly one case of it in classic animation. It was called "Gruesomestein's Monsters" and it was really pretty good. But the money ran out and they didn't do any more of it.

As you say, irrespective of how desirable it might be, it isn't going to happen. Not any time soon, anyway.

4 Thaaank you. Well put. Exactly what I was getting at. My fault for conflating it with the "companies can't enforce copyright and I'm tired of caring" line.

5

On his blog, Author makes the following comment about me:

This topic seems to degenerate quickly, and I’m a part of the problem. So I’m not going to respond to Steven’s accusation that “Red Hat has never made a schedule and stuck with it”. It seems like my tangential point about ignorance was more true than I realized! Now if you excuse me, I have some bugs to fix, because the schedule for RHEL 5 U3 includes them (the same schedule which we never made).

And he still misses the point: he is one of those people covered by my "UNLESS" clause. He's being paid to work on the stuff, in order to make deadline.

Maybe he also does some of it volunteer, but a project developed primarily or exclusively by volunteers doesn't make schedules for major feature releases. They ship major features whenever they're done.

6 I just wish it was easier to pay for the stuff I like.

I'm buying the Haruhi DVDs as they came out.  I bought the first few Bleach DVDs, because I liked the first dozen or so episodes of Bleach.  I have Midori no Hibi; I have Ikkitousen (which is honestly pretty bad); I have two copies of Haibane Renmei  (a mistake, but not one that greatly upsets me.)

But most of the time I neither want nor watch the DVDs.  There may be some nice extras (I bought the Cutey Honey DVD set just to get clean OPs and EDs), but I don't often listen to dubs, and DVD subtitling is terrible.  So the primary reason I buy them is because I downloaded and enjoyed the fansubs, and I want to send some money back to the studio that created the show in the first place.

But they probably get about $3 out of the $30 I pay for a DVD, so just about anything else would be more efficient.

7 While I don't disagree at all with the general thrust of this discussion, I do want to point out an existence proof of what one person can do pretty much all by himself, Shinkai Makoto's Hoshi no Koe AKA Voices of a Distant Star.

Back in 2001-2 he spent 7 months of quality time with 2 or more Macintoshes and produced 30 minutes of absolutely stunning detailed animation. He and his SO also did an alternate soundtrack (there's only two main character voices plus one "extra" who gets a few lines).

I think he's still considered to be about the best at doing beautiful cloud filled skies.

However, this doesn't help the business model problem, I think all his works but the 5 minute short She and Her Cat were paid for in one way or another, that short plus I assume his other paid graphic design work for a video game company got him a grant for Hoshi no Koe. Plus he's exceptionally talented, Wikipeida says "Shinkai has been called the new Miyazaki in several reviews including Anime Advocates and ActiveAnime, though he disagrees with this comparison, stating that 'it is an overestimation'."

- Harold

8 Hmmmm.

Let me refine the argument, then, to just the collaboration point. There's little writing collaboration, either with multi-author works or a full "writer-editor-publisher" sequences. (I don't count multi-author weblogs, as they could easily be many separate weblogs, usually.)

All your other examples pale in comparison to even niche programming collaborations, though; Django, a Python web framework that I use for my site, has tens of contributors, easily, and at least ten major contributors, and that's nothing compared to the kernel. Clearly, programming is a different world, for reasons so obvious that I wouldn't load this comment down with them.

9

A huge problem with distributed anime production would be thematic/quality control. There needs to be a Director, and that person's going to need to be in a million places at once.

One thing the Japanese do right that I think the North American localizers should take a more serious look at is getting all the VAs in a room for a recording session. Dubs are terrible because it's obvious that the VAs were standing in a both reading sterilized lines without feeding of the lines of the other characters. I see this only getting worse if people are phoning in performances from hither-and-yon without a Director looking over their shoulder.

The same thing happens with a lot of video game mods. Somebody gets a great idea, finds a few friends to help out, then it all derails when it becomes apparent the guy in charge can't be everywhere at once (or isn't cut out for a leadership role).

And I'll shut up now because this is getting far afield of the original copyright topic.

10 "One thing the Japanese do right that I think the North American localizers should take a more serious look at is getting all the VAs in a room for a recording session."

I've always thought the R1 failure to do that was based on cost: the VAs are part-timers, often aren't even in the same city, and there's not enough money in the R1 market to pay them a full-time wage.

Of course, if more folks bought their anime, instead of insisting on having it as a free right, maybe they could pay that much.  There's a reason I do "obligation buys."  Even if I don't buy every single DVD of something I've seen on fansub, studios need to be rewarded for producing the good stuff, and companies need to make a profit importing it.  Otherwise, like Steven says, it won't be produced, and it won't be imported. 

As for the rest of the OS flap, sorry, I don't do religious arguments.

What's funny is, I bet the majority of the people ranting about their right to free fansubs are hypocrites underneath, should you scratch that top layer of selfishness.  (Oh, wait, it's not hypocrisy if you think of it that way.)

What I'm saying is, if any of them ever produced and sold an actual original work, making a significant amount of money on it (say 1/3 or more of their current yearly income), they'd suddenly become upset if someone translated their work and gave it away for free.   I know  I would.

11

Doing it that way is largely about cost, but being part-timers, they have other obligations to meet that probably makes scheduling a huge nightmare as well.

But while R1 companies are experimenting with new modes of distribution (streaming and LQ downloads for a few bucks) to get product out faster, I'd like to see a bit of experimentation on the production side to improve quality as well. If I know the translation for volumes 3 through 5 of Nekomimi Meido Panty-Fighter Extreme are scheduled to be ready for dubbing at the end of March, I tell my VAs "Be in Houston for a weekend recording session the first weekend in April." Being an outsider, I don't know if a weekend would be enough to dub that much material (figure 9-12 episodes), but I think that with everyone in a room, the recording would go smoother, with fewer takes, and better quality.

But really, all the above was just blather to add respectability to a comment noting that ubu's IP hash is "Furries."

12 I remember a Usenix conference years ago where Lessig gave a big speech about the "napsterization" of content (that's the original Napster model, of course), using player-piano rolls as an early example of the market forcing change on those greedy copyright-holding bullies. He was preaching to the choir, and they cheered him.

I didn't. Because I was spending at least four hours a week dealing with people who were downloading photographs from my web site, printing them out, and selling them on ebay as "real prints from the negative". One enterprising clown was making keychains and coffee mugs. One of these New Media Entrepreneurs even told me that he owned the pictures because he'd bought them as a royalty-free CD at a show. And I wouldn't be surprised if he did.

Their "napsterization of my content" wasn't taking money out of my pocket, but it was hurting the professional models who were trying to sell other pictures to fans, and finding my pictures on ebay led some of them to believe that I was the seller. And I didn't have the right to sell their pictures, because I hadn't paid them for it. Free web gallery linked to their official sites, yes; that was agreed to up front. But without a model release, I couldn't legally or ethically have sold prints or coffee mugs, so Lessig's argument that someone else should be able to fell flat for me.

To be honest, I don't even look at price tags on R1 DVDs; I was already buying most of it from Robert and Amazon before Suncoast went under, and even the worst episode-per-disc series are cheaper than going to a movie. I'd like for Japanese media companies to someday grasp the concept of discounting older material; perhaps if they had more domestic DVD sales, they'd be healthier. As it is, their high prices help the pirates feel good about themselves.

Case in point: one of the things I bought in Akihabara was the DVD box set of "Sentou no Musume?!", a live-action series. The ten-disc box is over $300 on Amazon Japan; I bought it used at Traders for under $100, and it's pristine. Each DVD case was opened exactly once, to rip the disc, and then carefully replaced, for maximum resale value. Most of their merchandise was like that.

-j

13

Harold, Avatar talked about that:

(Yes, yes, Makoto Shinkai. He's the iron man! But it took him two years to produce ONE episode, and the moment he could set up a studio and get other people working with him, he did. Should tell you something about the feasibility of doing it solo...)

Yeah, if we can find iron men like him to create animation for free, and if we're willing to wait 9 months between half hour episodes, then we're golden.

But he's the only one. If this were really going to become any kind of factor, others like him would have sprung up by now.

14

I was going to mention  Shinkai Makoto but Harold beat me too it.

Here in the UK we get anime DVD's 6 months+ after the US, and they're twice the cost (£20 a disk in the high street stores) Therefore I buy R1 mainly or R0.

I've only really discovered fansubs in the last few months, and have come up with the following policy: If a show is already out, I'll watch the first episode ONLY. If I like it I'll buy the rest of the series. Pretty much every (non anime) DVD I have I've seen before buying, either on TV or the cinema. I often watch a new sitcom (for example), buy the first series when it comes out and will the tape subsequent series. I'd imagine japanese fans do this too.

For series that aren't licensed/released (Bandai- hurry up and release Lucky Star!), I'll watch them until they are available and will the buy the disks.

Andy

15

Forgot to post this link.

Andy

16 'But really, all the above was just blather to add respectability to a comment noting that ubu's IP hash is "Furries."'

PIXY!!!!!!!  I need to have a word or two with your hash routine!

"NICE BOAT" would do, in fact....

In an attempt to add respectability to this death threat to Pixy's code, I'll note that it would be really hard to blog about all 46 open source versions of "Shana II" even if it would be emotionally rewarding to take a hatchet to all the angst.

17 Some music kinda allows for an open source approach. After all, an entire sub-genre has popped up around the "Amen Break" (a 5 second drum solo from a 60s soul song), and you can go out and buy various componentized sounds/beats/whatever you call them, and mix and mash them together to get something new. This isn't possible for all music, of course, but it definitely dominates in electronic-based stuff.

Shamus mentions that he can't just take the components of Cowboy Bebop and reuse them to make something new, and he's correct. But I could see tools being developed at some point in the future that would allow a componentized approach to animation. For example, if the creators of Cowboy Bebop didn't just release their raw images, but instead gave tools that would allow you to move Spike or Jet in any way you wanted, and then also provided backdrops, etc... you could probably create something pretty decent. Audio would be a challenge, I guess, but it would certainly be possible.

But that's all just speculation. Today, it's not really possible. I'm not that familiar with the whole fansub debate, but it seems to me that the whole reason they exist is that the shows aren't available in R1 yet. The "free" part is just a sorta added bonus (if you're not squeemish about IP issues). From what I know of the issue, I don't see any real way to fix the problems, which I suppose is why it's such a contentious issue...

18 I find it amusing when fansubbers get upset because another group, or even worse, a licensee, has made use of their translation. "How dare they swipe our hard work!".

Translation is one of the few places where the "open source" development model can usefully be applied to this market, and it's being done for things like the Haruhi novels. You still need a solid core group for QA and release management, but there's room for volunteers to contribute in their free time. If they had permission from the publisher, it'd even be a good idea.

If some good genre writers started releasing their stories under Creative Commons licenses that allowed derivative works, you might eventually end up with a pool of quality source material for making free manga and anime, but there's still the problem of funding the artwork and production. Much of the fansub community is funded by parents who think they're paying for college educations; how much would be produced if they all had real jobs?

I think the real killer for the concept of open-sourcing anime is that very few people are willing to do the hard, thankless parts of the job for free. This also applies to open-source software, where documentation and QA are often minimal or missing, and it's not at all unusual for old bugs to reappear in new releases, because there aren't that many people who really have a passion for those jobs. The people who can do it don't love it enough to do it for free.

New features are sexy; tracking down race conditions is not. A year and a half ago, I had a lengthy back-and-forth with Author over a long-standing Linux bug that was biting me several times a month, and when we finally nailed down the cause, he tried to make a quick fix, and then realized that it would take a lot more work and testing. It's unlikely to be fixed until someone needs it enough to pay for it, or someone else develops a passion for this particular area and rewrites it from scratch, introducing a whole new class of bugs.

-j

19

Shamus mentions that he can't just take the components of Cowboy Bebop and reuse them to make something new, and he's correct.

AMVs try to do exactly that.  The best of them might actually succeed once in a while.  Which isn't the point of this thread, but I couldn't let that comment go by.