To address the last line in the story - "As one would expect, many backers are questioning exactly what all of that half a million dollars was spent on - something Stephenson hasn't addressed publicly in the two years since the project was funded."

Not hard to imagine. $500k less taxes is pretty much enough to pay for two years of two full time developers and a poorly paid 3d artist.

Yeah, I feel like there are members of the public who think a $20,000 dollar goal should pay for 3 or 4 people to work on a thing for a while... IT DOESN'T! It might pay 1 person to work on something for a few months.

This is why kickstarters should already have substantial progress made on a project before asking for crowdfunding, but if people want to bet on the Longshots with nothing to show but an idea, they should expect to get burnt from time to time.

This is a dangerously low-quality comment.

For some people knowing that you disappointed people ("a fan" would be right word in case of Stephenson) that believed in you so much could be easily worth of committing seppuku or something.

For all of the threads recently discussing failure and founder depression, you flippantly suggest someone should kill themselves for "disappointing some fans". Have you ever failed as ambitiously and publicly as Neal Stephenson did here? Maybe you read the story and thought he did not sound as apologetic as you'd like, but having read the rest of the backers' notes I'm pretty sure that he and the rest of the team have put their lives into this for years.

It's kind of like wondering how 100 people managed to eat a whole extra large pizza. Like, the pizza might seem quite large to you personally, but it's really very little food compared to what 100 people would normally eat even as a snack.

One shouldn't have to imagine - there should be a basic accounting in cases that fail and raise more than some fairly low threshold of cash. I don't mean like an IRS audit level of detail, just a 1 or 2 page breakdown and maybe a notarized declaration.

Can't say I'm very surprised, the whole project reeked of nerd hubris.

So now the whole game is gone? The work that has been done will be flushed down the drain?

Seems like it would be relevant to release the whole shebang as open source with potential commercial license incentives so that anyone truly interested in the community can work on it, and another studio can even pick it up where it was left off while benefiting the original devs as well.

It would be great if as a standard part of a Kickstarter pitch, people would specify what they will do if they fail to deliver. I assume promising to open source the project would be the most common choice, because that's what I think most backers would want.

If they don't deliver the rewards, they should return the backers money (I think Kickstarter says this, but I haven't carefully checked).

This quickly leads to not promising tenuous rewards, but hey, that's fine.

Most failed Kickstarters fail because they run out of money. There is no money to return.

I disagree, when I've backed KickStarters that have failed I've never wanted my money back. It's a risk, it's money I'm throwing away to help a project exist, and I'm aware it might not happen. To me, that's just fine.

I don't think people should avoid high risk projects, I just think they should avoid promising to deliver them. Like I said in an earlier comment, make the reward some novelty trinket, not the complicated, never-been-done video game.

> they should return the backers money

Yes, though it generally comes up only if there is no money left to return. I think there should be a statement about what will happen if the money runs out. Note that CLANG is a project by Subutai Corporation and thus that company (not Neal) owes it's backers. Someone as rich as Neal Stephenson could promise to pay out of pocket if the company goes bankrupt, but that is not practical for many.

The money is already spent. In the article it states they returned about $700 to the dozen-or-so people who requested it. But the rest of that money, in addition to a lot of personal runway, is gone.

The comment I replied to is talking about Kickstarter projects in the abstract, not just CLANG.

In the case of Clang, he did actually fulfill all the rewards

He didn't. The prototype doesn't fulfill his promises made in the Kickstarter video.

http://www.youtube.com/watch?v=nV39GiK4Hw0

https://www.youtube.com/watch?v=XrYVnBe3XjE

https://www.kickstarter.com/projects/260688528/clang

I backed CLANG and as far as I'm conceded they did fulfill the requirements. Am I disappointed it didn't result in something better, yes. However, I was prepared for that, I saw my money as funding research in how to make a sword based video game.

How do you return money that has already been spent?

The easy way is to not promise any rewards that depend on completing a complicated project. Like, send people a novelty pack of gum and also figure out how to get them a copy of the game if it happens to be completed, just don't promise them the game.

Realistically, Kickstarter is a store and people won't pay if they aren't getting the result of the project.

Well, this discussion is about people who paid and aren't getting the promised result.

I think Kickstarter is neat, but I think people should probably approach the claims of the typical project with quite a bit of cynicism. That means (as a backer) not treating it anything like a store.

They did. The new ToS softens the language somewhat. Which IMO makes sense because if I had enough money to refund backers, then why did I need backers? Maybe there are some marketing benefits but they may or may not be worth Kickstarter's cut.

I think you're optimistic about the benefits of just dumping a bunch of code onto a web site and saying: Go for it! (Even assuming there are no commercial libraries and the like to strip out.) I'm not saying it would never work but presumably things are quite far from completion or they wouldn't have been shut down.

Or, you know, keep the rights to your work and try again later.

You can do both.

This was doomed from the start. By trying to make it more realistic they exposed the limits of technology.

A key problem is that the weight of a blade is a huge part of swordfighting. A light bladeless controller can't capture the the experience. Especially since there's no feedback from hitting another blade.

On the other hand they might be able to recoup their investment.

In the leadup to Star Wars 7 there's a huge potential market for a lightsaber duel game. There's still enough time to develop a tie in game, and I can see the tech being bought for over 500k.

It seems like a risky move to announce that, oh btw, we gave $700 in refunds to some random people from the forum. Now the comment thread is full of backers saying "if THEY got a refund, then I want one as well!". That could get quite ugly.

That statement was on the level of: look we tried to return money to those who asked. Of course most people don't know that the project failed.

How will they deal with those? Ignore, Abort, Retry...

According to his post, he apparently wrote a book about the whole thing. https://www.kickstarter.com/projects/260688528/clang/posts/9... If he's not going to open source the project (my vote) maybe he could give his backers an epub!

One thing that bothered me about "Clang" from the start is that I was never really convinced that a Kinect-like system couldn't do it better. Per Clang's FAQ, their position was that the Kinect had too much latency. The first generation Kinect had a latency of 90 ms and the second generation had a latency of 60 ms. A bad, but not atypical display has a lag of 3-4 frames, so the newer Kinect roughly doubles the lag (or more) of a typical gaming system. Not so good! However, for how long will that remain the case?

Arguably, a better approach would be to work on a low-latency form of Kinect. First, this would provide richer input that includes the players body position in addition to his blade, which could make hit detection more realistic. Second, a Kinect system with lower lag would have more general utility than just playing sword games. e.g. Unarmed combat would be an option too. Additionally, authoring sword combat software for Kinect-like systems would allow you to release something imperfect on existing hardware. It's great to shoot for the stars, but picking off some low-hanging fruit first can help you get there!

Clang controllers used a few methods to give users feedback about hits. Obviously, there was no way to make the controller stop in mid air as though you'd hit a concrete pillar with your sword. However, it could use rumble packs in concert with audio cues to simulate glancing contact with an opponents sword or light cuts to his body. There's no reason a Kinect-compatible stick couldn't include the same sort of feedback.

So what's the next realistic step for advancing the state of video game swordplay? If Clang is open-sourced, it might not be an unrealistic task for somebody to hack up an imperfect Kinect version for people to try out. If Clang "wasn't fun" with it's own controller, it's unlikely a Kinect version that's laggier and lacks haptic feedback will be fun either, but at least it will give people a chance to try it out.

Personally, I think VR displays are the key missing ingredient for swordplay games. Seeing your stick in front of you and a sword mirroring its position on the screen is one thing, but seeing just the video game sword as though it were in your own hands is quite another! VR systems are going to have to work out low lag displays and body position data anyways but, once they do, it will pay well to be positioned to take advantage!

I agree this could be done with a vision system, but I think you're underestimating the challenge of it. You're proposing they build an experimental < 50ms vision based tracker which would need a high FPS camera and a special coupled processor with novel algorithms and PC drivers which were tuned to rapidly ingest that processed signal. Interfacing with a 6-axis sensor over bluetooth seems way more sane for a project of their scale.

I disagree with your second point. The slight fuzziness of kinect and the additional 50ms-80ms of lag is absolutely enough to ruin the game and thus it's not a good test of the other mechanics. To see why, watch this short and enlightening video on why most motion games are not fun: http://www.penny-arcade.com/patv/episode/kinect-disconnect

If anything, they should have prototyped it with PS3 Move or Wii+

Kickstarter just updated their terms so it is possible this triggered that: https://www.kickstarter.com/terms-of-use#section4

"The creator is solely responsible for fulfilling the promises made in their project. If they’re unable to satisfy the terms of this agreement, they may be subject to legal action by backers."

However the CLANG project did make a prototype but it would be nice if projects fail they have to be open sourced or able to be observed or carried forward if possible.

They gave their game as part of the rewards. I wouldn't say that a prototype fulfills this promise.

Perhaps Kickstarter shouldn't allow to give the product people back with money as reward because if the project fails this product reward goes down the drain too. I know like 90% of all projects do that and its kind of what makes Kickstarter but still its not really logical. Or another option would be a rule that backers have no rights at all. This would force the project makers to publish budgets and more details about the project otherwise nobody would back anything.

Neal, please release what you have made. We can do something with it, I promise. Go build something bigger than this.

> make sure I have something to show [...] even if it's just one level,with bad AI and bad controls.

Kickstarter funders did in fact get a downloadable demo/prototype.

https://www.youtube.com/watch?v=nV39GiK4Hw0

Browsing through the backer comments, I find that about half the reactions to that demo are similar to this one:

> I too am very disappointed with the bare bones tech demo you created. I had high hopes for CLANG - I bought the Razer Hydra expressly for it and then realised how pathetic the prototype really was.

> I don't believe you properly fulfilled the reward pledges at all and it sounds like our backer money was simply wasted. I would love a refund, however failing that I would like to see code open sourced and released to the public.