Thought: delayed release

10 respostas [Última entrada]
Joined: 05/30/2012

I'm just going to throw this out there, not expecting any sort of answer, but just because I want to share the thought. It was brought on by a post on Reddit about funding libre software development.

My idea is a variation on crowdfunding. It goes like this: when a program is ready for use, a release date quite a long time away (say, a couple years) is announced; the program is to be released by that date. But then donations are asked for, and a target goal announced. As money is donated, the wait decays exponentially. If the goal is reached, the program is released immediately.

I thought of this idea as a way to get rid of the uncertainty associated with normal crowdfunding. If you donate, and the target isn't reached, it doesn't mean the donation is for nothing, because instead of having to wait, say, two years, you may only have to wait two months. I think this could be particularly useful for games.

Joined: 08/11/2014

That was the plan for LibertyBSD, wasn't it?
I guess it'd work, but not without some exposure or advertising.
Perhaps it'd also work in increments. Say, the program is released, but extra features are under delayed release?

Joined: 05/30/2012

I'm pretty sure LibertyBSD wasn't going to be released at all until the goal was met, right? Anyway, I don't think this scheme makes any sense for an operating system. This is an idea for programs that can at some point be considered "done", and not need much if any further maintenance.

Joined: 05/18/2013

It seems like a good idea to me because of the greater incentive to donate.

Michał Masłowski

I am a member!

I am a translator!

Joined: 05/15/2010

LibertyBSD has shown two problems with what you propose: we need to
trust the author first that they are able to write and release the
program, and this method of funding prevents any cooperation with other
people on that software.

Not all crowdfunding campaigns use threshold systems (which have the
uncertainty that you mention), I think there was a well-known one for a
free software project that didn't.

Joined: 05/30/2012

All crowdfunding campaigns necessarily have uncertainty; it's just a matter of where the uncertainty is.

I don't think you can raise money for a video game unless you've already got it about halfway done. So with crowdfunding, there's a dilemma: either you can require a certain goal to be reached, and either abandon it entirely (wasting the effort you've already taken) or release the half-finished game anyway if the goal is not met; or you can not require a certain goal, at which point what stuff promised in the crowdfunding campaign gets cut if the goal is not reached can be arbitrary. Either it's a gamble for the developer of total failure, or it's a gamble for the funders of not getting quite what they wanted.

My idea is to instead change the risk to what proprietary software developers risk: the developer not getting as much money as hoped. This is better than the risk of not getting anything at all, and at the same time avoids any risk to funders.

Joined: 04/22/2013

But a developer will only invest time if he knows that he will get payed.
That's the advantage of ordinary crowdfunding.
Your proposal is more risky... if he doesn't get any money, ok, the release will be in a distant future, but so what?
He did the work unpayed...
it's hard to calculate that way or make a living of it.

But i guess your approach is very useful for not-so-professional development.
Like, if somebody wants to release the game for free (and as free software) anyway, but with this try he may take the chance to get some bucks.

Joined: 05/30/2012

I don't think you can crowdfund the development of a game without being about halfway done, anyway. An idea for a game is not enough to tell whether or not donating to it is worth it; good execution needs to be proven.

I don't think risk-taking is necessarily a problem. The main business today of proprietary video game development is developing a game in a gamble that people will buy copies of it.

Joined: 04/22/2013

"I don't think you can crowdfund the development of a game without being about halfway done, anyway."

Well, unless you have a very big name and a strong fan base, i actually agree.

I think you really have a point when it comes to video games.
However, it was always comparatively easy to make money out of free software video games, since developers could just keep the non-functional data non-free, and yet the sector is so heavily dominated by proprietary games.
So maybe it's not that they don't have an option to make money out of libre games...they just don't want to because people buy non-free games anyway? Maybe it's like that...

Joined: 08/13/2014


I understand that the idea is to develop software and release it sooner and sooner as more money is raised.

If the parameter to the exponentially decaying function is the money raised so far, and the meaning of the function is the time from the fund raising campaign opening to the release date, then I see a problem with using that type of function: diminishing returns. If the time to release is D=exp(-M/G) where M is the money raised so far and G is a scale constant (It can be thought of as the “goal” parameter, though there's no well defined goal), the inverse function is log(D)=-M/G (Note the logarithm).

As more money gets raised, contributing any fixed amount reduces the time by a ever-decreasing amount. For instance, if the money to halve the release time is 100 000 USD (This is given by G≈69 315 USD), then when the remaining time is 1 year, donating 100 000 USD decreases it by 1 year[1]; when the remaining time is 4 weeks, also donating 100 000 USD decreases it by 2 weeks only. In addition, since the release date is bounded (We know it will get released at or before the release date for 0 money raised) in my view this promotes raising very *little* money.

I have no experience with crowdfounding. If everything gets eventually released, then people may not contribute. I think that the best model would be to set partial monetary goals in advance, with associated features that get developed when those partial goals are reached. I don't know whether it's better to release as they're finished or wait until the fund raising campaign to release the software with all the features whose partial goals were met.

Are there examples of free software that was developed (Rather than released as free) through crowdfounding?.


[1]: Ignoring the difference in length of years, which is small.

Changelog: Added missing word.

Joined: 05/09/2014

I really like this approach: