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