Platonides wrote:
... And if you want WMF to have its employee do X, the pay would be 'I give Y money to WMF if they fix this first'? That seems a bit awkward.
It would be best to follow the pattern that the Red Cross uses, by offering either "where needed most" as the default, or a handfull of alternative options: http://american.redcross.org/site/PageServer?pagename=ntld_main
Jean-Marc van Leerdam wrote:
Well, if you want to keep some control over destinations of the donations you could allow to earmark up to 50% of the donation....
Yes, you could do that (with a footnote or similar disclaimer), and/or associate a certain amount with some of the earmark options after which those would be no longer available for selection (i.e., after they were fully funded.) Earmarking options could be offered in the order they score as maximizing total giving, until the closed-ended items with a maximum budget are fully funded. (Each could have its own goal thermometer shown. After all the closed-ended earmarks are satisfied, only the open-ended projects would remain in the order that donors find them most inspiring.)
Platonides wrote:
The idea of earmarking for minor donations is good, but it should not be readily available ... while not completely hidden, either.
Absolutely; a multivariate linear regression test to determine the extent to which each of the earmark options tends to maximize total contributions should be run in advance, with a sample size (assuming 30 earmark possibilities offered four at a time in a variety of different languages and locales) of between 5000 and 30,000 donations.
The dependent variable would be total amount given, while the independent variables should be the binary flag of whether the option appeared in each test (donation.) Here are some links to statistics to help with multivariate linear regression: http://www.statmethods.net/stats/regression.html
As for the earmarks, in addition to the five I suggested earlier, and the ten approved but un-slotted Google Summer of Code projects, and Sue has a list of 15 open-ended goals which could be used. There is ample opportunity to run more than just 30 earmarking options. I'm sure people could suggest others, either that they think of or find on their favorite mailing lists or village pumps.
Best regards, James Salsman