In the mechanism repo we now have files describing past, present, and possible crowdmatching mechanisms.
Perhaps a README would help that could specify the requirements for a complete mechanism definition.
I’m also wondering the best process for discussing and evolving the defined mechanisms. There could also be perhaps a hierarchy or naming scheme to identify sets of related mechanisms and highlight only their differences. But maybe it would be better to have a master template of sorts and the mechanisms are defined simply by the ways they fill-in or vary from the main default?
I was reading the patron-only proposal from @mray and it seems too undefined to me. Since there’s no definition of “pledge goal”, it could be completely anything. To be absurd, a patron could set any measurable goal at all (e.g. the number of government offices that are registered users of the project). I’m assuming the intent is to have the goal based in some metric that is itself internal to the mechanism… right? Maybe it could be specified that the goal is in an internal unit such as number of patrons or sum of pledges. Also, I suppose the project could choose to restrict pledges to a particular unit, right?
I would like to see added a set of unarguable observations for each proposal. In other words, things that are inherently true because of the definition and thus don’t need to be stated but might help people simply recognize facts about the mechanisms. I suppose this gets into the questions on what we next do with these definitions once they are all clear and consistent.
Also, I suggest we add other mechanisms (even non-crowdmatching ones) for comparison, using the same format. I would do this only after we’ve optimized the format for these definitions. But then, we could mention bounties, one-off-threshold-campaigns, non-matched donations… could also define FundOSS’s quadratic funding in our format (that’s the closest to crowdmatching, arguably a form of it) and preshold.
Having an objective set of definitions that can be compared one to one easily is a huge value.