OK -- I added the "I wouldn't pay option". So far we have 7 votes out of nearly 80,000 user accounts, so I don't think we've warped the statistics yet
Spamgourmet is currently running in the red -- google ads paid for it from the time we got our expensive server and bandwidth contract up until about the time they went public when for some reason the payout fell off (coincidence ?
). Recently cost has been one or two hundred dollars/month more than intake (including donations).
*I think* we can restore the reliability (that we never promised) by refactoring the software to run in-process with sendmail. CPU is our limiting factor now, and we've long known that our code (*my* code, in this respect) was running stupidly innefficiently. We're going to look at doing that soon. When we upgrade, there'll probably be more problems -- sorry in advance -- that's the software business. After that we should be OK for awhile.
If we offered a paid option for the current service, we'd need to provide some features that were inaccesible to free accounts (maybe unlimited forwarding). What would suck is if, like in many of these scenarios, we had to cut back on what was currently offered for free and move that into the paid catetory. Some sg users now are in the extreme category (one address shows that it's forwarded over 11 million messages!!!), so I wouldn't mind putting large but reasonable limits on things like number of addresses and the like. It seems doubtful that most paying users would want the ability to do the things that these folks are doing, though.
In any event, the paying subscribers would be subsidizing the non-paying subscribers, and so would need to get something extra (we already accept donations, and thanks!!! to everyone who's donated -- they don't add up to a whole lot, though). So if you voted that you'd pay for spamgourmet, I'd like to ask you to be specific about you'd be paying for.
My main fear about charging for the service is administration. It's a lot cheaper to be free than it is to be cheap, at least so it seems to me. Merely keeping track of who paid for what and for how long would be a big task -- doable perhaps, but worth considering.
Beyond that, if there's a payment, there probably needs to be some level of guaranteed support. And insurance... And accounting... - you see the quandary here -- merely accepting a payment in exchange for something creates a bunch of cost that we don't currently have, so if we did it wrong, it's possible that doing so would cause the service to, counterintuitively, lose a whole bunch of money (then it would be very hard to justify to our wives!). I guess I'd just say that any movement in this direction needs to be *very* carefully considered.
Syskoll's idea about a separate paid service avoids pretty much all these concerns -- like the rest of you, I have a lot of existing sg addresses, and so I'd hate to give those up (I guess we wouldn't have to?). Does it seem like there's enough demand to support salaried personnel? It'll take more than 7 subscribers