If I have Zope running, using ZODB3 on Data.fs, can a separate process also use ZODB3 to access Data.fs? I have my suspicions, but I'm not sure. I'm interested in this because I'm in the middle of a balloting period. People email me their ballots. I will need to check them for validity, send out confirmation or error messages, and eventually tally the results. I definitely want a persistent store, so ZODB seems good. Also, I eventually would like this to be a Zope product, with people able to access the results on the web (and, in the future, vote on the web). But getting it Zoped may be too much work in the short run. So, I'll have a Zope server running on my Debian system, and another process will be going through the mail. It would be more straightforwad if that other process worked directly with the database, rather than going through Zope via http or xml-rpc or something similar. So I'm wondering if that's possible. I can always take the indirect route if, as I suspect, that's necessary. I'm cribbing from parts of the Poll product as appropriate. The polling machinery is considerably more complex though: a restricted list of voters, various muckings around to get the info out of the email, and preference voting (1 for first choice, 2 for second, etc with an instant run off). I know Zeo is more in the multi-process space, but I don't want to use it for this project because I don't have time to learn it; it's alpha; and it seems like overkill for my problem.