Friday, April 02, 2004

Mass Mailing

Sent out a few more mass mailings over the past few weeks. We've been seeing a lot more returns (around 10%) than in the past. I suspect we're butting heads against SPAM filters. It looks like our biggest problem is the address from which we're sending the email (postmaster@project2061.aaas.org). I believe a number of sites are now checking the domain on that address against MX records in the DNS. Since that domain doesn't have an MX record it's being rejected. I'm going to work with Dave on setting up a new server name on project2061mail.org (lists.project2061mail.org sounds good to me) from which we can send mail. Hopefully that will prevent a lot of the problems we've been seeing.

We do have a few emails being caught by content filters, but I don't know that there's a lot we can do there. Unfortunately, a lot of SPAM uses the same language we do because it is indicative of a legitimate message. On the other hand, a lot of spammers have resorted to random text in their messages, so maybe over time we'll be less likely to be cause. I don't know how the new HTML-based newsletter we're developing will fare. I suspect a lot worse, but I should have time to test against spamassassin before we send it. Plus we should have taken care of the domain issue by then as well.

Ed's talking about using the list server built into OS X. I'm not so sure about that, my experience with a lot of list software is that they use their own proprietary format. It's be nice if it could be modified to hook into a database like mysql or mssql, but I doubt it. I'll probably stick with my current system for the time being.

Speaking of systems, I still haven't had a chance to work on the user profile/contact system that I'd like to set up to replace the communications database. Though I've talked to Serita and Barb about what I'd like to do neither of them seemed to be concerned. True, a lot of what currently done manually will be taken care of automatically with the new system, but I suspect they haven't really thought about how this will affect the random work they do with the communications database.