Rackspace Email & Apps Blog

3 Responses to “How We Use SPEWS at Webmail.us”

1. David Hewson says:

   January 11, 2006 at 3:43 am

   However you use SPEWS it doesn’t make for a satisfactory anti-spam solution for me. I moved our accounts to Webmail in December. On every other point I think you’re fantastic. But your spam handling seems diabolical – it is letting through messages that are such obvious spam (pharmaceuticals, ‘doctor’s’; stock reports, all manner of rubbish, and all from obviously fake addresses, some with the same subject line time after time). Worse, there’s nothing I can do to tackle them myself because your filters don’t work once your system has labelled mail spam. In other words you actually protect spam from being deleted by my own custom filters (instead of running filters BEFORE the spam filtering as most other systems appear to do).
   To be honest I am at the end of my tether with this subject. My last email provider was a small two man outfit with minimal support but very low costs. Its anti spam system – which included filters that acted on spam and a greylisting feature too – was a hundred times more effective than yours. Can you not learn something from people like Mailsnare? Greylisting is an option that won’t appeal to all people (it makes mail from unknown senders get sent twice, which rules out most fake addresses, and kills most spam, but introduce a slight mail delay). But it would be nice to have it as an option.
   As things stand at the moment I either have to change email addresses or change email provider if I want to get less spam. That would be a shame.
2. Bill Boebel says:

   January 11, 2006 at 4:48 pm

   Hi David. It was nice talking to you through Support earlier today. As we discussed, by “it is letting through messages that are such obvious spam”, you mean that we are delivering more spam than you would like to you Spam folder rather than flat out rejecting it during the SMTP session. So we are identifying it as spam, but you don’t want to see it at all – and I can understand that.
   About half of the spam that we block is at the SMTP level, and the other half is in the filters after the SMTP level. If we increased the strength of the SMTP-level filters much beyond its current level, there would be an unacceptable level of false positives, which creates a bigger problem for companies.
   In most cases, our customers prefer us to error on the side of caution and deliver tagged spam to the Spam folder rather than flat out reject it… that way the customer can still get to the messages if there was a false-positive.
3. Ben Hubbard says:

   January 15, 2006 at 7:28 pm

   For situations where a Webmail.us, Inc. user does not want to see spam at all, even mail we identify as spam, there is an option in the “Spam Filtering Preferences