Slashdot Mirror


Microsoft Going After Hotmail Spammers

Mirkon writes "Quoth The Register: "Microsoft has targeted spammers with a lawsuit aimed at bulk mailers who harvest email addresses of Hotmail subscribers in order to bombard them with junk." Details are apparently sketchy at this point, but it's nice to see America's favorite monopoly putting its power to good use." The original news.com.com story is slightly more informative.

3 of 314 comments (clear)

  1. Just do this by eonblueye · · Score: 5, Informative

    Just keep your name out of their Member Directory and you will be spam free. I've had my Hotmail account for years spam free.

    --
    +++ David Watts 5495 0.0 0.5 1888 884
  2. Re:I took Hotmail spammers to mean ... by hoggoth · · Score: 5, Informative

    > spam that I get is from accounts like hotmail.com or yahoo.com

    Take a look at your full headers, those are forged.
    I filter out mail from @yahoo.com|@msn.com|@hotmail.com|@aol.com where the connecting host does NOT end in yahoo.com, msn.com, hotmail.com, or aol.com
    Just this alone got rid of 20% of my mail (all spam, never a false positive).

    --
    - For the complete works of Shakespeare: cat /dev/random (may take some time)
  3. Re:So what.... by babbage · · Score: 5, Informative
    I've been wondering about that since the Spam Conference last month, where both an engineer from Microsoft Research and a representative from Brightmail spoke about how they're trying to filter spam from large networks such as Hotmail and MSN. The scenario you describe is a common perception -- the most obvious explanation for the way even unused, "funny looking" (not dictionary words, numbers, etc) Hotmail addresses get so much spam is that the company must be selling their subscription list to spammers. But if that were actually true, then why are they putting so much effort into filtering out spam at both the network & mail client levels?

    A different idea that came up at the conference was what I'll describe as "bigger targets attract more arrows". That is, an ISP with millions of subscribers (Hotmail, Yahoo, AOL, Earthlink) is a more appealing target for things like dictionary attacks than, say, my personal DynDNS account with two legitimate users behind it.

    If you're going to carry out a dictionary attack against a domain, diminishing returns will take over for the little one (one billion tries, two hits -- 2e9%), but for the big one you can expect a reasonable hit rate (one billion tries, 3 million hits -- 0.003% -- and in fact a reasonably big fraction of all users on the network).

    In practice, this means today that the bigger the netwowrk, the greater the current spam volume, to the point that of the largest ISPs and corporate networks around today, something like 40% to 50% of their mail traffic is now spam.

    I think this is a better explanation for what's going at Hotmail et al., and it also does a better job of why they want so badly to control the spam issue. The explanation they'll give to the public is that this is good customer service, and to an extent that's true. But at the same time, trying to handle all this network traffic is probably a technical nightmare (and comments about the migration from FreeBSD to Win2000 are not helpful here :). For a free service, having to handle that much unwanted traffic is probably killing them, and bringing it under control for that reason is probably at least as important as maintaining customer good will.