Spam: my slides from the presentation I gave at Toorcon 2004, ‘Spam Forensics: Reverse-Engineering Spammer Tactics’, are now up. Hope they prove enlightening ;)
Comments closedMonth: September 2004
Travel: Toorcon was great fun! Lots of interesting conversations.
Unfortunately they had a cruddy internet connection, so I’m majorly backlogged, and can’t write about any of it just yet ;)
Comments closedSpam: SpamAssassin 3.0.0 is now released! w00t! Only 4 months late this time ;) Announcement, techie details, Slashdot. New logo too:
(Note: if you’re running SpamAssassin 2.x and plan to upgrade, this is a new major release cycle — so we’ve taken the chance to break some backwards compatibility. Be sure to read the UPGRADE doc!)
Comments closedPolitics: So I was listening to that this morning. Did I hear correctly? Did Bush really say that one of the good side-effects of Iraq’s invasion, was that there were now hopefully less attacks inside other countries? sure looks like it:
‘Coalition forces now serving in Iraq are confronting the terrorists and foreign fighters so peaceful nations around the world will never have to face them within our own borders.’
I’m sure the Iraqi civilians will love that. ‘Hey guys, sorry about all the missing limbs, but you’re doing a really good job of being flypaper so we don’t get hurt. Cheers! Have a 15% corporate tax rate!‘
Comments closedConferences: Hey — I’m talking at ToorCon 2004 down in San Diego this weekend! Come along and check it out, if you can.
I’d better hurry up and file my presentation slides pronto ;) The topic is:
Comments closedSpam Forensics: Reverse-Engineering Spammer Tactics
In this talk, I’ll discuss how the SpamAssassin project has identified reliable signatures indicating that a message is spam, by reverse-engineering spammer tactics from the spam mails themselves. I’ll also discuss several specific features that we have identified, how we found them, and why the spammers add them.
Patents: Now that the summer break is over, software patents are back on the EU’s agenda. The FFII (via EDRI-gram) reports
On 24 September 2004, the European Council will probably meet to rubber-stamp the ‘political agreement’ achieved on 18 May 2004 on the highly controversial software patents directive (2002/0047 COM-COD).
According to the FFII the text was designed to mislead ministers about its real effects. ‘It consists of many sentences of the form or ‘software is … unpatentable, unless … [condition, which, upon closer scrutiny, turns to be always true]’.’ And, states FFII, ‘It can be said with certainty that only a minority of governments really agrees with what was negotiated, but several governments were misrepresented by their negotiators, who broke intra-ministerial agreements or even violated instructions from their superiors.’
More info:
- FFII Urgent call to National Governments and Parliaments (04.06.2004)
- FFII call for action against software patents (07.09.2004)
- FSF Europe open letter to EU presidency (06.09.2004)
Web: Flickr‘s latest trend — using just an eye (or similar minimalist face part) as your avatar pic:
Life: Luke writes:
Lean and I were joined by Lara at 6.10pm on Saturday 28th September. Lara is a little (8lb 7oz, so not _that_ little) girl. And she is gorgeous. Of course.
Congrats! I’ll be dropping in on the three of them next week, looking forward to it…
Comments closedFunny: The Daily Show’s GWB reelection film: ‘George W. Bush — Because He Says So’ (Quicktime MOV, 6MB). This is the funniest thing I’ve seen in ages.
Remember — don’t listen to the facts — listen to the words!
(thanks to anaxamander for the file. This URL is cached through CoralCDN, so pass it on!)
Comments closedWeb: My Nearly-Live Planetary Desktop Backgrounds site is now using NYU’s Coral Content Distribution Network instead of FreeCache.org. (FreeCache wasn’t caching the files, because they were too small. drat.)
Coral is a ‘decentralized, self-organizing, peer-to-peer web-content distribution network’, using a distributed sloppy hash table and peer-to-peer DNS redirection infrastructure.
At least, apparently. ;) I haven’t read the papers yet, but what I do know is that so far, it seems to be working perfectly — each file is requested exactly once by the CDN servers:
193.10.133.129 - - [31/Aug/2004:16:50:31 +0100] "GET /xplanet/tmp/200408311455.399750/day_clouds_800x600.png HTTP/1.1" 200 706936 "-" "CoralWebPrx/0.1 (See http://www.scs.cs.nyu.edu/coral/)"
and never requested again. That’s a big saving… nifty!
Comments closed