mad props!! Re: [olug] Fw: OLUG.PHPCONSULTING.COM

[Christopher Cashell]

At Mon, 29 Sep 03, Unidentified Flying Banana Brandon Lederer, said:
> I dont necessarily consider that "mangled".  If I were writing a harvesting 
> script it would look for the std name at domain.com
> 
> and name at domain dot com, switch at to @ and dot to ., stip whitespace, and 
> there ya go.... about 2 or 3 lines in perl.

Remember, though, that even e-mail harvesting spiders that just pulling
in the standard user at domain.com will pull in a *huge* number of false
positives.  You end up with tons of e-mail addresses that are no longer
valid, that are spam-proofed in some way, things that look like e-mail
addresses but aren't (user at host masks from irc logs, etc).

Trying to pull in 'mangled' e-mail addresses is going to increase your
number of dead or bad addresses enormously.  Remember too, that there
are dozens (if not hundreds) of minor variations that you can do to the
'user at host dot com' type of mangling, which can make programatically
converting them to usable e-mail addresses very complicated.

>From what I've seen of most spammers, they're not exactly people who are
interested in anything that might resemble increased work or time, so I
doubt they'd bother with any complicated unmangling attempts.
Especially when there's no need to bother.  Processing web pages and
newsgroups will get you all the e-mail addresses you want as fast as you
want them.

> however I do appreciate the effort.

I've only actually seen one e-mail harvesting program up close and in
use, and that one didn't even try to read mangled addresses.  I can't
say for sure that none of them do, but I'd be very surprised if many of
them bother with it.

-- 
| Christopher
+------------------------------------------------+
| A: No.                                         |
| Q: Should I include quotations after my reply? |
+------------------------------------------------+