[ale] (no subject) SPAM talk...

[Bjorn Dittmer-Roche]

On Sat, 17 Apr 2004, George Carless wrote:

> > Ouch! Well drew does have a point about obfuscation and it's backed up by
> > actual RESEARCH. I find it frustrating to read a thread like this where
> > noone has actually backed up their claims with facts. Of course this kind
> > of research becomes outdated fast, so maybe that's why noone bothered, but
> > FWIW here's one very good citation with solid research:
> >
> > http://www.cdt.org/speech/spam/030319spamreport.shtml
>
> http://www.unicom.com/chrome/a/000388.html

This article says that obfuscating the addresses on the web can still be
read by the software they tested.

The article I cited says that obfuscated e-mail address are seldom
used by spammers in practice.

I think the article I cited is more relevant since we are asking the
question "does obfuscation reduce spam?" Spammers in the future may change
their ways. (although that is far from certain, why bother spamming people
who will probably also have a spam filter, after all)

> The problem is that obfuscation can only ever be a (very) short-term
> solution.  The people developoing the spamming software are relatively
> quick, and relatively sophisticated - and they have an obvious vested
> interest in keeping track of new methods of obfuscation (and rendering
> them ineffective), while the rest of us who're running mailing lists
> etc. don't have the same time/resources/etc. to dedicate to the problem.
> The logical conclusion of which is that all obfuscation of email
> addresses can lead to is more and more obscure/ugly archives with no
> real benefit in terms of preventing spam.  And the more
> additional/superfluous code that needs to be written to support this
> obfuscation, the greater the load upon admins -- increasing the
> difficulty of administering a list/archive/whatever, reducing the amount
> of time available for more important tasks, and increasing the risk of
> admins misconfiguring something when faced with an increasing complex
> environment.

If it's really that hard to configure mailman to do a little obfuscation
I'll take your word for it, since I've never worked with it. My only point
is that obfuscation DOES reduce spam.

Apple's Developer site provides another solution: password protect the
site but make it easy for a human to figure out the password, and,
personally, I appreciate any site that makes an effort in this area as my
spam filter is *far* from 100% effective. (although apple's solution makes
them ungoogleable, which is too bad)

I am only casting my vote that some effort be made.

        bjorn