[ILUG] Obfuscation et al...

[Colm Buckley]

I know I said that my previous post on this subject would be the last,
but I'm afraid I must break my word...  I don't plan to perpetuate this
argument, but there are a couple of important points here.

(>> is me, > and >>> are Adam.)

>> [...] but *really* - you're posting to a *public* *mailing*
>> *list* and you expect your email address to be kept private?

> Private from other regular users, no, that would just be stupid. Private
> from spiders, yes, and why not?

Three reasons:

1) The software in its current state doesn't allow it, and I don't have
   time to maintain a patched version.

2) It's throwing effort into a black hole - the next step for
   spam-harvesting software would logically be to subscribe to the
   lists and gather addresses as they came in.  It's what I'd do, and
   I'm assuming that the 'pro' spam-harvester authors are at least as
   competent as I.

3) If the spam-gatherers are as proficient as you seem to think, there's
   not much benefit to be gained from obfuscating the addresses in a
   mechanical, systematic fashion.  If it can be
   mechanically-obfuscated. it can probably be mechanically-clarified.

>> [... possible filtering solutions ...]

> I run in Windows/Linux 50/50. But anyway - if there were like-minded
> people who were bugged by it - it's an an awful lot of hassle for many
> people, for a problem that can be fixed by one.

As I mentioned earlier, you're the only one who's complaining.  You
can't drag 'many people' into your argument without backup.  Other
evidence (an earlier message on this list) indicates that this is *not*
a problem.

As I see it now, you're concerned and incensed to be receive unsolicited
email, yet you can't/won't take technical steps to resolve it yourself
and you continue to post to a public forum.  I'm sorry, but I really
don't see this as my problem.

> Reply-to sender vs Reply-to list is valid because I get two messages,
> when I only require the one. More bandwidth, bigger phone bill.

As I mentioned before, it's not *my* fault that people send replies to
both sender and list.  And I *strongly* doubt that you get enough
replies to make any significant difference to your download time,
especially given that ILUG mails are text-only and generally fairly
short.  Even if it were the case, it wouldn't override my concerns
(explained in my last message) about the dangers of reply-to sender.

>> I've taken the steps in
>> http://mosa.unity.ncsu.edu/~brabec/antispam.html to help cut down on
>> email-address harvesting.

> Well, at least that's something. But that'll only stop the amateur
> spiders - it won't stop the pro's.

And what, exactly, is stopping these ruthless 'pros' from doing
something as simple as subscribing to the list?  If they're willing to
spend so much effort on one site for a return of a hundred or so
addresses, I'd expect them to find a way to get around whatever measures
I put in.

> And I'm not suggesting that you fork the mailman distribution, I'm
> suggesting that you add a patch. How often do you upgrade it for
> gods sake?

'Adding a patch' is the *definition* of forking the distribution...  and
I upgrade mailman whenever a new version appears in the Debian
distribution.  Or, at least, mailman is updated automatically whenever
such a new version appears.  I don't want to miss important security
fixes by switching to a locally-maintained version.

>> And, for the Nth time.  The reason replies go to the sender by default
>> instead of the list [...]

> This has been argued N times, and I'm not going to go into it again. I
> still haven't been presented with an argument that changes my mind.

Your obstinacy isn't my concern.  With the greatest respect, it's not
your mind which counts, it's mine.  *I* run the list software, not you.
System administration is not a democracy; the users place their trust in
the administrator to maintain the systems in the best fashion she or he
can - the administrators take *account* of the feelings and wishes of
the users, but in the end it is the admins' own intuition and expertise
which determines the course and configuration of the system.  If you
can't stand the heat, the living room is elsewhere, so to speak.

>>> A little politeness wouldn't go astray.
>> Matthew 7:3-5
> Laugh? I nearly did.

It wasn't a joke.

       Colm

-- 
Colm Buckley BA BF | NewWorld Commerce, 44 Westland Row, Dublin 2, Ireland
colm at tuatha.org (personal) | colm.buckley at nwcgroup.com (business)
+353 87 2469146 | whois cb3765 | http://www.tuatha.org/~colm/
The most common withdrawal symptom is a wet spot.