On Thu, 19 Oct 2006, Roland Perry wrote:
> > "simple" being the key word there. If spam detection could be done with
> > simple rules, there wouldn't be a spam issue in the first place. So the
> > simple rules are likely to cause more harm than good.
>
> Not at all, the simple rules can eradicate the majority very quickly.
>
> For example, of someone sends an email to [log in to unmask]
>
> ... then I immediately classify it as Spam, because there's no-one of that
> name here. [A real example, received today].
Why did your mail system accept a mail for an address that didn't exist in
the first place, for you to have to go on and (manually?) classify it as
spam?
The most effective way to reduce spam is to employ technical measures at
the MTA to refuse to accept connections on various criteria (HELO checks,
greeting delays, sender/recipient validity checks, DNSBL checks, syntax
checks, and others - which broadly speaking cover the situations of
improperly configured mail servers, zombied PCs, signatures of spam
software, and lack of use of the proper email standards). My mail servers
refuse to accept about 80% of connections from the Internet offering us
mail on criteria like these, the remainder go on to be 'spam' checked,
whatever that may mean. At this point, simplistic checks don't do much
good.
(We are veering wildly off-topic now into mail administration stuff, and I
don't propose to continue this discussion any further on-list).
> > > If someone wants to send an email to lots of different people, without
> > > breaching privacy, they should use a mailing list application.
> >
> > I really see this as an issue for the ad-hoc "send this thing to lots of
> > people", rather than regulated, repeated sends over a period of time.
> > People seem unlikely to bother to prime any sort of mailing list
> > application for an ad-hoc mailout, in my experience.
>
> IMHO, a County Council is likely to have enough cause to email large
> numbers of concerned citizens about one project or another that they
> should have a proper system in place.
Quite right they should, taking this example. Maybe they do but the
employee 'forgot' -- who knows?
> > Are we bored of this thread now? :)
>
> The number of postings in a short period indicates that the list members
> are much more interested than average!
Heh.
Jethro.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Jethro R Binks
Computing Officer, IT Services
University Of Strathclyde, Glasgow, UK
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
All archives of messages are stored permanently and are
available to the world wide web community at large at
http://www.jiscmail.ac.uk/lists/data-protection.html
If you wish to leave this list please send the command
leave data-protection to [log in to unmask]
All user commands can be found at : -
http://www.jiscmail.ac.uk/help/commandref.htm
Any queries about sending or receiving message please send to the list owner
[log in to unmask]
(all commands go to [log in to unmask] not the list please)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
|