On Thu, 2014-06-12 at 19:21 +0100, G.W. Haywood wrote: > Hi there, > > On Thu, 12 Jun 2014, Patrick O'Callaghan wrote: > > > On Thu, 2014-06-12 at 16:04 +0100, G.W. Haywood wrote: > >> The right place is the mail server, which should recognize that > >> the mail is junk AND REJECT IT. ... > > > > This is far too simplistic. Not all mail identified as spam is in fact > > spam, and only the final recipient can make the call. There is no one > > canonically right answer for every situation, so there is a place for > > client-side spam filtering in many use cases. > > Please pay attention. > > I didn't say that the final recipient shouldn't make the call, I said > that he should make the call before the mail is accepted by his mail > server, so that the spammer doesn't get another $currency_unit for > sending his spam. This necessarily means that no mail client sees the > spam, which in turn means that Evolution is not the place to do this. > > Clearer? > I'm perfectly sure that poc knew exactly what you were saying, no matter how condescendingly you put it. The issue is that when you reject mail at smtp time you are explicitly relying on the accuracy of an automated system to determine what is, or is not, junk. Unless, of course, you sit and watch the SMTP traffic for all the mail you receive and press a button to accept or reject it.
Sure, running your own mail server gives you the ability to implement all manner of heuristics and grey listing and scoring so you can get close to the nirvana of a 100% spam classification - but you will never get 100% and, more importantly, not everyone has the skill set or the resources or the desire to become their own mail admin. Clearer? P. _______________________________________________ evolution-list mailing list evolution-list@gnome.org To change your list options or unsubscribe, visit ... https://mail.gnome.org/mailman/listinfo/evolution-list
