If the system is put into place that [-SPAM-] can go to a folder or specific mailbox, there is no need for this extra step with mxcore now deleting messages silently.
That was the plan. Having said that, feedback so far has actually been very positive. I expect we will revisit this next week and make a decision then.
Yeah, well no surprises there esp as they don't necessarily know or understand what might be going on and PN aren't really saying either ! All they are seeing is little or none of the explicit spam (which is a good thing) but are blissfully unaware that there might be implications of other mail going missing either now or at some point in the future.
What I simply don't want to end up with here is PN tweaking the system to reject more and more spam so that on the surface it looks as though there really isn't any significant problem following the breach - but with a side effect of "a very small number of customers might possibly get a very small amount of genuine mail rejected" to quote the sort of wording one could reasonably expect to see in a PN service.status announcement ! Whilst PN might well consider that to be an acceptable situation I don't. I do appreciate efforts being made to reduce the impact of spam in general but not if it errs on the side of over enthusiastic or erroneous deletion.
What often seems to happen with other ISPs who appear to be doing this kind of detection (simply to protect their servers by reducing volume rather than anything else) is that genuine mail IS deleted or bounced and other related problems DO occur. For instance, a colleague is subscribed to various yahoo and similar groups and receives individual e-mails from these. Every time some @rse spams the group(s) with iffy looking or sounding messages, the ISP bounces them as 'content rejected' or suchlike which results in yahoo stopping all further messages being sent to that address until it is reactivated. Anyone here who uses yahoo groups no doubt knows that they are a bit quick in blocking addresses but very slow in letting you know that they have. The result is that one or more times a week, one or more addresses are blocked by yahoo due to a single bounce, lots of genuine messages are lost and it's a right old PITA all round. Yes I appreciate that PN appear to be silently deleting rather than bouncing so this kind of thing may not be a problem but I still want to understand fully what risk there is of any genuine mail getting silently deleted. Similarly, consider again the hotmail (and others) problem with the Booking Confirmations from a Ticket Agency that I mentioned earlier. The decision to silently delete all those 1000's of e-mails was apparently the result of a scoring-type system deciding that the messages were most likely spam rather than being some of the 150K genuine and desperately wanted messages that were sent out in a very short space of time. The users of the services who rejected these messages were completely unaware that their service provider could and indeed was deciding which mails they could read and which they could not. The deletion was automatic on receipt and occurred regardless of specific users' account settings to use a 'junk' folder or similar.
So, am I going to get more official detailed info as asked for (particularly on this ClamSpam thing) or should I just stop asking because PN are going to completely ignore the requests ?
Unfortunately some peoples mail systems treat these lists as black blacklists and reject the mail out of hand. Plusnet use these as part of the scoring system which results in our mail being tagged.
AIUI, the change Plusnet have made is to treat more of the blacklists at the blacker end of the scale as outright rejections because monitoring has proven that no genuine mail has come from servers on this list.
Understand all that fully and it all sounds very good of course BUT (and there's always a 'but' isn't there) How do I know that PN isn't one of those mail systems that could screw up when making the decision to delete ? How do I know that monitoring has in fact 'proven' that no genuine mail has ever been deleted ? I mean, in all fairness rather than intended to be insulting, PN doesn't have a particularly good track record in testing and monitoring anything does it. The only thing likely to flag a problem is customers complaining and in perhaps the majority of cases they wouldn't even be aware that genuine mail had been deleted. Also, I still get the impression that we are talking about 'silent deletion' on receipt here and not simply 'tagging' iffy stuff. Tagging, whilst being a bit of a PITA to have to manually check 'junk' folders when you haven't ever needed to before is not so bad but it's the possibility of silent deletion or worse still silent bouncing that concerns me in all this. It's all too easy for any ISP to dismiss any claims of stuff going missing as "just one of those things that happens sometimes and nothing directly to do with us" so even if you do know that something has got deleted by mistake, no one is ever likely to listen or do anything about it.