If spam really isn't a problem, then the efforts involved in ultimately engineering a real solution, which in my opinion has to use economic forces, and the efforts involved in engineering easy user interfaces, will not be worth it. If spam really is a problem and is exponentially leading to email becoming unusable in most cases, then these engineering efforts are worth it and if the protocol aspects are not done in a standard manner, they will be done in proprietary ways. Some systems that would like to be closed would probably be delighted to provide a control whereby their users can very simply and very easily decided binarily whether they are connected to outside Internet mail or not, hoping that any foolish enough to choose to be connected will find their mailbox overflowing with so much spam so quickly that they will be driven back to only allowing communication with others in the closed system. Alternatively, if a user can't or won't configure any whitelist and their system just defaults to charging, say, 5 cents for each incoming email, and this behaviour were common, what would the effect be? If most spam went away, their cost for sending mail would be approximately cancelled by their income from receiving mail. Even if they never receive mail and send 5 messages a day, you are only talking about less than $8 a month. Whatever residual spam there was would respresent an income stream for them or their ISP. They wouldn't be able to subscribe to mailing lists (unless the mailing list operator went to a bunch of work and allowed people to pay for subscription to cover the delivery costs) but how bad is that? Better than email becoming worthless, I'd say. And how hard would it really be to have a good user interface for just whitelisting mailing lists? This is a complex space. I've always said it would be enormously difficult to deploy such a system. But I'm also not impressed by glib responses claiming its completely impossible because of user interface problems. Even if that were true for 95% of users, which I don't accept, it seems to me it would be better to have email be useful for 5% of users than for 0%. Donald ====================================================================== Donald E. Eastlake 3rd dee3@torque.pothole.com 155 Beaver Street +1-508-634-2066(h) +1-508-851-8280(w) Milford, MA 01757 USA Donald.Eastlake@motorola.com On Mon, 19 Aug 2002 Valdis.Kletnieks@vt.edu wrote: > Date: Mon, 19 Aug 2002 13:35:45 -0400 > From: Valdis.Kletnieks@vt.edu > To: Donald Eastlake 3rd <dee3@TORQUE.POTHOLE.COM> > Cc: ietf@ietf.org > Subject: Re: on UCE: Possible Interest (fwd) > > On Mon, 19 Aug 2002 09:22:59 EDT, Donald Eastlake 3rd <dee3@torque.pothole.com> said: > > > The cost is be set by the recipient. They maintain a white list of > > people / mailing lists that can send them mail without charge. They get > > to choose whether to trust the From address or require mail to be > > digitally signed to qualify for white list treatment or allow free mail > > if a special password header is present or whatever. Almost all mailing > > Been a while since you worked a help desk, huh? :) > > An amazingly large percentage of users have trouble understanding the > "please reply to this message to confirm your subscription" mail that is > sent by most popular mailing list management software. > > For that matter, an even more amazing number of these "please reply" messages > bounce because the user didn't configure their OWN e-mail address correctly > (today's favorite - John...Smith@mail..domain...com who obviously needs to > debounce that '.' key ;) > > And you want them to cut-and-paste the list address into a form AND get > all the right options set too? I think you're giving the users too much > credit for possessing operational neurons.... > -- > Valdis Kletnieks > Computer Systems Senior Engineer > Virginia Tech