I think the issue here is how we can plan a better transition than what we have done for SMTP. That is, if we start thinking about "8+ everywhere" instead of "ACE everywhere" scenario, then how we can get there from here. That is the question for the IDN group I believe. Edmon ----- Original Message ----- From: "John Stracke" <jstracke@incentivesystems.com> To: <ietf@ietf.org> Cc: <idn@ops.ietf.org>; <ietf-smtp@imc.org>; <owner-ietf@ietf.org> Sent: Tuesday, April 02, 2002 9:54 AM Subject: Re: [idn] Re: 7 bits forever! > D. J. Bernstein wrote: > > >It is not reasonable to demand that this > >7-bit garbage be allowed to remain on the network forever, at the > >expense of all future implementors and users. > > OK, you're obviously in a better position than I am to appreciate the cost > to implementors. But users? Suppose you could streamline qmail by > dropping support for talking to 7-bit servers. How do you convince a > postmaster to install it? The bandwidth cost of having to use 7-bit > encoding is real, but small; the cost of losing connectivity to sites is > potentially high. Worse, it's almost impossible to put an upper bound on > the cost of lost connectivity. For a company, it could mean losing > messages (and sales) from important customers; businesses have failed over > such things. > > For commercial implementors, it's a no-brainer: dropping 7-bit support > will lose customers, so don't do it. For open-source implementors such as > yourself, naturally, things are less clear-cut. > > /==============================================================\ > |John Stracke |Principal Engineer | > |jstracke@incentivesystems.com |Incentive Systems, Inc. | > |http://www.incentivesystems.com |My opinions are my own. | > |==============================================================| > |"This [OLE] isn't an API, this is a shaggy dog story in C!" --| > |Unknown | > \==============================================================/ > >