Received: from localhost by CS.UTK.EDU with SMTP (cf v2.9s-UTK) id QAA08065; Thu, 1 Jun 1995 16:12:22 -0400 X-Resent-To: drums@CS.UTK.EDU ; Thu, 1 Jun 1995 16:12:20 EDT Errors-to: owner-drums@CS.UTK.EDU Received: from wilma.cs.utk.edu by CS.UTK.EDU with ESMTP (cf v2.9s-UTK) id QAA08051; Thu, 1 Jun 1995 16:12:18 -0400 Received: from LOCALHOST by wilma.cs.utk.edu with SMTP (cf v2.11c-UTK) id QAA15428; Thu, 1 Jun 1995 16:12:17 -0400 Message-Id: <199506012012.QAA15428@wilma.cs.utk.edu> X-URI: http://www.cs.utk.edu/~moore/ From: Keith Moore To: Mark Crispin cc: drums@CS.UTK.EDU, moore@CS.UTK.EDU Subject: Re: changes In-reply-to: Your message of "Thu, 01 Jun 1995 01:31:47 PDT." Date: Thu, 01 Jun 1995 16:12:11 -0400 Sender: moore@CS.UTK.EDU > > 1) Simplify the syntax, remove obnoxious restrictions: > a) Remove ".", "[", and "]" from specials. > b) Redefine local-part as word. > c) Redefine domain as atom. > d) Remove sub-domain, domain-literal, and domain-ref > > This is what most people have to implement in their parsers anyway, in order > to cope with: > From: Joseph T. Mooch @begin(ex-cathedra) I *do* consider this to be within our charter. We can make slight changes to the 822 grammar if there's a need to do so; especially if we know there are problems with what's in 822. And we can remove features of 822 that are never used. But we need to remain very compatible with any existing practice which is legal according to 822. @end(ex-cathedra) On brief inspection, it looks like your suggestion would be a great simplification of 822 without imposing much of a burden. (I especially like getting rid of the need to treat "." specially where surrounding white-space is concerned.) My main concern is that there may be user agents that refuse to deal with addresses of the form , and you can't blame them for doing so. Another problem with simplifying the grammar as you suggest is that we would then have several layers of grammar. If "." is no longer a "special" to 822 it will still be a "special" in a DNS name. This could be confusing. Even if we do make this change, user agents are going to need to generate things that are 822-compatible for the next several years. So it will be a long time before we benefit from any decrease in complexity. (I just recently removed the rule in my sendmail config files that rewrites " at " => "@", but I wouldn't have felt comfortable doing so until recently.) But that doesn't mean it's not worth trying to simplify 822 if we can. Keith