- From: Ian Hickson <ian@hixie.ch>
- Date: Sat, 6 Dec 2003 13:50:36 +0000 (UTC)
- To: "B. Johannessen" <bob@db.org>
- Cc: www-archive@w3.org
On Thu, 4 Dec 2003, B. Johannessen wrote: > > http://www.hixie.ch/specs/html/forms/xforms-basic > > The draft directs comments to htmlforms@damowmow.com and > www-archive@w3.org which i guess are mailing lists. www-archive is the public archival system at the W3C: http://lists.w3.org/Archives/Public/www-archive/ ...and htmlforms@damowmow.com is an alias for ian@hixie.ch. >> email: >> "An e-mail address, as defined by [RFC822] (the mailbox token, >> defined in RFC822 section 6.1)." > > First of all, if you're going to refer to an RFC it should probably > be RFC 2822 which obsoletes RFC 822. Good idea. > Second; please don't force client and server programmers to parse > the horrible syntax of an RFC 2822 "mailbox". If it's unclear why > this is a bad idea, I can come up with a few examples of valid > RFC 2822 "mailbox"-es to show how bad it can get. Or you can just > believe me when i say it's a disaster :-) Would you rather require that every form author write a parser instead? Currently, forms generally try to validate e-mail addresses and frequently do so in a naive way, leading to many people not being able to use those forms with their e-mail addresses. (For example, people with periods in their username.) By requiring that the UA do this instead of the author, we reduce the number of implementations that need to be debugged by maybe six orders of magnitude. >> maxlength: >> "...When this attribute is specified on text or password >> controls, the server must not submit a form with the control >> having more..." > > You probably meant to say "..the client must not..." Yup, thanks. -- Ian Hickson )\._.,--....,'``. fL U+1047E /, _.. \ _\ ;`._ ,. http://index.hixie.ch/ `._.-(,_..'--(,_..'`-.;.'
Received on Saturday, 6 December 2003 08:51:02 UTC