- From: Stephen R. Savitzky <steve@rsv.ricoh.com>
- Date: 28 Sep 1999 09:31:48 -0700
- To: keshlam@us.ibm.com
- Cc: www-dom@w3.org
keshlam@us.ibm.com writes: > Quoth Stephen R. Savitzky: > > Can one pick and choose in this way, and > > still say you have "an API to this information set"? > > The Infoset is still a Working Draft, and the standard W3C boilerplate says > It is inappropriate to use W3C Working Drafts as reference material or to > cite them as other than "work in progress." > So statements about consistancy with it should be taken as indications of > intent; they aren't yet testable, since the Infoset may still be redesigned at > any moment. In which case the statement is in urgent need of rephrasing. > The DOM itself is still a work in progress, and doesn't yet do everything > people want it to. Trying to do everything that everyone wants is a guaranteed road to disaster. I'd rather have distinct DOM and InfoSet interfaces than DOM Level 50 piled on top of 49 ever-more-bloated special-purpose interfaces and years of regrettable decisions. The InfoSet API would then be free to model _exactly_ the information in the document, and nothing else. > Ignorable whitespace is an open issue... I'd settle for an "isWhitespace" flag and let the application decide whether it's ignorable. In fact, that's what I added in my implementation. > If we had it all to do over again, I'd want Infoset complete (with > namespaces and datatypes and schemas and so on) before folks nailed down > XML's syntax, never mind the DOM. But when dancing on the leading edge, > some of the parts you want just won't be ready in time; all you can do is > prioritize and go forward incrementally. But you can hardly claim even to _intend_ to have Level 2 be an API to the Infoset in that case. The best you can do is say that Level 2 is intended as an API to a subset of the Infoset as defined at the moment, and that _maybe_ Level 3 will be intended as an API to the Infoset as it might become. > <smile>Have patience please; keep raising the issues and helping us understand > where the priorities are, but be aware that we're already making mistakes just > as fast as we can...</smile> <smile>Then you need to get a computer. <quotation>A computer is a machine capable of doing the wrong thing a million times per second</quotation></smile> -- Stephen R. Savitzky <steve@rsv.ricoh.com> <http://rsv.ricoh.com/~steve/> Quote of the month: Death is nature's way of telling you to slow down. Chief Software Scientist, Ricoh Silicon Valley, Inc. Calif. Research Center voice: 650.496.5710 front desk: 650.496.5700 fax: 650.854.8740 home: <steve@theStarport.org> URL: http://theStarport.org/people/steve/
Received on Tuesday, 28 September 1999 12:32:22 UTC