W3C home > Mailing lists > Public > www-dom@w3.org > October to December 1998

Re: XML APIs

From: Stephen R. Savitzky <steve@rsv.ricoh.com>
Date: 03 Nov 1998 09:54:36 -0800
To: DOM List <www-dom@w3.org>
Message-ID: <qc1znkslur.fsf@gelion.crc.ricoh.com>
Jonathan Robie <jonathan@texcel.no> writes:

> Stephen R. Savitzky wrote:
> 
> > At this point I'm probably going to stop hoping that iterators get back into
> > the DOM at some point, and instead count on programmers to design
> > appropriate classes of their own that don't depend on whatever bizarre
> > single-application-specific behavior somebody decides to throw into the spec
> > because they think they know better than I do what my application wants to
> > do with a parse tree.
>  
> I can't let the phrase "bizarre single-application-specific behavior
> somebody decides to throw into the spec" slip by, though. What "single
> application" are you talking about?

Consider live nodelists for getElementsByTagName, for example.  This,
according to previous postings on this list, exists to support scripting in
browsers which attempt to start running the script while the document is
still being read in.  It supports at most two application programs:
Microsoft's and Netscape's browsers.  It does so at the expense of either
requiring getElementsByTagName node-lists to traverse an average of half the
tree for each call to "item", or adding a significant overhead to every
element and every editing operation.

Keeping the children of EntityReference up-to-date when the corresponding
Entity is changed is a similar nightmare.  The whole treatment of entities
is aimed at applications that _read_ XML documents, and is unuseable
(without extensions) by applications that write or edit them.

>                                     Have you looked at the membership of
> the DOM WG? SoftQuad, ArborText, Texcel, INSO, Microsoft, Netscape, IBM,
> Sun, iMall, webMethods, Merryl Lynch... we do radically different things.

...but the process appears to have been seriously distorted by the desire to
get something out the door in time for the next release of the two browsers
mentioned above.  

> And "somebody decides to throw into the spec"? In fact, one of the
> criticisms people made was that we spent too much time agonizing over
> various design decisions, and Level 1 took a long time to come out.

During which time I made major implementation decisions based on the
expected presence of iterators.  I can still keep _my_ iterators, but I have
little confidence in the ability of the WG to come up with something in the
spec that I can use with any degree of efficiency.

-- 
 Stephen R. Savitzky   Chief Software Scientist, Ricoh Silicon Valley, Inc., 
<steve@rsv.ricoh.com>                            California Research Center
 voice: 650.496.5710   fax: 650.854.8740    URL: http://rsv.ricoh.com/~steve/
  home: <steve@starport.com> URL: http://www.starport.com/people/steve/
Received on Tuesday, 3 November 1998 12:49:18 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 22 June 2012 06:13:46 GMT