From: "David A. Black" <dblack@...>
Date: 2001-09-30T06:38:46+09:00
Subject: [ruby-talk:21833] Re: ANN: HtmlDomGenerator 0.1

Hello --

On Sun, 30 Sep 2001, James Britt (rubydev) wrote:

> <snip />
> 
> > Well, since the DOM API is a very specific, ummm, specification, it's
> > misleading to call it a DOM parser unless it implements that API.  As
> > for lack of validation, just toss "non-validating" in front of
> > whatever else you call it :-)
> 
> There are all sorts of document object models, any of which
> could be referred to as a "DOM".  The W3C has a particular
> specification that defines their view of a DOM, and, as
> they did with XML schemas, they just unfortunately happened to 
> use a generic name for the spec.
> 
> So, if someone wants to refer to the W3C DOM, they should
> probably use the phrase 'W3C DOM Level 1', or whatever level
> is of concern.  Other people shouldn't be precluded from
> referring to a particular document object model as a DOM.
> 
> Now, that said, I understand that many people assume that
> there is only One True DOM, and it's perhaps best not to
> confuse people if it can be avoided.  But this could be
> cleared up in the RAA description with a statement that this
> is a distinct DOM API, not a W3C-defined API.

I'm not quite that doctrinaire :-)  but given the usage, I think it
is best to be clear, as you say.  There's a lot of this generic
naming around; imagine if you wanted to write a hypertext
markup language....


David

-- 
David Alan Black
usually dblack@candle.superlink.net
writing from dblack@saturn.superlink.net