RE: DOM Level 1 Becomes a W3C Recommendation

At 06:42 PM 10/1/98 +0100, msabin@cromwellmedia.co.uk wrote:
>John Cowan wrote,
>> msabin@cromwellmedia.co.uk wrote:
>> 
>> > > * added text about HIERARCHY_REQUEST_ERR being raised
>> > > by Node.insertBefore/replaceChild/appendChild when the
>... 
>> Better that than allowing loops in the tree.
>
>... How much hand-holding do DOM users need?

Exactly. Apparently the DOM is really geared towards naive ECMAScript authors
for Netscape or IE browser clients.
Node.appendChild() is probably one of the most common method calls when
building
server-side documents (or parsing existing documents) and to purposely inhibit
efficient implementation is extremely short-sighted (I'm being polite).

I'm even more convinced there needs to be a split now between a "server" DOM
and a "naive-script-writer" DOM. If Level 2 does not address these issues then
I think a new group should be formed soon, otherwise what is the point? It is
obvious from the traffic on this list that a lot of the implementations are
getting
around these silly issues by exposing non-standard APIs. This undermines the
whole idea of a public recommendation and the possibility of
cross-implementation
compatibility.

Regards,
Claude Zervas
Adobe Systems, Inc.

Received on Thursday, 1 October 1998 15:09:44 UTC