Re: Comments on W3C Document Object Model Requirements

> From:          "Peter 1. Jones"<peter.jones@abs.gov.au>

> 1  For General Requirements item 8
> >>   8. It must be possible to read in a document and write out a
> structurally identical document to disk
> >>  (both documents can be represented by the same raw structural model)
> delete the words "to disk" because they are unnecessarily
> restrictive (and not matched by a constraint on reading).

So you think the concept is clear enough without saying "to disk"? We 
want to be clear that different implementations may or may not share 
the same way of implementing the DOM, but that results of the DOM 
operations will be the same. 

> 2  For Document Manipulation item 3
> >>  3. Operations must restore consistency before they return
> clarify the meaning of "Operations" in the context of the two
> preceding items [ie it may be necessary to combine several
> add/remove/change actions into a transaction-type operation if the
> individual actions do not maintain consistency].

Good point. We'll be talking about this when we get to the 
concurrency and threading discussions. 

> 3  For Event Model item 3
> >>  2. There will be interaction events, update events, and change events
> clarify the intended distinction between the three classes of event
> (especially, how they relate to the add/remove/change actions used
> elsewhere in the document).

We'll be clarifying this once we've started the discussions on the 
topic and know more about the in-depth requirements.

> 4  Remove (if it is not significant) or clarify the difference in
> wording between
> >>  Document Meta Information
> >>  These are requirements for what must be exposed ...
> >>    1. There will be information about the document and ...
> and
> >>  UA Information
> >>  These are requirements for information about the user agent environment
> ...
> >>    1. There will be a way of obtaining relevant information about ...
> [They currently read as "there must be exposure, and it must cover
> information" and "there must be information, and it must be
> exposable".]

We'll take a look at this before doing the next public release.

Thanks for your comments.

regards,


Lauren

Received on Monday, 23 February 1998 12:08:07 UTC