W3C home > Mailing lists > Public > public-xml-processing-model-comments@w3.org > June 2007

Re: decomposing viewport

From: Norman Walsh <ndw@nwalsh.com>
Date: Wed, 13 Jun 2007 14:53:27 -0400
To: "James Fuller" <james.fuller.2007@gmail.com>
Cc: public-xml-processing-model-comments@w3.org
Message-ID: <87hcpbr8ig.fsf@nwalsh.com>
/ "James Fuller" <james.fuller.2007@gmail.com> was heard to say:
| On 6/13/07, Norman Walsh <ndw@nwalsh.com> wrote:
| in any event, this raises the question of  p:insert, p:delete, and
| p:replace steps....
|
| one suggestion, why not roll up the 3 steos into one step called p:update ?

There's obviously a natural tension between lots of small steps with
limited functionality on the one hand and a few steps with more
functionality on the other. I don't think we've developed any hard and
fast rules for deciding where on that spectrum we want to be.

My own bias is towards steps with discrete functionality. One reason
for that is so that implementors can tune the individual steps more
easily.

|> | p:view-port
|> |
|
| and what about p:read-port ? perhaps a bit too 'active' in meaning.

Yes, that's too "verby".

I think p:viewport is a pretty good name: the step gives you a sliding
window, or viewport, into the source document and lets you make
changes to what appears in that viewport.

                                        Be seeing you,
                                          norm

-- 
Norman Walsh <ndw@nwalsh.com> | It is good to have an end to journey
http://nwalsh.com/            | toward; but it is the journey that
                              | matters, in the end.--Ursula K. LeGuin

Received on Wednesday, 13 June 2007 18:53:36 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:21:42 GMT