Re: Agenda for XML Core WG telcon of 2005 April 27

/ Norman Walsh <Norman.Walsh@Sun.COM> was heard to say:
| / Paul Grosso <pgrosso@arbortext.com> was heard to say:
| [...]
| | 7. xml:id.
|
| There's been a long thread on public-xml-id over the last few days
| suggesting that the name xmlid would be better than xml:id.
|
| I find four arguments almost, but not quite, convincing.
|
| 1. We are free to use any name we want, xmlid is equally reserved.
| 2. The xml:id attribute does have different semantics than the other,
|    existing xml:* attributes.

Bzzt. As was pointed out to me in the call, xml:base already doesn't
inherit as an attribute. So xml:id doesn't change the world with
respect to the semantics of xml:* attributes.

| 3. Local attributes are easier to access from an API than global
|    attributes.
| 4. The thread was started by an outside developer at a member company.
|    (They misunderstand some things, but they're as close to
|    "real customers" as we ever get, I think.)
|
| I may have missed some arguments, but those are the ones that stand
| out to me this morning.

                                        Be seeing you,
                                          norm

-- 
Norman.Walsh@Sun.COM / XML Standards Architect / Sun Microsystems, Inc.
NOTICE: This email message is for the sole use of the intended
recipient(s) and may contain confidential and privileged information.
Any unauthorized review, use, disclosure or distribution is prohibited.
If you are not the intended recipient, please contact the sender by
reply email and destroy all copies of the original message.