Re: Re Deprecate/Undefined (was Request for status dump and issues check)

me> forbid is OK so long as it's clear it will stay forbidden 

> I'm not sure that this is a reasonable constraint. Many systems 
> evolve discontinuously, and that's one reason to have version 
> numbers. In the standards world, unlike the software world, the 
> pressure to upgrade can be resisted indefinitely if an urgent need is 
> met by the old standard but not a new one

Yes but there is a big difference between the general principle that
version n+1 of a standard will need to make some changes from version
n, and deliberately making a change now with the intention of changing
again later.

One possible argument for "forbid" (I think for instance this was Tim
Bray's argument) is that relative URI are wrong in principle for
namespace names.  If this is the reason for taking the forbid option
now then so be it, but that should be the decision and only changed
for namespace version 2 if there is some unforseen change of
circumstances or an overiding principle forces a change (neither of
which is particularly likely).

Another reason for choosing "forbid" now is just to put this discusion
on hold, so that the entire discusion has been a waste of time, and no
decisions will be taken at this time. The long term intention in this
case would be to make yet another incompatible change (after changing
from "literal" to "forbid") to change in version 2 to something else,
most likely one of the options already discussed on this list.

I fear the latter option is what is being suggested.

> Later versions, if they are marked in 
> the data, can then redefine or forbid relative URIs after an 
> appropriate discussion.

It is hard to see how any discussion would be any more appropriate
than the discussions here (or the earlier discussions on xml-plenary,
or xml-dev or presumably in the original WG). I can't see how it will
help to delay deciding this now. If the descision is for "forbid"
the wording of the spec should say why it is forbidden, and that
should state sufficiently clearly that the reason is that relative URI
are inappropriate as namespace names. If there isn't consensus that
they are inapproporiate then the forbid option shouldn't be taken.
Otherwise it is guaranteed that this whole discussion will start again
with equally cyclic results in a year or 18 months time, whenever NS 2
comes up as a possibility.

David

Received on Saturday, 24 June 2000 18:29:07 UTC