It's an interesting problem, though. I would be curious to see a
heuristic analysis. I have a boatload of these kinds of choices and I
have been meaning to get methodical about looking at them for ages
now. (Chronic postponement is also due for heuristic analysis.)
I use huge bloated graphics programs, especially 3D packages and
video packages, of which I am inordinately fond, but they have big
classpath and dll hiccups (especially with anti-virus programs) for
which, over the years I have resorted different regexs if not
outright different OS versions. Plus new versions of Java have always
been big tradeoffs in terms of developing personal systems. However,
whenever I have attempted to get whole systems synchronized, I always
go back to just building compatible sets and rebooting whenever I
need one that doesn't work with the current config on the box on
which I happen to be working. I keep promising myself that "Someday
I'm gonna get this right..." but it hasn't happened yet. OTOH I have
developed a cross-platform, cross-OS, cross-Language philosophy which
has ended up deciding for me that just being adaptable is the best
path for me. However, I do curse under my breath a lot.
Ciao,
Rex
At 8:28 AM -0500 6/10/05, Bullard, Claude L (Len) wrote:
>You could use a regex or separate them by element name.
>The latter simplfies the validation as to container
>but it does nothing with regards to the actual content
>unless you microparse (the regex). Actually, given
>any use of a typed text node where the text node
>value tells you the type, you always have to resort
>to microparsing for safe validation. It isn't that tricky.
>
>len
>
>
>From: Art Botterell [mailto:]
>
>On Jun 9, 2005, at 7:44 PM, Renato Iannella wrote:
>> Since a URN is part of the URI family, then we could have one element
>> that covers booth cases:
>
>Except that the valueListName (the URN) needs to be mandatory while
>the valueListLocation (the URI) would be optional. Not sure how to
>enforce that if they share an element.
>
>It's sort of like the dual use of <uri> in <resource>... relying on
>the particular data value to inform the semantics seems like it may
>be unnecessarily tricky.
>
>---------------------------------------------------------------------
>To unsubscribe from this mail list, you must leave the OASIS TC that
>generates this mail. You may a link to this group and all your TCs in OASIS
>at:
>https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php
--
Rex Brooks
President, CEO
Starbourne Communications Design
GeoAddress: 1361-A Addison
Berkeley, CA 94702
Tel: 510-849-2309
