... But that's not because of any inherent difference in the language. It is because of the way people are _using_ the language. ... Java, yes. C, no (except...
Oren Ben-Kiki
orenbk@...
Nov 1, 2000 10:09 am
3852
... Is problematic when you need to hold the data between calls... ... Not really. The same problem happens in UNIX pipes, Tcp/IP sockets, etc. You just live...
Oren Ben-Kiki
orenbk@...
Nov 1, 2000 11:56 am
3853
... We have Clark's proposal. ... We have the proposal to copy the namespace from the parent element. ... I don't think we have a concrete proposal to that...
Oren Ben-Kiki
orenbk@...
Nov 1, 2000 12:06 pm
3854
... From: Oren Ben-Kiki <orenbk@...> ... And the way we are attaching namespace to parent element is ... ? Rgds.Paul....
Paul Tchistopolskii
pault12@...
Nov 1, 2000 4:27 pm
3855
... Yes, but the problem is, for a filter of N stages, it requires N copies of the relevant information be performed. ... I was just thinking along the lines...
Clark C. Evans
cce@...
Nov 1, 2000 10:00 pm
3856
I'm thinking that our API to MinML should be completely based on an "iterator interface"; where the "strength" of the iterator is explicitly defined. ...
Clark C. Evans
cce@...
Nov 1, 2000 10:01 pm
3857
Ok. Given a MinML text, how do we define references (XLink/XPointer?) Just wondering if anyone out there had any thoughts on the matter... Is liking "much...
Clark C. Evans
cce@...
Nov 1, 2000 10:01 pm
3858
... Paul, in the syntax, we use: xmlns='http:\\server\namespace-resource.minml' in the MinML text -- where the namespace of a child is inherited from its...
Clark C. Evans
cce@...
Nov 1, 2000 10:02 pm
3859
... Just as long as it is not assumed that the Lexer interface is "lower level" than the other interface. For example, if we are using the API over a MinML...
Clark C. Evans
cce@...
Nov 1, 2000 10:03 pm
3860
... element. ... When I use the word 'bytecode' I mean MinML syntax ( because it is what MinML syntax really is ). Sorry, I'l not use confuzing terminology any...
Paul Tchistopolskii
pault12@...
Nov 1, 2000 11:47 pm
3861
Looks like there'll be a Common XML presentation at XML 2000! Right now, it's just Bob LaQuey (if he can make it) and myself on the podium. I'll post more...
Simon St.Laurent
simonstl@...
Nov 2, 2000 12:16 am
3862
... Hmm. Well, I don't think we officially voted on that yet... Clark...
Clark C. Evans
cce@...
Nov 2, 2000 12:22 am
3863
... Wrong. It wasn't even a formal proposal. Clark made a proposal, I described a variation of MinXML I was using to support namespaces in my projects which...
Don Park
donpark@...
Nov 2, 2000 5:23 am
3864
... No it doesn't. The ultimate consumer provides a buffer, and the rest can just use that space. Admittedly, this ultimate consumer has to provide a ...
Oren Ben-Kiki
orenbk@...
Nov 2, 2000 9:46 am
3865
... Well, why should one use something silly like a lexer API to present a database as a MinML object model? Convert the database to the MinML object model...
Oren Ben-Kiki
orenbk@...
Nov 2, 2000 9:46 am
3866
... I'd vote for the alternative - let's give it a name, SNIP (Syntax for Namespaces using Implied Prefixes). SNIP in a Nutshell: Use prefixes, as per Clark's...
Oren Ben-Kiki
orenbk@...
Nov 2, 2000 9:54 am
3867
... Surely, either our original reasoning on attributes was wrong and attributes should be allowed back, or your namespace proposal should look like this: ...
Philip Nye
philip@...
Nov 2, 2000 10:36 am
3868
... an ... always ... So far as I can see the scope of the namespace is the same in both SNIP and SNUG. SNUG uses a "special" attribute while SNIP uses a...
Philip Nye
philip@...
Nov 2, 2000 12:20 pm
3869
... Before Don's suggestion, I was leaning towards SNIP without any defaulting mechansim. After the 'verbosity' complaints, I'll modify my position to SNUG...
Clark C. Evans
cce@...
Nov 2, 2000 1:01 pm
3870
... Ahh. (Thwaks self on head). ... Kewl. Ok. We have two interfaces then, the "parser" interface and the "lexer" interface. ... Nice review / requirement...
Clark C. Evans
cce@...
Nov 2, 2000 1:05 pm
3871
... What's wrong with: <urn..loc.gov..books.local-name> <urn..ISBN..0-395-36341-6.local-name> Or any other reasonable encoding mechanism? ... I don't see this...
Oren Ben-Kiki
orenbk@...
Nov 2, 2000 1:18 pm
3872
... Date: Thu, 2 Nov 2000 09:00:48 -0500 (EST) From: Clark C. Evans <cce@...> To: Oren Ben-Kiki <orenbk@...> Subject: RE: [sml-dev]...
Clark C. Evans
cce@...
Nov 2, 2000 1:51 pm
3873
... Nothing wrong with this - remind me what the objection to colons for MinML namespaces is. Philip...
Philip Nye
philip@...
Nov 2, 2000 2:04 pm
3874
... Use of "prefix:local" without a corresponding "xmlns:prefix='...'" won't pass through an XML system. Clark...
Clark C. Evans
cce@...
Nov 2, 2000 2:08 pm
3875
... I think this is a mirage. It only applies to totally static namespace (no changes in content over time) or when the system has access dynamically (and ...
Philip Nye
philip@...
Nov 2, 2000 2:11 pm
3876
... Thanks, An observation: This won't pass a namespace aware XML system but as I understand it the XML namespace spec is already in conflict with some other...
Philip Nye
philip@...
Nov 2, 2000 2:59 pm
3877
... From John Larmouth's book: <quote> The underlying concept for object identifiers is a tree-structure. Each object identifier value corresponds to precisely...
Philip Nye
philip@...
Nov 2, 2000 2:59 pm
3878
... When my computer programs encounter a new namespace (a particular snapshot of a given vocabulary in time) they want to know: (a) what are the acceptable...
Clark C. Evans
cce@...
Nov 2, 2000 3:00 pm
3879
... Why? SNIP uses '.' to separate a prefix from the local name. The prefix is anything up to the last '.' in the tag name; if there's no '.', it is considered...
Oren Ben-Kiki
orenbk@...
Nov 2, 2000 3:17 pm
3880
... Round trips via XML - it might rewrite the prefixes. Have fun, Oren Ben-Kiki...
Having problems with message search?
Fill out this form to ensure your group is one of the first to be migrated to the new message search system.
