Re: [XSCH] possible note skectch

Thanks Jeff,

can I suggest:

I concentrate on getting the value side nearer finished, and then you 
add a new subsection on the DL reasoning side of things. I think that is 
important but distinctly more in your domain of expertise than mine.

I had a further thought that it would be worth having a new short 
section on duration during the readers attention to the fact that F&O 
(??? or is it some other XSLT/XQuery WD) does solve the problems with 
duration by replacing it with yearMonthDuration and hourMinuteSecond 
duration (or something like that). This is another leftover from the 
recommendation round.

Jeremy


Jeff Pan wrote:
> On Wednesday, October 27, 2004 3:08 PM,  Jeremy Carroll wrote:
> 
> 
>>I have done a first pass at a note,
>>
>>http://lists.w3.org/Archives/Public/www-archive/2004Oct/att-0049/xsh-sw-note.html
>>this is very rough and ready but shows the sort of ground I would like 
>>to cover.
>>I hope to polish it a little tomorrow getting some of the links into place.
>>I'm hoping that during the F2F we can identify what is missing, and then 
>>post F2F to get it up to publishable quality.
> 
> 
> Here are a few comments/suggestions on the current draft. As the current draft is mainly a rough outline, the wording "... is unclear" in the following comments should be read as suggestions for further versions of the draft. 
> 
> 1) User defined datatypes:
> 
> - We should provide more details on why the DAML+OIL solution is a non-standard approach to fragID and why the id solution is only partially endorsed by RFC XMLMIMETYPE.  
> 
> - Example: we can modify the datatype name "foo" as "adultAge" if we replace 1700 with 18.
> 
> 2) Comparison of values
> 
> - We should present the current XML Schema solution before the three "new" solutions. Furthermore, we should compare RDF datatypes with XML Schema datatypes in order to make the situation clear. 
> 
> - The motivation of the "all primitive types different" solution is unclear.
> 
> - It is not clear how the XPath 2.0 eq operator solve the problem.
> 
> - We should provide a section about DL reasoning and datatypes, and then discuss its relations with the three "new" solutions.
> 
> Greetings,
> Jeff
> 
> --
> Jeff Z. Pan  ( http://DL-Web.man.ac.uk/ )
> School of Computer Science, The University of Manchester
> 
> 
>  
> 
> 
>>Jeremy
>>
>>
>>
>>
>>
>>
> 
>

Received on Wednesday, 17 November 2004 22:52:12 UTC