[Om3] Two more CDBase questions

[Michael Kohlhase]

Christoph LANGE wrote:
> Dear James,
>
> On Monday 12 May 2008 22:50:20 James Davenport wrote:
>   
>>> And here's now one more question about a thing that either is not
>>> specified or
>>> that I'm too stupid to see. For OpenMath objects we say that if a cdbase
>>> is not given for a symbol, it is inherited from a parent. But what if
>>> there is an
>>> OpenMath object within a CD (say /CD/CDDefinition/FMP/OMOBJ) where no
>>> cdbase is defined? Does that mean that the default CDBase
>>> http://www.openmath.org/cd
>>> applies, or does the CDBase of that content dictionary apply, i.e.
>>> /CD/CDBase?
>>> I think the latter would make more sense, but on the other hand it would
>>> make
>>> processing easier if we forced CD authors to explicitly specify the
>>> cdbase for
>>> every OpenMath object, if they want it to be different from the default
>>> one.
>>>       
>> No - I think the former would be much easier to use. Otherwise a piece of
>> OM cxan change its meaning if it is cut/pasted around. I would like to be
>> able to read a definition of a symbol from a CD without having to check
>> whether ordinary-looking symbols had been redefined.
>>     
>
> Indeed, that sounds convincing; I had not thought of the copy/paste case.
>   
I think that whatever we do, the current setup with cdbase being 
inherited from parent elements (whether that be OM Object-internal 
inheritance as specified in OM2 or the case for CDs and CDSignatures 
where you are essentially proposing document-level inheritance), we will 
have problems with copy and paste. The only safeguard against this is to 
put full cdbase into every OMS. But this is what we decided against in 
the name of legibility. But of course for OM3 we would decide 
differently on the grounds that we have learned something.

Michael
> But when I think about the CD editing use case, there is actually another
> thing that is not clear to me. Imagine somebody is authoring a new CD
> http://example.org/math/my.ocd, and most symbols in this CD "my" are defined
> using symbols from the standard CDs at openmath.org. In any syntax that I can
> imagine this would look quite clusmy, so I should offer some support for it in
> SWiM, but I'm not yet sure what and how. Consider for example
>
> <CD>
>   <CDName>my</CDName>
>   <CDBase>http://example.org/math</CDBase>
>   <CDDefinition>
>     <Name>foo</Name>
>     <CMP>foo + 1 = bar</CMP>
>     <FMP>
>       <OMOBJ>
>         <OMA>
> 	  <OMS cd="relation1" name="eq"/>
> 	  <OMA>
> 	    <OMS cd="arith1" name="plus"/>
> 	    <OMS cdbase="http://example.org/math" cd="my" name="foo"/>
> 	    <OMI>1</OMI>
> 	  </OMA>
> 	  <OMS cdbase="http://example.org/math" cd="my" name="bar"/>
>         </OMA>
>       </OMOBJ>
>     </FMP>
>   </CDDefinition>
>   <CDDefinition>
>     <Name>bar</Name>
>     <CMP>bar - 1 = foo</CMP>
>     <FMP>
>       <OMOBJ cdbase="http://example.org/math">
>         <OMA>
> 	  <OMS cdbase="http://www.openmath.org/cd" cd="relation1" name="eq"/>
> 	  <OMA>
> 	    <OMS cdbase="http://www.openmath.org/cd" cd="arith1" name="minus"/>
> 	    <OMS cd="my" name="bar"/>
> 	    <OMI>1</OMI>
> 	  </OMA>
> 	  <OMS cd="my" name="bar"/>
>         </OMA>
>       </OMOBJ>
>     </FMP>
>   </CDDefinition>
> </CD>
>
> Neither the CDDefinition of foo nor the one of bar is really easy to edit
> manually, so I think that an intelligent editor could probably auto-generate
> cdbase attributes if it recognizes that one of the well-known default CD names
> is used, or that the name of a CD in the local knowledge base (e.g. cd="my" in
> this example) is used -- as long as none of these is ambiguous.
>
> What do you think?
>
> Best,
>
> Christoph
>
>   
> ------------------------------------------------------------------------
>
> _______________________________________________
> Om3 mailing list
> Om3 at openmath.org
> http://openmath.org/mailman/listinfo/om3
>   

-- 
----------------------------------------------------------------------
 Prof. Dr. Michael Kohlhase,       Office: Research 1, Room 62 
 Professor of Computer Science     Campus Ring 12, 
 School of Engineering & Science   D-28759 Bremen, Germany
 Jacobs University Bremen*         tel/fax: +49 421 200-3140/-493140
 m.kohlhase at jacobs-university.de http://kwarc.info/kohlhase 
 skype: m.kohlhase   * International University Bremen until Feb. 2007
----------------------------------------------------------------------

-------------- next part --------------
A non-text attachment was scrubbed...
Name: m_kohlhase.vcf
Type: text/x-vcard
Size: 320 bytes
Desc: not available
Url : http://openmath.org/pipermail/om3/attachments/20080513/9bb81f3b/attachment-0001.vcf