content, format, ctype, or xtype ?

[Rob Seaman]

Surely we don't need another inscrutable meta-meta-concept?!?  What is  
different about this and Norman's idea?

	http://www.ivoa.net/Documents/latest/utype-uri.html

It looks like the phrasing "explicit method" from Norman's quoting of  
the VOTable specification was itself later replaced with a more  
explicit example, but the underlying concept remains that a utype  
might denote a specific representational method.  "I am STC-S", not  
just "I am STC" - or "I am a restricted subset of ISO-8601", not "some  
might refer to me as a time-like quantity".  Heck section 4.5 even  
mentions STC:

	http://www.ivoa.net/Documents/WD/VOTable/VOTable-20080914.html#ToC29

At some point all this data modeling has to engage with the real  
world.  If utypes aren't where that happens, what's the point of 'em?   
Is there an advantage to pedantically having to say "I am a time-like  
quantity" in addition to "I am expressed as ISO-8601", when the latter  
always implies the former?

Why not a single enumerated vocabulary of utypes that includes a  
modest list of useful variations?

Rob
---

On May 4, 2009, at 12:33 AM, Francois Ochsenbein wrote:

>
> Yes it has to do with a semantic content of the field, but
> the 'utype' that I tried to propose several times does not
> seem to be satisfactory to TAP . There are 2 cases currently
> "iso8601" and "stc-s" to specify time stamps and STC expression.
>
> Personnally I still would prefer the 'utype' which has the
> advantage of being more accurate in its definition, and is
> versionable...
>
> -- Francois
>
>>
>> Hi Francois -
>>
>> It would help to have a list of common "data representations" which
>> we are trying to characterize with this attribute.  I originally
>> thought it was the numerical format, but it now sounds like it may  
>> be a
>> broader concept having to do with the semantic content of the field.
>> If we just list all the currently forseen values, perhaps we can
>> identify the concept.  What is "x"?.
>>
>> 	- Doug
>>
>>
>> On Sun, 3 May 2009, Francois Ochsenbein wrote:
>>
>>>
>>> Hi Pat, Mark, Doug,
>>>
>>> I would like to come to a final wording for the new attribute
>>> in VOTable which would specify the data representation of
>>> columns having a special meaning like STC-S or time. None of
>>> the various propositions (encoding, format, content, ctype)
>>> got real support; it seems that 'content' was felt as a
>>> bit more acceptable than the other terms. A last proposal
>>> could be 'xtype' (some condensed way of "expression type")
>>> would that be acceptable for tap ?
>>>
>>> See you soon,
>>> Francois
>>>
>
> = 
> ======================================================================
> Francois Ochsenbein    ------   Observatoire Astronomique de  
> Strasbourg
>   11, rue de l'Universite 67000 STRASBOURG  Phone: +33-(0)390 24 24 29
> Email: francois at astro.u-strasbg.fr (France)    Fax: +33-(0)390 24 24  
> 17
> = 
> ======================================================================