Definitions (was something else)

[Rob Seaman]

Let's emphasize the agreement here:

> At the moment, the definitions are optional, and I do not think this  
> will change. (Certainly it will not change in the skos standard.)

Now then:

>> Can we instead collect a separate master list of numbered  
>> definitions that are linked to each entry, rather than embedded?
>>

> I do not favour this approach. Since we are only needing definitions  
> for one vocabulary, these should be placed directly into the  
> vocabulary.

Well, I'll extend my argument by pointing out that definitions like:

1722 - JohnsonU:  a bandpass of the "JohnsonPhotometricSystem"; see  
"TransmissionCurve" "JohnsonUCurve"
1723 - JohnsonB:  a bandpass of the "JohnsonPhotometricSystem"; see  
"TransmissionCurve" "JohnsonBCurve"
1724 - JohnsonV:  a bandpass of the "JohnsonPhotometricSystem"; see  
"TransmissionCurve"e "JohnsonVCurve"
1725 - JohnsonR:  a bandpass of the "JohnsonPhotometricSystem"; see  
"TransmissionCurve"e "JohnsonRCurve"
1726 - JohnsonI:  a bandpass of the "JohnsonPhotometricSystem"; see  
"TransmissionCurve" "JohnsonICurve"

...already have external links, either explicitly or implicitly.

We might also want to link internally:

1722 - JohnsonU:  a bandpass of the "JohnsonPhotometricSystem"; see  
"TransmissionCurve" "JohnsonUCurve"
1723 - JohnsonB:  see 1722; see "TransmissionCurve" "JohnsonBCurve"
1724 - JohnsonV:  see 1722; see "TransmissionCurve" "JohnsonVCurve"
1725 - JohnsonR:  see 1722; see "TransmissionCurve" "JohnsonRCurve"
1726 - JohnsonI:  see 1722; see "TransmissionCurve" "JohnsonICurve"

By relying explicitly on links to external sources, including  
definitions, the need to link internally is avoided.

What is usage in other fields?  Do they provide definitions?  How are  
the definitions organized?  Can definitions be linked to supporting  
information such as graphs and tables?

The statement is that we should embed the definitions since we  
currently only need these for one vocabulary.  What do we do when we  
later decide that we need a second vocabulary?

For that matter, let's assume our efforts will be flagrantly  
successful and will be embraced by the larger community.  Thus, that  
the maintainers of these external vocabularies we're adopting will  
cheerfully decide to build on the joint efforts and will want to add  
definitions.  Do we want multiple cross-referenced lists with  
duplicate definitions?

While we're at it (and since my daughter is spending a semester abroad  
studying Tibetan culture) - comments about multi-language support?   
With external links it would be trivial to link English terms to  
French definitions - and also English and French definitions to  
Russian terms - that is, a many to many mapping.  What is best  
practice for this in other fields?

English is (currently) the lingua franca of science, but the VO has  
many requirements for public outreach (that will be emphasized more- 
and-more as we go forward).  Surely these requirements will be quite  
evident in VO vocabulary efforts.  Let's not build in unnecessary  
limitations.

We should:  1) minimize initial effort, and 2) maximize future  
capabilities.

Rob