RE: Supporting arrays of concepts : node labels

Stella,

You've hit on exactly the issue that concerns me, which is how the network 
client knows what to do with what it gets back from the server. The 
ultimate aim, or user case if you prefer, is supporting a human-browsable 
display, both for alphabetical and for hierarchical types of display such 
as the AAT examples Doug talks about in his JoDI paper.

As you say, sometimes the order will be alphabetical and sometimes it will 
not. I think this mixture is even more frequent with taxonomies than with 
thesauri.

1. If it's not alphabetical, but some logical sequence, the client will 
want to respect the order in which the relations have been returned by the 
server.

2. If it's alphabetical, then of course we can ask "Alphabetical according 
to which language"? A series of relations sequenced by English labels will 
clearly not work if you want to display the French labels. That may not 
matter that much, if the client accepts the job of re-sorting all the 
relations by label for every language for every set of relations for every 
record that we want to display.

But the harder question is, How does the client know when the sequence is 
alphabetical (or is irrelevant) and when it's not (and is significant)?

Ron
>Ron Davies
>Information and documentation systems consultant
>Av. Baden-Powell 1  Bte 2, 1200 Brussels, Belgium
>Email:  ron@rondavies.be
>Tel:    +32 (0)2 770 33 51
>GSM:    +32 (0)484 502 393

At 19:15 9/05/2004, Stella Dextre Clarke wrote:
>Ron,
>Not sure I understand this question. Is it about the sequence among 
>sibling narrower terms? If so, then very often the sequence does not 
>matter and alphabetical order is the most convenient. But sometimes there 
>is a natural order, such as the order of size, or of age, or of location. 
>Presentation in the natural order helps people to understand the 
>underlying concepts better, detect omissions, overlaps etc. The person who 
>determines this is the thesaurus editor. The presentation becomes even 
>more helpful if node labels are inserted as in Leonard's example. (In this 
>example, notice that he found a natural order in some of the groups but 
>not others.)
>
>I hope I've been answering the right question?
>Stella
>
>*****************************************************
>Stella Dextre Clarke
>Information Consultant
>Luke House, West Hendred, Wantage, Oxon, OX12 8RR, UK
>Tel: 01235-833-298
>Fax: 01235-863-298
>SDClarke@LukeHouse.demon.co.uk
>*****************************************************
>
>-----Original Message-----
>From: public-esw-thes-request@w3.org 
>[mailto:public-esw-thes-request@w3.org] On Behalf Of Ron Davies
>Sent: 09 May 2004 13:26
>To: public-esw-thes@w3.org
>Subject: RE: Supporting arrays of concepts : node labels
>
>Let me apologize in advance for asking what might be a very naive question 
>at this stage, but there is a tremendous amount of material to get through 
>to try to get up to speed on current developments. Stella brings up an 
>issue that I was just trying to get a grip on, namely, the sequence in the 
>different concepts presented as of the same property. Is there any implied 
>sequence among, say, Narrower Terms for a given concept? If so, what is it 
>(or who determines it) and where is it indicated?
>
>Thanks very much.
>Ron
>
>Ron Davies
>Information and documentation systems consultant
>Av. Baden-Powell 1  Bte 2, 1200 Brussels, Belgium
>Email:  ron@rondavies.be
>Tel:    +32 (0)2 770 33 51
>GSM:    +32 (0)484 502 393
>
>At 13:43 9/05/2004, Stella Dextre Clarke wrote: