[topicmapmail] PRD structure/content (Public Resource Descriptors)

[Simon Grant]

Having read the whole of Steve's (very useful) paper as well as the 
slides, I've changed the Subject line, because that is where I feel 
constructive discussion could focus. Recalling some extracts from 
Steve's earlier post...

At 14:16 2006-06-20, Steve Pepper wrote:
>I think we need to be very careful about what kind of 
>machine-readable content we suggest that people put in their subject 
>indicators (or Public Resource Descriptors, to use the terminology 
>of [1]). Having said that, some kind of referral mechanism, by which 
>publishers can deprecate their own (outdated) PSIs in favour of 
>others, would probably be useful.
>[...]
>In general I think we should encourage people to put as few 
>assertions as possible in their PRDs (subject indicators): the 
>purpose of a PSI/PRI is simply to help us know when we are talking 
>about the same thing, not whether we hold the same views about that 
>thing; including assertions (even "innocent" assertions such as 
>near-equivalence) are therefore not only irrelevant, they are 
>directly counter-productive.
>
>On the other hand, that kind of assertion could certainly be 
>justified as part of a service intended to aid PSI discovery.

I suggest:

1. deciding about how the structure and content of a PRD can help, in 
several ways
2. embodying that in a simple spec for XML docs that can be used with 
a CSS to produce an easily readable web page - a bit like FOAF I guess.
(The alternative would be an XHTML microformat: I wouldn't want to 
argue the point either way.)

My earlier suggestion, which amounts to including equivalent PRIs in 
a PRD, would help in two ways:
a) by aiding discovery of equivalent PRIs
b) by simply representing consensus (or lack of it) between different 
authorities in a machine-processable manner
c) by providing greater stability than any one domain provider can

To use Miles Thompson's example...
- a number of us could be interested in information about companies
- we could all publish PRI/Ds
- in cases where the company structure was very simple, we could all 
include each others' PRIs in our PRDs.
- in more complex corporate structure, someone could publish a set of 
PRI/Ds just to capture the structure of (say) IBM.
- others could choose, from the descriptions, the best one to include
- in actual topic maps, an author could choose a small number of PRIs

One other feature now seems really attractive to me: to include a 
statement of *non*-identical PRIs, where the purpose of the PRI/D is 
partly to distinguish a set of related or confusable subjects. 
Compare Wikipedia's disambiguation pages. This mechanism would also 
be a good way to enable one authority to distance themselves from 
another dubious authority who was either putting in spurious 
equivalences or inaccurate descriptions.

OK, I  won't go on to step 2 now, let's leave it at that, but just to 
say that there would be ways of finding documents with that 
particular XML format - even just with Google.

No central registries. No top-down, except for agreeing on the spec. 
The only down side I can see is that the minting process is a little 
more complex. If you like, it's like outsourcing the work that would 
have had to go on within a central organisation to the periphery. 
While we're waiting for the spec (needn't take long) people could 
start off following Steve in just putting a human-readable web page.

Simon

--
Simon Grant  http://www.simongrant.org/home.html
Please continue to use my established e-mail address
a (just by itself) (at) simongrant.org