[topicmapmail] PSIs - alternatives

[Patrick Durusau]

Alexander,

Well, I think it is time to simply admit that having unique persistent 
identifiers presumes that everyone would use the same identifiers for 
the same subjects.

If that were true, then we would not have the problems integrating 
information that we do now. Reinventing the notion of a universal name 
set that everyone will use, with or without repositories, is simply a 
non-starter. Even if it is prefaced with the "sacred" prefix http://.

What we can do is encourage people to write PSIs and treat PSIs as 
subject proxies and merge different PSIs that in at least the merging 
author's view, represent the same subject. I cannot really imagine US 
authors using PSIs coined by the Taliban nor vice versa, and both (at 
least in my opinion) are equally legitimate.

Rather than chasing the chimera of a universal language or naming 
convention, why not simply embrace the fact that for all sorts of 
reasons, some good, some less so, depending on where you stand, people 
want to identify their subjects differently?

PSI's have a vital role to play even in a diversity senario because they 
are a public notice that here is how I identify subject X. So if you use 
some other identifier and merge it with mine, then you can find that 
subject in either your materials or mine.

Allows all the "wild dogs" ;-) and others to do what is just in their 
own eyes and still enables integration of diverse information resources.

Hope you are having a great day!

Patrick

Alexander Johannesen wrote:

> On 6/20/06, Simon Grant <asimong at btinternet.com> wrote:
>
>> There could be a impartial host for PSIs - maybe topicmaps.org? -
>> respected by the entire community. That might work, if it were open
>> for everyone to put them in freely or for a small fee.
>
>
> I think this issue has been discussed ad nausseum, and I think the
> general thinking is that it really is too hard for a central
> repository to work; people throw in all sorts of stuff, nobody checks
> if something similar is already in place, people don't bother
> maintaining their PSI's (unless it brings cash for doing so), and it
> will all end in tears. I also think the general thought is that if
> some PSI's are truly valuable, they will organically be maintained in
> the systems that use them, and by that use evolutionary principles
> what stays and what goes. And since your PSI doesn't actually have to
> point to anything that exsists, it doesn't matter that the PSI as a
> URI works, and that means that people not only can but *will* break
> them over time.
>
> Some people take some ownership of certain PSI sets, and people reuse
> them, or publish new ones, and systems tries to stay somewhat up to
> date with it all. It comes down to persistant identifiers, not just in
> the TM world, but with any information system. It's a hard nut to
> crack, and people from all camps are trying to sort it out, not just
> us TM'ers. Haven't seen any one system that does it right, though, and
> the organic way most TM engines deals with it somewhat works, but this
> would be a good time for people in the know to throw a bone to us wild
> dogs.
>
>
> Regards,
>
> Alex


-- 
Patrick Durusau
Patrick at Durusau.net
Chair, V1 - Text Processing: Office and Publishing Systems Interface
Co-Editor, ISO 13250, Topic Maps -- Reference Model
Member, Text Encoding Initiative Board of Directors, 2003-2005

Topic Maps: Human, not artificial, intelligence at work!