[topicmapmail] Question of the week : the TMDM PSIs in TM engines

[Lars Marius Garshol]

* Robert Cerny
>
> If you have would have a system where anyone can make decisions  
> about subject identity and those would be propagated in one way or  
> another, it would turn out as a tower of babel. And on the other  
> hand: whenever i use a PSI, i might am making a subject identity  
> decision ("my subject is the one that the PSI describes"), so  
> whenever i use a PSI in a system where there is a lot of topic map  
> exchange, i might perform an act of "pollution" with possibly  
> devastating effects.

That's true. Ann Wrightson did a presentation on this at Extreme in  
2002:
http://www.idealliance.org/papers/extreme/proceedings//html/2002/Wrightson01/EML2002Wrightson01.html

And "devastating" is no exaggeration. I made a mistake in an autogen  
script once that gave all topics the same subject identifier. The  
result was a topic map with just one topic that had all names,  
occurrences, and associations. Not very useful.

--Lars M.
http://www.garshol.priv.no/blog/
http://www.garshol.priv.no/tmphoto/