Re: [xml-dev] "XML is just syntax" versus "Use semantic markup" (Is this

[Steve Newcomb <srn@coolheads.com>]

I think we're not understanding each other, Len.  I guess I 
didn't understand your note.

Maybe you're talking about a community's process of 
authoring a map, thinking that there has to be a starting 
point for such a process: a "base map".  I wasn't talking 
about that; my mistake.

I was talking about the importance of making the information 
of a given community (or individual) accessible to other 
communities and individuals, in terms that are different 
from the terms of the originating community.  I meant to say 
that, although despair about this possibility seems popular, 
it is not really warranted.  True, such accessibility 
requires the commitment of *human* effort, and such efforts 
are not nearly as cheap as Google's machine cycles, but 
there's no technical reason why the fruits of such efforts 
can't be easily re-used and re-exploited indefinitely, nor 
why technology can't be used to make such re-exploitation 
much cheaper than re-developing equivalent information in 
different terms would be.  (Note: not dirt cheap, and not 
100% automatic, but much cheaper, anyway.)  Even without 
cheap re-usability, the returns on mapping investments can 
be reasonable and attractive, as librarians and indexers 
have been demonstrating for many years.

Still, I think my comment is at least a little bit relevant, 
because of the problem of coming up with a base map in the 
first place.  It could be advantageous to start with 
somebody else's map.

Len wrote:
> Base maps aren't authoritative.  They are a means for sharing consensus so
> we can achieve more in community than we can alone, but a base map can be
> verified against the terrain before a symbology is applied.  The symbols are
> the ontology.

> If a map is incompatible with another map, that can be noted.  It can't
> always be resolved unless that commonly mapped is consulted.  If there is no
> commonality, there is no conflict.
> 
> len
> 
> 
> 
> From: Steve Newcomb [mailto:srn@coolheads.com]
>  
> John Sowa's "Lattice of Theories" notion is interesting. It 
> recognizes that it's useful to express intersections between 
> different universes of discourse governed by incompatible 
> ontologies.  The Topic Maps Reference Model is interesting, 
> too. It establishes a standard rhetoric for expressing such 
> wormholes.  In both cases, there's no requirement for a 
> "base map".  I think these kinds of ideas show the way 
> forward, because they sidestep any requirement that 
> everybody agrees about anything before information from 
> different perspectives can be integrated, or before 
> information expressed in terms of a given perspective can 
> become useful to people who don't share it.