'Meta' terminology (was RE: N3 contexts vs RDF reification)

pat hayes <phayes@ai.uwf.edu> wrote:

>>pat hayes <phayes@ai.uwf.edu> wrote:
>>
>> >>
>> >>My understanding:
>> >>
>> >>Summary
>> >>=======
>> >>There are at least two competing proposals for representing contexts in
>>RDF.
>> >>The concept of 'context', although similar, differs slightly with
respect
>>to
>> >>'higher-order' statements, ('reification' and making statements about
>> >>statements).
>> >
>> >I wonder, could I make a plea that y'all change your terminology here
>> >slightly? The term 'higher-order' already has an accepted usage now
>> >for about80 years, and it isn't this, so this is likely to cause all
>> >kinds of confusion and wasted time. What you are talking about is
>> >meta-language statements (statements about other statements), not
>> >higher-order statements.
>>

[snip]
BTW, I forgot to mention how impressed I was at your grasp of these matters.

>>Yes, higher-order is incorrect terminology.  I am a bit confused about
>>calling it a meta-language, though.  It depends on whether RDF is a
>>'language' or not.  Agreed certain RDF 'vocabularies' can be thought of as
>>languages, so in that sense it would hold true.
>
>Yes, if we are being very persnickety Im not sure of using any
>logical terms when talking about RDF.  But 'meta' seems appropriate
>to the 'reification' terminology; at least, thats how I have been
>understanding reification in the RDF context, ie if A reifies B then
>A is a statement in the metalanguage of the language of B. I'm not
>sure that I really grock reification in RDF, though.
>

Not 'persnickety' ;-), but following your lead on trying to be as accurate
as possible in describing statements about statements.  I totally agree with
you that the terminology should be right to avoid any source of confusion.

Meta-languages have a different meaning to me - i.e. a *higher-level* (if
that term is ok with you) language used to *create other languages* (I could
be wrong).

I agree the term 'meta' is useful, my interpretation is that it just means
'a higher kind of'.  Perhaps we could call them 'meta-statements'?

>>Though presumably I could more generally call it "meta-metadata" - the
first
>>layer of statements represent data about data (metadata) - the next layer
is
>>considered to be data about metadata (meta-metadata), ad nauseam.
>
>We need a new term. How about metatadata?

If there are no objections, I like the existing terminology of 'metadata'
(data about data).  Now, metadata is also data, hence 'meta-metadata',
'meta-meta-metadata', etc. all collapse to be called simply 'metadata'.

I.e. all statements (regardless of layers of abstraction above the raw
resources they describe) can generally be called metadata, or else if the
specific layer of abstraction is important in the context they are spoken
of, then meta-meta-metadata is clearly 3 layers of abstraction above the raw
data (just count the number of 'meta' words).

More precisely, any level of abstraction, n, is also a member of the n-1
level of abstraction recursively (ie. also n-2, n-3, etc).

I would propose describing raw resources as 'data', statements as
'statements' or 'metadata', statements about statements as 'meta-statements'
or 'meta-metadata', etc.

>
>Pat Hayes
>
>---------------------------------------------------------------------
>IHMC					(850)434 8903   home
>40 South Alcaniz St.			(850)202 4416   office
>Pensacola,  FL 32501			(850)202 4440   fax
>phayes@ai.uwf.edu
>http://www.coginst.uwf.edu/~phayes


regards
Lee

Received on Friday, 27 April 2001 02:30:26 UTC