On Wed, 15 Dec 2004, Mike Taylor wrote:
>> From: Dr Robert Sanderson <[log in to unmask]>

>>> Conclusion: the URIness or otherwise of a term's structure tells
>>> us nothing about whether that term is to be interpreted as a
>>> string or a term.  URIness is orthogonal.

>> Not at all. It tells you how you should break up the term.
>> foo.text any "http://a.b.c/d/e z3950r:a.b.c.d/e?f=g"

>> If you were to tell me that the term structure was word, I would find you
>> the following words:
>>    "http a b c d e z3950r a b c d e f g"
>>
>> But if you tell me that it's a string, I'll find you one string. If
>> you tel me it's a URI, then I'll find you two URIs like you
>> intended.
>
> Wow.  That is one contrived example!  :-)

Not at all.  When searching full text, finding URIs is very useful,
especially in footnotes or reference sections.

Another example:

zeerex.set =/cql.uri "info:srw/context-set/1/dc-v1.1"
vs
zeerex.set =/zeerex.oid "1.2.840.....3"

>> weather.temperature = 3
>> I'm going to treat that as a number unless told otherwise because

> deweyDecimalCode = 123.0
> You'd better not treat that as matching records with DDC "123".

Sure, because the index is DDC, not numeric.


>>> Let me re-state it this way: when I search for
>>>        foo < fruit
>>> is "fruit" to be interpreted as a string or a word?
>> Totally unspecified, I think.
> I agree.  That's my point.

Any default that you come up with will be wrong in some very obvious
cases.  You don't want to ever string search dc.description, whereas you
almost always want to string search dc.identifer.

As per my date example, I think the default interpretation is index
dependant.

Rob


       ,'/:.          Dr Robert Sanderson ([log in to unmask])
     ,'-/::::.        http://www.o-r-g.org/~azaroth/
   ,'--/::(@)::.      Dept. of Computer Science, Room 805
,'---/::::::::::.    University of Liverpool
____/:::::::::::::.  L5R Shop: http://www.cardsnotwords.com/
I L L U M I N A T I