> In all, I think that it is difficult to say that three searches that
> all aim for the same concept, but where each one retrieves a different
> result could be called a concept search. It seems that by definition,
> a concept search should retrieve the same results.
>
OK - this is good, as it leads to a clear difference in definition. For
me, a concept search doesn't, by definition, retrieve the same results.
I'm not sure I have a pithy definition of a concept search
unfortunately, but I'm reasonably sure that I don't see consistency as
part of that definition (not to say consistency isn't nice).
Owen,
This conversation seems to me very valuable, and I don't want to see
persons talk past one another. I think James is partly right and you
are partly right. Perhaps we can "both/and" this one.
I know its Friday, but I'm going to take a stab at getting philosophical
here, and I beg any real philosophers (hopefully, who have a practical
bent as well) to set me straight. :)
First, I am not a Platonist in that I do not hold that all words /
concepts have something like a perfect representation in a higher realm.
And yet - is there anything - facts, words, ideas, *concepts*, things,
objects, you name it - that are entirely consistent - or, said another
way, are of a unique kind? I think "yes". For example, despite
different languages, different cultural contexts, and languages
limitations, we all learn some shared concepts: "thirsty", "clouds",
"tears", "sad", "food", "mother", "father", etc. I think this is where
Jim's intuition comes in. At the same time, it is also obviously true
that in regards to the persons and things of this world we create unique
words / concepts (and even new meanings of words, although hopefully we
are up front about this) all the time (think of things like "hammer" and
"bottle" [and remember the movie "The Gods must be Crazy"! :)]). I
suspect this is where your intuition comes in, Owen. In any case,
despite this variability, it seems to me that there is a common human
foundation here.
Or... do we believe that words are only "power tools" we use to
"manipulate our environment or others", and are not far more deeply
significant? Or... do we believe that all facts / words / ideas /
concepts are hopelessly in dispute due to their being "impregnated by
culturally-constricting-conceptual-schemata" born of diverse purposes
and / or rivalry? I don't think this is true.*
So what does all of this mean, practically speaking? It means that
James is definitely on to something when he says "it seems that by
definition, a concept search should retrieve the same results". After
all, when we talk to or even write one another, we assume that persons
can understand us in a very real sense - and this assumption is often
right (I think it's especially amazing when we consider how important
body language is in communication). In other words, in conversation, we
"retrieve the same results" (consistency is nice indeed :) ). Could
this be because as human beings we share incredibly similar - perhaps
sometimes identical - concepts?
I think that historically, cataloging (hear the "Genius of Cataloging"
by Francis Miska, which I promoted here) takes this concrete reality for
granted, and builds on it (with "Scope notes", "See", "See also",
related, narrower, broader terms, etc.).
Also, it seems to me that in order for the semantic web to have any real
hope of working on a global scale (I am quite skeptical about this), it
must assume the same thing. But I am more confident about
human-to-human communication using the concept of "concepts" :) than
machine-to-machine or machine-to-human.
OK philosophers, fire away. I must limit myself to one post today and
won't be back until Monday.
Regards,
Nathan Rinne
Media Cataloging Technician
ISD 279 - Educational Service Center (ESC)
11200 93rd Ave. North
Maple Grove, MN. 55369
Work phone: 763-391-7183
* Also, as I've said before, when we deal with the world, all of us must
observe and categorize: we locate, describe (sometimes measure,
sometimes not), identify, etc., even if we are talking about enchanting
mythologies or the inner experiences of persons. We all perform
"either-or"-ing sometimes.
I suspect that these all "basic beliefs" we share ; in other words,
they are core assumptions *we can't not have* - because if they are not,
how can we speak of language as if it is related to the admittedly rich
and fuzzy concepts of "understanding" or "knowledge" in any real sense?
-----Original Message-----
From: Next generation catalogs for libraries
[mailto:NGC4LIB_at_listserv.nd.edu] On Behalf Of Stephens, Owen
Sent: Friday, January 04, 2008 9:32 AM
To: NGC4LIB_at_listserv.nd.edu
Subject: Re: [NGC4LIB] Relevance ranking: was Aqua Brow
Jim wrote:
> And I meant that
> searching "wwi" will get you primary documents *if* someone
> has tagged them in that way.
Yes - but that's true in a library catalogue as well of course - change
tagged to 'catalogued' of course, and only allow librarians to do this
;)
>
> In all, I think that it is difficult to say that three
> searches that all aim for the same concept, but where each
> one retrieves a different result could be called a concept
> search. It seems that by definition, a concept search should
> retrieve the same results.
>
OK - this is good, as it leads to a clear difference in definition. For
me, a concept search doesn't, by definition, retrieve the same results.
I'm not sure I have a pithy definition of a concept search
unfortunately, but I'm reasonably sure that I don't see consistency as
part of that definition (not to say consistency isn't nice).
> Additionally, perhaps if people point to the wikipedia page
> about Dostoyevsky using "dostoievski," that may work for the
> wikipedia page, but it still doesn't mean that by searching
> dostoievski, you will be finding what is available in Google
> about the Russian author.
>
This may well be true - but does it matter (a Friday afternoon question
if ever there was one)? Perhaps a more pertinent question is 'is it ever
going to be possible to find all available material on a topic given a
large body of available material?' - I would suggest that the answer to
this question is 'no' - so it becomes more an issue of doing the best
possible job. I don't know if Google does the 'best possible job' -
probably not in fact - but it does a reasonably good one for a large
number of people - and I suspect this is the best we can hope for.
I've been trying desparately to decide what this all means for libraries
(if anything), and although I keep getting glimmerings of ideas I don't
seem to be able to pin them down. However, a couple of bits of reading
seem relevant:
http://www.booksquare.com/quick-thoughts-on-book-search/ - which in turn
triggered:
http://radar.oreilly.com/archives/2006/12/book_search_sho.html
Received on Fri Jan 04 2008 - 13:02:58 EST