Friday, December 09, 2005

Whence predicates?

I was creating a new issue about improving the searching for issues (love those self-reflexive ones) this morning when I got to the bit about:

"This should be added to the Search Page so that the improvements can be combined with the other" ... thingies, yokies, whatchamacallems, ... predicates. Yup, that's the word - what does it mean?

So away with me to, but they had no definitions at all (!), so fall back to Google, which looked OK, in particular this one suggested it was right, and a further search confirmed the context, so I went back to the issue page and

Hang on one gosh-darned second there Mister Self, how come I knew it was the right word, if I didn't know what it meant?

Of course, even a semi-cynical world-view would accept this easily: how many people have you heard who rabbit on about stuff they don't understand (and can keep doing it till the next election). And there's that old favourite : "How do I know what I think till I see what I say?" (Why not "till I hear what I say?")

And I know that we do not "know meanings" in the same sense as a dictionary does. Classic examples such as "consciousness" and "intelligence" (where there is no agreed definition) spring to mind rapidly.

But this was at a different level - this wasn't just "not able to define", this was "haven't a clue what it means", and from it's coming up in my mind till searching found some definitions, I was aware of the dichotomy - I knew it was the right word, *and* I knew I didn't know what it meant.

"You knew at some unconscious level". That's just a cop out - we know everything at "some unconscious level". What is interesting here is which unconscious level ? And how does it work.

My hypothesis is probably biased by the MSc, but I think Copycat is the best explanation of this interruption (Work ! oh yeah, must do some of that):

I would suggest that the concepts that were floating around the particular issue "dragged in" other concepts, and words with them, one of which was "predicate". And this was strongly re-inforced by the phrase "search predicate". I'm a coder - of course I've heard that many times, and it was the first expansion of the search I tried (see above).

Maybe it is simpler than at first apparent: there are two concepts: "search predicate" and "predicate". One is used in programming, and the other turns out to be used in formal grammars and logics (I did know it was "an academic word"). The former was the correct concept but usage rules suggested dropping the "search" adjective - because the context was all about "searching for issues", and to use that word again would probably have been tautological, or at least irritating. Stripping off the "search" from "search predicates" leaves me with "an academic word" that I knew far less well, so away with me to ...

Hmm... that actually looks like a reasonable explanation. Another great mystery solved ! Back to work I guess.

Before I go - I don't actually know what "search predicates" means either, but that is closer to the classical: I can't define it (without Google), but I'd have a good idea of how to use it. And if there were users around I could define it for them.

Oh, yes: to be fair did come back fairly quick. But their definitions were all of the "academic word", so I wouldn't have solved the mystery if they had been been responsive (love those co-incidences too).

So, that's it then. I guess.


Well, coffee first.

"this one" - going back over the search for link insertion in this blog, I notice the link is not appropriate. Far better to notice that it was the 3rd item on Google's page which convinced me. And to notice that I did not notice the first item on that page, which was far more directly relevant.


Post a Comment

Links to this post:

Create a Link

<< Home