Monday, September 29, 2008

just when you think the writing is done .. dilemma.

by my estimate, i have two subsections to write, then a full-scale edit, and then i will have a preprint. let's hope it will be ready for the AMS secti0nal in october ..

.. that is, if anyone bothers to come to my talk and even then, if they bother to ask about my work.

inevitably i'm always scheduled against a more popular figure in a parallel session and the masses goes to his (or her) talk. a few colleague-friends stick around sometimes, if only to be nice.

you'd think that, upon seeing the finish line to be so near, this would be cause for comfort. then again .. come on.

if you've been reading this blog for a while, then you'd know:
there is never any cause for comfort or joy. q:

on friday i vowed to leave the grading in the office ..
(which, oddly enough, was a vow that i actually kept)

.. and worked on writing. when i tired of writing, i thought about where these ideas were going, what i want to prove next. that usually means a game of "let's pretend we know a lot and see what happens" and degenerates into depression, once the game stops and reality sets in.

then again, what's life without a little risk?

so i've started a new batch of research notes -- a kind of mathematical diary, if you will -- pertaining to this one c0njecture that i've been obsessing about. some months ago i told my mathematical sibs about an idea of mine which is related to other things. at the time it convinced them, but oddly enough, not me ..

.. partly because they didn't check the details and thereby failed to realise what might go wrong.

i don't blame them; i'm loathe to check the details of the ideas of others, too, mostly because i'm lazy. if i were left unchecked and i thought harder and past their ideas, it would also feel like stealing.

now it seems like the idea works. so what's the rub? my soon-to-be-preprint is supposed to be a special case of the conjecture.

i've proven all these lemmata just so the machinery runs the way that i want it to, in that one little case.

there would be similar machinery involved if this newish idea works -- similar, but sufficiently different that the proofs i've written already would require small but consistent modifications ..

.. and probably 2 more pages of some generalities. argh.

so, the dilemma: if i don't do some rewriting, then i have to reprove little tweaks of everything in the next paper as well as the newer machinery i need to run the newish idea.

i'm bleeding pages, already;
29 pages and counting, for the current writeup.

on the other hand, this is assuming that the idea really does work, and that i'm not having delusions of grandeur again. so unless i'm absolutely sure, why risk changing something that's perfectly fine?


i hate to say this, but sometimes i wish i could ask my late advisor for .. well, advice. i wished i asked him more often, when he was still alive.

1 comment:

Leonid said...
This comment has been removed by the author.