Thursday, January 19, 2012

the more things stay the same, the more they change.

so tomorrow is the fourth [1] time that i'll be giving this particular talk, regarding differentιable structures on metrιc spaces. i'm hoping that it will be the last, for a little while.

call me paranoid, but i feel uneasy when i talk about the same thing too often; i worry that i'm not getting new ideas or making enough progress on my research.

some of my colleagues have attended this same talk several times, those poor souls .. but i think i see their motivations. i've done the same thing myself. at least for me, it's hard for me to learn something new in the short span of a talk, even if it is a 2-hour period, so often a good talk is worth hearing again.

once, this caused no end of confusion:
it was right after my ph.d. defense, and there were two back-to-back conferences. i was worried that i'd bore my colleagues to death by giving my thesis talk twice. so in one i stuck to my thesis, and in the other i discussed a completely different project in preparation.

makes sense, right?
besides, maybe people will be impressed at my versatility ..! [2]

of course, i was wrong. once i told my colleagues, they were alarmed and made a fuss about it. you see, both conferences held parallel sessions of contributed talks, and my talk was running opposite a well-known and prominent researcher in my field. so a lot of people, expecting to hear it next week, skipped my talk and went to his.
oh well; we're all young and stupid for once in our lives, right?

so i've learned my lesson .. somewhat. i still can't bring myself to give the exact same talk. besides, i've not yet given the version of the talk that i wanted!

you see, the driving mechanism behind the theorem consists of genuinely new techniques for dοubling metrιc spaces, which are inherently geοmetric in nature. in every past talk, i've spent too much time expositing and only hinted at this.
perhaps it's a risky proposition to discuss techniques of proof during a seminar, but i want to show the audience why the theorem is actually geοmetric.

the seminar is two hours long (2 x 45min), which must be enough time if i forgo the pleasantries. i think i will say a lot and write little, at least regarding the parts that should be "well-known."
at any rate, after this i'm retiring the talk for a while. besides, there's this new project that i want to talk about instead, in mid-february .. (-:

anyways, i'm rewriting part 2 of the talk now. back to work!

[1] it also depends on how you count. last march i talked about it at an AMS Special Session, but i had promised different things in the abstract and only advertised the theorem. besides, it was a weaker theorem then.

[2] thinking back, it was a particularly stupid decision on my part. i ended up writing and preparing two talks where i could have written one. to add to the context: this was also during the three weeks that my thesis corrections were due, and time was very much a factor ..!

1 comment:

Leonid said...

Disclaimers are getting old. There's a better solution: invite Jεnna Burκe to guest-blog here.