some motivating problems
Even though I've written a few papers (and a M.S. thesis), I'm still learning what it means to do research.
I have had some idea of where this (last six weeks of) work is going, even if I know I'm still, sometimes, anxious about doing Research the "Right" Way.
(There's another, older, essay that I wanted to link to, but I've forgotten how to find it...)
Reading through what others have written about what it means to grow up as a researcher, I realize that this summer's work is the first time I had insight into an important problem, questioned an underlying assumption, did more than a little follow-up -- thinking with some independence -- and tried to reason through, justify, and communicate some partially formed results, without "asking for permission".
It's that last bit that's new to me, "permissions", and I know I'm not used to it. It feels weird, not asking for permission as I have done before.
I guess in return, I ought to produce something more coherent and structured than my ad hoc stumblings here, which make sense to me but aren't linearized and (in)formalized for more general consumption.
As a start, I have to justify the importance of this problem, describe previous work and my approach, show some results, and make sure to justify those as well.
But before I do those things, let me step back and give context to this work.
I see distributional algorithmics as perhaps a chapter in my thesis, rather than as the main gist of it. I'm not sure, but a larger chunk of it might be organized as follows.
Theory: distributional algorithmics
Practice: (codename) Rspace
Systems: VLSD
I think the question of VLSD is worthwhile to pursue, but I think it's better to derive (systematic) principles after you've tried to build at least one system first. Otherwise, you'll be like a donkey trying to push your shopping cart around rather than try to pull it with you.
That said, Rspace is a big project, and perhaps not something I can do working only part-time over the summer by myself some 2,991 road miles from the Stnaford campus. So I might as well get my feet wet with a less systems-y project, write up some initial work, and connect the dots from there.
Why write a thesis anyway? Why journey towards a PhD at Stnaford?
A professor at Berkeley asked me this, and I told her that two of my signature strengths include creativity and diligence (slash self-discipline). Love of learning is somewhere up there too, and it makes sense to put yourself in a place to exercise your signature strengths every single day.
On a more pragmatic level, education means a lot to my family and the live cultures that I grew up in. It's also worth noting that, as was also mentioned in Condi, higher education takes on a different shade of meaning if you don't have passing privilege as a white male.
More women are earning college degrees, but, on average, it "means" less to them in terms of their economic independence. The storyteller in Condi notes that if you're a white man (or just share that statistical spirit?), you have the general choice of working for a corporation and moving up slowly through the promotional hierarchy, but that, observes the author, does not seem to work for women and/or minorities.
Although it's true that APIs are overrepresented in certain fields, industries, and academic institutions, I have to note that, as happens with women in America, higher education also translates more weakly into economic independence for API-Americans, on average, than it does for non-API-Americans.
When we intersect these dimensions, we should note that even though the gender gap (in average pay) between API women and men is non-trivially smaller (15% I believe, both in America /and/ China, interestingly) than the gender gap for non-APIs, API-Americans themselves are seriously underrepresented the more you scale up the levels of corporate hierarchy.
Hmm...
We must also note that this seems relatively specific to API-Americans -- it's worse for APIs in America, statistically, than it is for Whites, Blacks, and Latinos.
Phew, that was a longer spiel than I had expected, but I might as well write about my motivation somewhere. Even if I wasn't born with a latent predisposition for this kind of work and then later biopsychosocialized into it, I should remember that a higher education in America "means" more to me than it does for that well-known roster of mostly-pass-as-white males that dropped out and changed the world along the way, even if, it means "less" in the end as well.
I have had some idea of where this (last six weeks of) work is going, even if I know I'm still, sometimes, anxious about doing Research the "Right" Way.
(There's another, older, essay that I wanted to link to, but I've forgotten how to find it...)
Reading through what others have written about what it means to grow up as a researcher, I realize that this summer's work is the first time I had insight into an important problem, questioned an underlying assumption, did more than a little follow-up -- thinking with some independence -- and tried to reason through, justify, and communicate some partially formed results, without "asking for permission".
It's that last bit that's new to me, "permissions", and I know I'm not used to it. It feels weird, not asking for permission as I have done before.
I guess in return, I ought to produce something more coherent and structured than my ad hoc stumblings here, which make sense to me but aren't linearized and (in)formalized for more general consumption.
As a start, I have to justify the importance of this problem, describe previous work and my approach, show some results, and make sure to justify those as well.
But before I do those things, let me step back and give context to this work.
I see distributional algorithmics as perhaps a chapter in my thesis, rather than as the main gist of it. I'm not sure, but a larger chunk of it might be organized as follows.
Theory: distributional algorithmics
Practice: (codename) Rspace
Systems: VLSD
I think the question of VLSD is worthwhile to pursue, but I think it's better to derive (systematic) principles after you've tried to build at least one system first. Otherwise, you'll be like a donkey trying to push your shopping cart around rather than try to pull it with you.
That said, Rspace is a big project, and perhaps not something I can do working only part-time over the summer by myself some 2,991 road miles from the Stnaford campus. So I might as well get my feet wet with a less systems-y project, write up some initial work, and connect the dots from there.
Why write a thesis anyway? Why journey towards a PhD at Stnaford?
A professor at Berkeley asked me this, and I told her that two of my signature strengths include creativity and diligence (slash self-discipline). Love of learning is somewhere up there too, and it makes sense to put yourself in a place to exercise your signature strengths every single day.
On a more pragmatic level, education means a lot to my family and the live cultures that I grew up in. It's also worth noting that, as was also mentioned in Condi, higher education takes on a different shade of meaning if you don't have passing privilege as a white male.
More women are earning college degrees, but, on average, it "means" less to them in terms of their economic independence. The storyteller in Condi notes that if you're a white man (or just share that statistical spirit?), you have the general choice of working for a corporation and moving up slowly through the promotional hierarchy, but that, observes the author, does not seem to work for women and/or minorities.
Although it's true that APIs are overrepresented in certain fields, industries, and academic institutions, I have to note that, as happens with women in America, higher education also translates more weakly into economic independence for API-Americans, on average, than it does for non-API-Americans.
When we intersect these dimensions, we should note that even though the gender gap (in average pay) between API women and men is non-trivially smaller (15% I believe, both in America /and/ China, interestingly) than the gender gap for non-APIs, API-Americans themselves are seriously underrepresented the more you scale up the levels of corporate hierarchy.
Hmm...
We must also note that this seems relatively specific to API-Americans -- it's worse for APIs in America, statistically, than it is for Whites, Blacks, and Latinos.
Phew, that was a longer spiel than I had expected, but I might as well write about my motivation somewhere. Even if I wasn't born with a latent predisposition for this kind of work and then later biopsychosocialized into it, I should remember that a higher education in America "means" more to me than it does for that well-known roster of mostly-pass-as-white males that dropped out and changed the world along the way, even if, it means "less" in the end as well.
posted by L. Wu @ 9:04 AM
0 Comments:
Post a Comment
<< Home