[nfbcs] Research

David Tseng davidct1209 at gmail.com
Sun Feb 26 08:49:37 UTC 2017


Missed this one earlier. Responding below:

On Thu, Feb 23, 2017 at 10:15 AM, John Heim <john at johnheim.com> wrote:
Well, maybe I am misreading your intent but you said, " Skip the CS
undergrad unless you're in a lower tier school [...]" I don't think it
matters what tier the school is in when it comes to willingness to make
accomodations. In general, the best thing for anyone's career is to go to
the best school they can get into. Having a degree from MIT on your resume
is better than having one from East Podunk College.

Most universities do everything they can to keep students from dropping
out. Here at the University of Wisconsin Math Department, we have at least
3 full-time people who's job is to keep students from flunking math
courses. Their job is to make  sure students are taking courses they can
handle and that they get the help they need if they are struggling. And
that's just the math department. Other departments probably have their own
people.

My observation was simply from about a ~15 year career in software
engineering at various big name companies, having the opportunity and
privilege to know lots of blind devs, the ones that were happy with their
CS studies, were generally those who went later as part of a masters. And,
who said anything about dropping out? I think most students just scrape by.
The killer courses are usually curved anyway, so even if you really blew
it, you're usually curved to a B-/C. I recall this one real analysis class
where my sighted peers and I all blew it :).

Whether my original assertion  is true or not, I cannot say. However, from
my own undergraduate experiences, I would say you're better off waiting.
If you take a look at the top 4 uni's for CS in the US (CMU, MIT, Stanford,
and Berkeley), and the kind of classes you have to take each
quarter/semester, you'll also find grad school might be "easier" since the
undergrad route is just plain harder. You're gonna take 2 technical classes
per quarter/semester all 4 years in addition to whatever else you're
required to take.
That's a hell of a way to go with coursework, learning new tools, judged
against your peers (curve), and fighting a11y issues.
If you're sure you're going to get fabulous support (apparently Stanford
does), then by all means, go for it. However, a masters from MIT looks far
more impressive than a bachelors where you scraped by and didn't have the
time to really absorb the material. Ultimately, it doesn't matter as long
as you have a degree in CS. In industry at least, it matters far more how
you fair on a technical interview than which school you went to. If you
blow someone away with your technical prowess, then we really don't care so
much what school you went to.
One guy I knew got his psychology degree then went off to cambridge for a
masters in CS. Another has a non-CS degree followed up with a masters in CS
from Berkeley. Another linguistics, then Ph.D from Cornell. There's more,
but omitting due to the ease to which someone could fingerprint the info
(and I don't have the permission of the individuals involved).
An additional reality is that at your public research university, the
classes are ridiculasly huge. Even in my upper division operating systems
course, there were like ~100 undergrads. That's another reason to go grad
school or smaller university (where classes are smaller).



More information about the NFBCS mailing list