HN Explorer

Ask HN: Best MS CompSci thesis work you have seen

1 points by FraaJad ↗ HN
I am starting MS in Computer science in Spring 2012. While MS is considered to be preparation for a PhD, there may be some thesis work which stand out on their own in terms of original contribution.

Please share the work that you consider important, exciting etc.,

3 comments

[ 2.3 ms ] story [ 13.7 ms ] thread
iqster
Probably BumpTop by Anand Agarawala @ UofT. It was great research (got a CHI paper), got lots of public attention (over a million people saw the YouTube video), and was the basis for a successful company.

That said, please don't try to aim for a brilliant M.Sc. thesis ... many grad students have fallen into that trap. You need to just do stuff. Keep reading papers, making prototypes and submit papers to conferences and workshops. Best suggestion I can give ... find an existing research topic (worked on by great PhD students and faculty), and contribute a well-defined but small part. You will get authorship in a good publication, and learn how to do research. Do this in the first 6-months to 1-year. This will actually help you when you write your own thesis (which can certainly be independent.)

P.S. I assume you are in Canada, eh?

ColinWright
Now that HN doesn't show upvotes, I'm largely compelled to add a comment endorsing this reply.

Yes, some people have written brilliant MSc theses. Yes, you might be one of them. But don't make that your aim.

By all means try to write a really good thesis, but your deadline is much, much more important. Work hard, produce something solid, contribute to other people's work, and that will give you a much better launching pad. It may be that your thesis ends up being brilliant, but if you make that your target, the overwhelming odds are that you will crash and burn.

ObHN: Most of the people who started huge companies did not have that as their target, it happened as a consquence of hard work, a good idea, a bit of luck, and some undefinable magic fairy dust.

FraaJad
Thanks for your reply. I'll look up the paper. I was away in class, so couldn't reply :p

Actually, I'm in the Midwestern USA.

I did talk to my professor about this today and asked him to show some his students' best MS works. That gave some idea about the depth and rigour of work required to be considered "good".

I think your idea of working with a great Ph.D student is very interesting and I will run this idea by my professor.