Hi guys, probably in other countries things are like here in Brazil, and in the end of the college degree in Computer Science you have to develop, document and present a final project in some area. I'm just about to finish the course and I haven't started the project yet, well, I did it a lot of times but I wouldn't decide about what to keep with, since it was not interesting at all. I have troubles to find a nice, interesting and emerging area in CS. I wouldn't like to make a project based on use of a language, like a portal, an e-commerce, etc, language must be the tool. But then I fall in this question, which areas in CS are emerging in late years? I'd like to hear some suggestions! Thanks.
+2
A:
An emerging area is ontology language development and practice. An ontology is a language that represents a web of concepts organized into a series of relationships where those relationships may even be hierarchical. The objective is to identify and represent context of language or expression so that a communicate can interpret and respond to human communication as a human might.
+2
A:
Cloud computing is hot right now and still largely undefined, unproven.
Geoff
2009-08-27 11:00:04
is cloud computing a CS topic?
SilentGhost
2009-08-27 11:00:46
isn't it?
Geoff
2009-08-27 11:08:03
is it?
SilentGhost
2009-08-27 11:12:16
Hmm. While many aspects of it are obviously CS-related, my impression is that cloud computing is *hot* mainly in the technology industry; business people, CIOs, etc. At least much more so than in the academia.
Jonik
2009-08-27 11:42:02
I see your point, but I kind of think that is why it's interesting. When I got my Masters, I wrote how corporations could take advantage of all their networked desktops to perform massively parallel operations like seti@home. CS isn't just about theory, its can be about real-world business uses and technologies.
Geoff
2009-08-27 13:09:06
Cloud computing is a term aimed more at the implementation imho, as the technology behind it worth CS research is usually the same as networking-problems, balancing, speed, accessibility and so on. And the rest is probably engineering. So it's more of an "application" than a CS research area, imho.
Etamar L.
2010-04-07 15:41:21
+5
A:
What about quantum computing?
http://en.wikipedia.org/wiki/Quantum%5Fcomputer
A prof at my old university was researching this area pretty heavily. Might be a tad advanced but definitely interesting.
Zoidberg
2009-08-27 11:04:21
I think it would be a little hard for me to make tests and develop something for it in a graduation thesis with few people (by few, you can read two), but for sure it's a very interesting area
Alaor
2009-08-28 02:33:11
LOL too true, good luck on finding something interesting. My graduation thesis was on regression testing and tools that may help select automated test cases. It wasn't OVERLY exciting but practical none the less, and sometimes thats just as exciting... at least for me anyways.Good luck!
Zoidberg
2009-08-28 10:26:35
+3
A:
Genetic algorithms continue to be big IMHO. They are fairly easy to implement (though tweaking the fitness function can be a long and arduous process), and can be applied to nearly any project. I know of cases where its been used to predict the stock market and also to find the best solutions for manufacturing processes.
espais
2009-08-27 11:21:38
+14
A:
The death of the RDBMS and what will supersede it.
e.g
1. map-reduce
2. key/value store
3. Document databases
4. Column Databases
John Nolan
2009-08-27 11:23:46
Really, why should people in the 21th century take care of mapping beautiful object hierarchies into stupid tables??? (Ah, and of course, mapping them back again...)
ivan_ivanovich_ivanoff
2009-08-27 11:58:44
Nice suggestions, even more because I really like working with data/rdbms, I tried some oodbms and it's clear to me that this area will increase in the future
Alaor
2009-08-28 02:31:50
+3
A:
I have this awful feeling that a lot of people are going to confuse computer science with computer technology
Remember the quote of Edsger Dijkstra: "Computer science is no more about computers than astronomy is about telescopes."
John Smith
2009-08-27 11:27:39
I dare say though that all advances in telescope technology has been done by astronomers who want more than they currently have.
Geoff
2009-08-27 13:11:42
I disagree. Some telescope advances have actually been from camera and lens makers.There is also a slight flaw in the quote which I hadn't seen before. The telescope is almost exclusively used by astronomers, whereas the computer is used by everyone.
John Smith
2009-08-27 14:36:35
Not when Dijkstra said it, though. At that time, the thought of everyone using a computer was completely ridiculous. Heck, not even every *country* could afford one. (Actually, for about one third of the population of this planet, the thought of using a computer, is probably *still* ridiculous. Remember: the world bank defines being "poor" as earning less than 1.25$ a day, and according to that definition, more than 1 billion people are poor. For them, buying even a 100$ laptop means no food, water, shelter, clothes, medical supplies for more than 2 months!)
Jörg W Mittag
2009-08-29 12:50:25
1) Cell phones are computers; 2) In the so-called "third world" use of old, refurbished cell phones is growing for disparate applications (for example, to transfer money - see the work of Jan Chipchase, anthropologist at Nokia research).
MaD70
2009-11-02 17:01:27
+7
A:
Anything that makes concurrent programming easier.
While concurrent programming as such has been researched and understood for decades (i.e. the theory is there), practical application has lagged behind a lot until recently, because single-core CPUs were the norm.
Only recently have multi-core CPUs become widespread, and languages like Erlang that support concurrent programming without the headaches caused by shared memory have lost their niche status and come into the limelight.
Michael Borgwardt
2009-08-27 11:55:39
+2
A:
Machine Learning algorithms. There are quite a few unsolved problems in computer science/bioinformatics/etc just waiting to be solved using them. Getting a computer to recognize patterns, or even generating data from what it was trained on is a powerful concept which has a wide range of applications.
Bill Casarin
2009-08-27 12:14:23
+1
A:
Automated testing methods are an interesting topic.
There are some emerging companies experimenting with model based testing. They say it's the cat's whiskers, I say it's the cat's turd in the lawn.
The truth will be somewhere in the middle I presume, though nobody really knows right now where in the middle.
NomeN
2009-08-31 23:27:57