Ideas for 3rd year project (CS and SE)

[unoho]

Hi everyone, glad to be back on this forum again.
I am going in to 3B term next semester and I see that I have a course related to design project. This project will be a continuation until 4th year and would be used as a final project.
I am kind of lost into what type of things would be a good choice. I am currently in Soft Eng. I have thought about making games or doing open source projects but few seniors told me that games/open source projects are usually frown upon. Has anyone ever experienced/dealt with this type of situation?
Also, I have done some brainstorming but whatever idea I come up with has already been made or the scope isn't big enough to be used as a final project.
So my question to you good folks would be what are some projects that you or your friends worked on? Any insights into how you got there/ protips for doing a 3 term long project etc. would be greatly appreciated!

[btiffin]

This is from an old guy, with old guy belief systems...

My favourite thesis's (thesesees?) to read when surfing the internet are always programming language based. (It's also my interest and bias).

As a suggestion, work out some nitty gritty details for adding security to stack frames. Narrow the spec to optimized loops. Or, go big, and develop something with an angle to it. Elements (variables) as pipes, part of the lexical specification. Or a VM that is actually based on a reduced instruction set that simply gets assembled for each target from macros so the VM is actually not, and code can run free and native again. Etc.

Poke around with some of the systems that tenured professors are up to. Icon Unicon from Arizona, or Pure (Haskell based) from Mainz in Germany or (well, there are lots, and most are hard to find).

Or, suck up to corporate thinking; detail some ways of extracting pennies from online activities by allowing people to buy extra tokens for Cloud boosts.

Or, do the world a bigger favour. Programmers still need a good garbage collector, well at least, a gooder garbager collecter.

Or, the lexical shenanigans to make controlling multi core processors in source code, readable by IQ normal programmers, a viable thing in a programming language.

Or, waste time on a toyol. Toy/Tool/Toil, maybe something to protect revision control systems from IQ normal programmers, errr, I mean to protect IQ normal programmers from the convoluted IQ bigger than normal developed systems we get to use everyday.

Mostly though, the biggest part. Love it, or leave it for another try. There is so much fun to be had.

Cheers

[unoho]

Thanks for the ideas, btiffin.

I think the biggest challenge like any projects would be where to start. As I don't have any idea where to start any of those above. But that's where the fun begins

[btiffin]

The cloudboost idea is likely a good one.

Dan, Tony. Get on it. Seriously.

No, really. Seriously. Play into peoples impatience ala Tapped Out and all the magic glass tablet tokens for dollars craze.

A turbo button for the cloud, and it'll only cost $1.99 to run that next job faster. Guaranteed, or your money back.

code:

$# start all processes on an equal footing $ nice 15 $ if yep-gotabuck; then nice 10; fi $ if nope-they-said-no; then nice 20; fi  # cheap bastards

Seriously. Get on this one. Perhaps a lot less facetiously implemented then the draft above, but I do think it could fund a fair sized enterprise. Start it up and you'll have my resume, but if you do, and I pass the interviews, I'd like a corner office, with a great big sign.

This is the guy that could have owned this building



Cheers