Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Is it just me or does anyone else always feels overwhelmed reading the project suggestions? I find the projects listed to be extremely challenging. Project suggestions - http://www.scs.stanford.edu/11au-cs240h/labs/project.html


I was amused at these suggestions:

> You could implement a 'bot that executes untrusted code along the lines of Lambdabot. Lambdabot was written before Haskell was typesafe, and as a result is fairly hairy code and very limited in functionality. Using type safety you ought to be able to achieve something much cleaner and simpler, and that provides far more functionality. > Implement some sort of end-user programmable web-site (like Wikipedia, but for code rather than for information). > Implement an online Haskell REPL environment such as Try Haskll. Using Safe Haskell and some of the recent support in GHC 7.3 for data declarations in GHCi you should be able to provide a far more powerful feature set then existing websites.

Try that, and you'll discover that much of the complexity in lambdabot/mueval/tryhaskell-mueval-fork is due to stuff that Safe Haskell won't help with! eg. good luck dealing with 'let x = x + 1 in x'


Some are easy, some are research level. They're all "open source"-style programming : i.e. add real features to real projects on a real time frame.

You could expect to apply to do these in the Summer of Code Haskell projects.

Challenging, yes, but not unrealistic for competent systems / language people.


Eh, yes. Please tell me that, guessing from the course's name, this isn't a 2nd year course in some degree?




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: