Hello, Yes, I realize it's mid-February right now and the summer is still months away, but it's probably not too early to think about the future. I am wondering if there are any Summer of Code projects that I would be able to do for the Haskell community. I will be graduating from my undergrad program this semester and am hoping to go on to graduate school in the fall, and I think a good Haskell project would be the perfect way to spend my transitional summer. I looked at http://hackage.haskell.org/trac/summer-of-code/query?status=new&status=assigned&status=reopened&group=topic&type=proposed-project&order=priority to see if there was anything that people already had in mind, but that page looks old and unkempt (judging by the abundance of advertisements) so I am not sure which of those are still available or needed. So, are there any projects that will need a student for the summer? If this is not the place to ask, where should I be asking? Bryan Burgers
bryan.burgers:
Hello,
Yes, I realize it's mid-February right now and the summer is still months away, but it's probably not too early to think about the future.
I am wondering if there are any Summer of Code projects that I would be able to do for the Haskell community. I will be graduating from my undergrad program this semester and am hoping to go on to graduate school in the fall, and I think a good Haskell project would be the perfect way to spend my transitional summer.
So, are there any projects that will need a student for the summer? If this is not the place to ask, where should I be asking?
Here's some general advice for those considering applying for the Summer of Code Haskell projects. We haven't prepared a list yet of desirable features, but going on last year's successful applicants, the following qualities will be required: 1. The project should have: * clear benefit to the Haskell community * there will be an emphasis on new libraries and on new development tools. We really prefer people work on things that the community needs: libraries and development tools. Some suggestions would be the http lib, a binary/bitstream parsec, or a light web framework (that works out of the box), ghc-api integration with emacs/vim. 2. The applicant should have: * demonstrated experience with Haskell programming * demonstrated experience working alone, (e.g. open source experience is a great asset) With over 100 applicants, and 9 positions available, competition is tough. You can stand out by contributing to projects *now* so that you gain experience with the current best Haskell development practices, and get some guidance from the open source hackers in the community about how to work efficiently and with community consensus. Start working now on your own library or app, get it in darcs, publish it, upload it to hackage, and when summer comes around, you'll have massively increased your changes of being selected. -- Don
Hello Bryan, Tuesday, February 13, 2007, 2:24:21 AM, you wrote:
I am wondering if there are any Summer of Code projects that I would be able to do for the Haskell community.
of 9 projects started last year, only 1 or 2 was successful. so i think that retaking one of projects selected last year may be a good idea to Donald: afair, you was written some report about this SoC. can you please explain status quo of projects started last year? is it possible to restart unsuccessful ones again? to all: i think, now it's the good time to start thinking about projects we need. last year one-week schedule was too fast i'm completely agree with Donald in that last year some participants was Haskell novices and this paid project was the first program they ever tried to write in Haskell :) -- Best regards, Bulat mailto:Bulat.Ziganshin@gmail.com
bulat.ziganshin:
Hello Bryan,
Tuesday, February 13, 2007, 2:24:21 AM, you wrote:
I am wondering if there are any Summer of Code projects that I would be able to do for the Haskell community.
of 9 projects started last year, only 1 or 2 was successful. so i think that retaking one of projects selected last year may be a good idea
7 were successful, 2 were unsuccessful. This was above average for Google SoC (they originally planned for more than 50% failure) and we can be considered very succesful! Some other .orgs had dramatically worse rates.
to Donald: afair, you was written some report about this SoC. can you please explain status quo of projects started last year? is it possible to restart unsuccessful ones again?
Certainly.
to all: i think, now it's the good time to start thinking about projects we need. last year one-week schedule was too fast
i'm completely agree with Donald in that last year some participants was Haskell novices and this paid project was the first program they ever tried to write in Haskell :)
As far as I remember, all of the participants had previous haskell knowledge. The point of the SoC though is to bring *new* hackers into the community, who hopefully become core contributors. We had several SoC guys at the recent hackathon, who wouldn't have been there if not for the SoC. You see SoC participants helping out in all sorts of ways in the community now. So it really worked rather well. I'm looking forward to another crop of hackers for this years projects! -- Don
bulat.ziganshin:
Hello Donald,
Wednesday, February 14, 2007, 1:51:26 AM, you wrote:
7 were successful, 2 were unsuccessful.
can you please name successful projects and their download pages?
Most of the projects have been hosted on darcs.haskell.org since the beginning, and students involved in the SoC continue to help out in the community. Here's a quick summary: 1. Project: GHCi based debugger for Haskell Jos� Iborra L�pez Now part of GHC head. Jos� was at the recent Hackathon in Oxford, and the debugger continues to be developed in GHC. He continues to be active in the community and #haskell. 2. Project: haskellnet Jun Mukai Completed, available from http://darcs.haskell.org/SoC/haskellnet/ Recently published a Haskell textbook in Japanese. Can be found on #haskell. 3. Project: Storable a => ByteString a Spencer Janssen Completed, available from http://darcs.haskell.org/SoC/fps-soc/ Spencer's now an active member of the community, and currently working on a new window manager written in Haskell 4. Project: Fast Mutable Collection Types for Haskell Caio Marcelo de Oliveira Filho Completed, and part of pugs. http://perlcabal.org/~audreyt/darcs/pugs/third-party/HsJudy/ 5. Project: A model for client-side scripts with HSP Joel Bj�rnson Completed, available from http://darcs.haskell.org/SoC/hsp.clientside/ 6. Project: Thin out cabal-get and integrate in GHC Paolo Martini Completed, part of Cabal (http://haskell.org/cabal), and the rapidly progressing hackage.haskell.org 7. Project: Port Haddock to use GHC David Waern Completed, and still active, http://darcs.haskell.org/SoC/ghc.haddock/ David also made it to the Hackathon, and is a #haskell regular These last two projects were not completed: 8. Project: Implement a better type-checker for yhc. 9. Project: Implement a parser for Language.C Further analysis and improvements to implement for the upcoming SoC will appear on the SoC website in due course, before the start of this year's SoC. -- Don
This project may be a success:
4. Project: Fast Mutable Collection Types for Haskell Caio Marcelo de Oliveira Filho
Completed, and part of pugs. http://perlcabal.org/~audreyt/darcs/pugs/third-party/HsJudy/
But there is no way to pull this out of pugs and install it as a library. The cabal file depends on being alongside http://perlcabal.org/~audreyt/darcs/pugs/third-party/judy/Judy-1.0.3 and while I can build and install HsJudy this way, I cannot clearly see how to install the Judy libraries it needs. If anyone *can* make HsJudy install and work, could you put this information on the haskell wiki? -- Chris
haskell:
This project may be a success:
4. Project: Fast Mutable Collection Types for Haskell Caio Marcelo de Oliveira Filho
Completed, and part of pugs. http://perlcabal.org/~audreyt/darcs/pugs/third-party/HsJudy/
But there is no way to pull this out of pugs and install it as a library.
The cabal file depends on being alongside http://perlcabal.org/~audreyt/darcs/pugs/third-party/judy/Judy-1.0.3 and while I can build and install HsJudy this way, I cannot clearly see how to install the Judy libraries it needs.
If anyone *can* make HsJudy install and work, could you put this information on the haskell wiki?
-- Chris
I'd just ping the auhtor, host the repo on darcs.haskell.org, and then fix it until it builds like any normal cabalised repo. It really could be a Data.HashTable killer, given a bit of build infrastructure love :-) -- Don
Donald Bruce Stewart wrote:
If anyone *can* make HsJudy install and work, could you put this information on the haskell wiki?
-- Chris
I'd just ping the auhtor, host the repo on darcs.haskell.org, and then fix it until it builds like any normal cabalised repo.
It really could be a Data.HashTable killer, given a bit of build infrastructure love :-)
-- Don
I kludged an install that makes it work in ghci now. But I don't have a tested recipe for this yet. Even downloading from the pugs repository was tricky. I could not see how to use darcs to get just HsJudy (when I tried to do so darcs was grabbing > 3k patches (using --partial!)). So use "wget --no-parent -r", which is hardly a step I could put in a general recipe. -- Chris
Google just announced the 2007 SoC
http://code.google.com/soc/
On 2/15/07, Chris Kuklewicz
Donald Bruce Stewart wrote:
If anyone *can* make HsJudy install and work, could you put this information on the haskell wiki?
-- Chris
I'd just ping the auhtor, host the repo on darcs.haskell.org, and then fix it until it builds like any normal cabalised repo.
It really could be a Data.HashTable killer, given a bit of build infrastructure love :-)
-- Don
I kludged an install that makes it work in ghci now. But I don't have a tested recipe for this yet.
Even downloading from the pugs repository was tricky. I could not see how to use darcs to get just HsJudy (when I tried to do so darcs was grabbing > 3k patches (using --partial!)). So use "wget --no-parent -r", which is hardly a step I could put in a general recipe.
-- Chris _______________________________________________ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
participants (5)
-
Bryan Burgers -
Bulat Ziganshin -
Chris Kuklewicz -
dons＠cse.unsw.edu.au -
Mathew Mills