Re: [rvm-research] GSoC 2012

[Erik B. <eri...@gm...>]

Hi,

On 20.03.2012 04:33, chi...@g2... wrote:
> I want to contribute to Jikes RVM and participate GSoC this year.

That's cool, we could really use more contributors :-).

> I am highly interested in academically-oriented project which maylead to a publication in a workshop/conference.

IMHO it would be more beneficial for the Jikes RVM project if GSoC 
students worked on non-research-related projects (e.g. missing features 
compared to commercial JVMs, infrastructure, bigger refactorings or bug 
fixes, ...). Those kind of contributions are less likely to come from 
the Jikes RVM community because they are not related to research.

However, I'm not involved in the selection of students so you don't need 
to listen to me ;-).

> My research interest is tool-chain and run time system. [...] Would you mind to give me some advice about project selection?

I'm not sure that I really understand what your interests are so I can't 
give any specific advice.

I'll try to give some general advice because I didn't see any replies on 
the list:

1. In case you have not already done so, you can look at GSoC proposals 
from previous years at 
http://docs.codehaus.org/display/RVM/Google+Summer+of+Code .

2. You can also look at the bugtracker at 
http://jira.codehaus.org/browse/RVM if there are any issues (especially 
of type "new feature" or "enhancement") that are interesting for you.

3. Depending on what you mean by tool-chain, it might be interesting to 
take a look at bootimage building. For example, the list of classes in 
the bootimage ( 
http://docs.codehaus.org/display/RVM/Primordial+Class+List ) is 
currently hand-crafted. One could write a tool that would automatically 
discover which classes are necessary for the bootimage.

4. If you propose an idea not listed on the current GSoC page, it is 
necessary to discuss your idea before submitting your final application 
to ensure that the scope is correct for GSoC.


Kind regards,

Erik Brangs