My name is William Panlener, and I am currently attending the University of Mississippi as a first semester Masters student in Computer Science. This is my first time participating in GSoC.
My proposal for GSoC involves choosing an application with high dependence on native dlls and implementing enough stubs so as to eliminate this dependency. I chose the game, League of Legends, and its installer, Pando Media Booster. Most stubs that require implementation are in msvcp90 and wininet.
I have two concerns with this proposal: 1) Is it unusual to work towards improving a specific application rather than focusing on a specific component of wine? 2) The scope of the project is difficult to gauge in advance since the next problematic stub cannot be identified until all the previous problematic stubs are taken care of. Is this an issue for a GSoC proposal?
I would be more than willing to revise my proposal (if this is allowed) if either of the above points is a problem.
On 14 April 2012 05:03, William Panlener wpanlener@gmail.com wrote:
I have two concerns with this proposal:
- Is it unusual to work towards improving a specific application rather
than focusing on a specific component of wine?
In general Wine development it's pretty common. For GSoC it's a bit more uncommon, but I don't think that necessarily makes it bad.
- The scope of the project is difficult to gauge in advance since the next
problematic stub cannot be identified until all the previous problematic stubs are taken care of. Is this an issue for a GSoC proposal?
Somewhat. We generally like GSoC projects to be fairly well defined tasks, so that it's clear up front how much work it's approximately going to be and when it's finished. You could probably make the proposal a bit more concrete by taking a look at the application's import table and checking which of those imports are unimplemented.
However, while I'm not that closely involved with GSoC, I was under the impression that the deadline for submitting proposals has already passed.