[GSoC] Proposal: make WebKit run on top of GNUstep

[Daniel Ferreira (theiostream)]

Hi there,

My name is Daniel Ferreira and I'd like to spend a Google Summer of
Code implementing the proposal to "port WebKit to GNUstep" (or, most
likely, a subset of it) as announced on the GNUstep ideas website, and
try to get the GNUstep browser issue closer to a solution. However,
aside from this thread

http://lists.gnu.org/archive/html/discuss-gnustep/2007-03/msg00314.html

and this more recent one:

https://lists.gnu.org/archive/html/discuss-gnustep/2015-06/msg00108.html

there is almost zero detail implementation-wise of how this goal would
be attained, and only architectural discussions about how to approach
the web browser problem. Despite being a little late to the party, I'd
like some help to be able to craft a decent proposal (or at least
figure out how to approach the problem).

I spent the last two days setting up a build environment for both
GNUstep and WebKit, and I hoped I would be able to grasp the dimension
of the effort by trying to build it with OSX as a target and see how
many linking errors I got when trying to build it on top of the
GNUstep libraries. However, it complains about the lack of frameworks
that I don't believe are even in the GNUstep roadmap (Accelerate,
AudioToolbox, IOKit), so I think I'm following a wrong approach here.

That said, I'd really appreciate an (updated) explanation of what are
the current obstacles to getting that port done, or what can I
do/build to get a better understanding of it :)

Also: despite the extremely short schedule, I'm totally up to writing
a small patch to GNUstep or something if you'd like some sort of proof
that I can go around the project.

About me: I'm a Brazilian student and I'll start college at Stanford
University this fall. I've done system modifications in the iOS
jailbreaking community ("tweaks") and some apps, and during that time
I could get some pretty decent experience with Core* Apple APIs
(although I really look forward to learn more about their internals
and how they could be implemented) as well as Objective-C. I've also
written a bunch of C/C++ code for my employer (a payment processing
company) and some on github.com/theiostream. Again, I'm really
interesting on how these internals are (or can be) implemented, and
that's why I got so interested in this project.

(I am sending this e-mail to both the summer-of-code and the
discuss-gnustep lists separately. I'm sorry if anyone gets this twice,
but I'm really unsure regarding in what channel this will reach
someone.)

Thanks, and hoping I can get some answers in time,
-- Daniel.