On Fri, 02 Jun 2006 01:42:31 -0400, Sander Hoentjen <[EMAIL PROTECTED]> wrote:
> On Thu, 2006-06-01 at 12:30 -0400, Youness Alaoui wrote: >> btw, I forgot to say... >> I just saw the wiki, it's ok for the first (MSNP13), but I think it's >> too >> little work for a full time 'employee' during 3 months.. > I don't think so.. when I mean msnp13 i mean full msnp13, and that is > one huge ass job if you ask me (yes it includes shared folders). It also > includes essentially rewriting the whole protocol handling, getting a > more unified way of handling messages (like file tranfer, DP transfer, > etc.) > oups, sorry, in that case yes, of course, I thought you just meant "finish MSNP13 in amsn"... but if it's the whole protocol, yes, it's a huge ass job... but I don't think he'll have time for shared folders, etc... only for implementing the basics, probably with MSNP2P too... if that's the case, it's a C implementation of the protocol as the MSN connection manager to telepathy... >> replace it by >> libmsn (we'll need it anyways since we move to telepathy for amsn2) >> For the XML2GUI, I don't think it's a good idea, maybe part of his >> assignment could be "find solutions for a xml2gui" because it seems >> there >> are already some exisint, look at tcl wiki page, I think under GUI >> builders... There is one even named "xml2gui" so... the student could >> do a >> research job into finding, testing xml2gui projects, then choose one >> that >> suits us, and probably modify it to suit our needs even more (I don't >> know >> how stable/complete the current implementations are, it would be his job >> to find that out) > So is xml2gui a good idea or not, not sure what you think according to > above text :) > yeah, sorry, I'm a bit confused myself.. I said that it's not worth it writing that, but if the student has time, his job can be a research job instead of an implementation job.. but Harry just said he'd do that.. maybe we'll find a solution/alternative before we finish... also, it will heavily depend on our final design, which is really far from being complete, we just have some ideas so far... >> Other project would be to build the other stuff from amsn2. the data >> structures, the telepathy thing, the farsight thing, the gui objects, >> the >> wrappers, the dbus extension, etc... > yes i was not sure on those project, IIRC you had a lot done on the > farsight stuff and jonne on the dbus extension. yes, farsight... almost done porting the code to the new API, but I still need to work on it (where is the time for that), and I also have the huge problem of : porting the code is not all, we did a LOT of work on the protocol stuff in msncam.tcl, none of it is done in the farsight plugin, doesn't support pausing yet for example, nor the reflector, it's really basic webcam support, it's far from being as stable and complete as the current msncam.tcl implementation... >> we need to review all that and create a real design, maybe with the >> student's help. > So you are saying: give the student what we have, let him create what we > don't have yet and make everything organized and clean? no, I'm saying that the design is far from being complete and we should have every tiny little detail covered before doing anything... a specific requirement spec file with detailed on what is needed... time estimates for each small part of the project, etc... once we got all that, then we're ready to say what we want and what should be done, then we go to the implementation part... we'll do the desing, but I just said that the student could help... >> Did you get students to apply ? that question about python, who asked >> it ? >> do you know how many students we can get ? > Students can apply to proposals, not to projects if i understand > correctly, that is why we need to have proposals. Also there aren't many > sponsors for the project so not a lot of students can participate, I > think we should be glad if we got one. Anyway we can still keep > proposals if they don't get done. Throughout the year we can remove/add > others and next year we will apply to both google's SoC and this one > again, etc :P ok, you mean we can't get any student for now as long as we have no proposals ? ok then, try adding the basic known ones, implement MSN connection manager (MSNP13) for telepathy, implement webcam plugin for farsight and 'related work' (for integration in the msn connection manager and the webcam protocol part, and making the hash algorithm for reflector work on mac (big endian), etc...) we'll see about the remaining stuff, we need to review it I think before specifying anything (we can't have "create data structures" as a proposal, right ? :P) KKRT >> >> KKRT >> >> On Thu, 01 Jun 2006 07:49:57 -0400, Sander Hoentjen <[EMAIL PROTECTED]> >> wrote: >> >> > http://amsn.sourceforge.net/wiki/tiki-index.php?page=SummerOfCode >> > here i listed 2 proposals already, should we add more? >> > also i got this question: "I was wondering if porting amsn to python >> to >> > create a kind of libamsn(2) would be accepted as a project" >> > What are your ideas about that? On one side I don't see a benefit in >> > doing that, but on the other hand if we can have a student do it (and >> be >> > ready in 3 months..) i don't see a downside (other then that it will >> be >> > in place of another project, that we *might* want more. >> >> >> -- KaKaRoTo _______________________________________________ Amsn-devel mailing list Amsn-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/amsn-devel
