p2psp team mailing list archive

[Nikhil Ramesh <rnikhil.2014@xxxxxxxxx>]

Hi,
    My exams are almost done. Only two to go :)
I am currently working on a proposal. Can I put it on the mailing list ?.
Or should I mail you personally ?.
Can I do both implementation of ACS, LRS rules and implementation of
DBS(peer) as an webrtc application as one complete GSoc project.? I think I
can do them fairly easily within the given time period. Is this allowed ?.

Also, I put a working copy of the code on GitHub. I will frequently commit
my changes/prototype there. I will create a new folder for all this. Link:
https://github.com/rnikhil275/p2psp.
As planned before, once my exams end on 18th, I will start working on the
project/prototype. I will also get back to your questions on the same day.

Thanks a lot for a commenting in the code base. It really helped me.

Thank You




On Thu, Mar 12, 2015 at 1:59 PM, Vicente Gonzalez <
vicente.gonzalez.ruiz@xxxxxxxxx> wrote:

>
>
> 2015-03-12 8:28 GMT+01:00 Nikhil Ramesh <rnikhil.2014@xxxxxxxxx>:
>
>> Hi,
>>     Awesome- Moving code to Github.
>>
>
> :-)
>
>
>>
>> No. I didn't mean a single class. Of course I understand the fundamental
>> difference between these two rules. I was asking why they are different
>> projects in the list. (Difficulty being easy) ?.
>>
>
> Well, I must confess that when we wrote these ideas (in the GSoC ideas
> page), the Set of Rules and the code were not so developed (we have keep
> working in these aspects for a while) and yes, for that reason, at this
> moment, they could be easy to address. However, these are still some open
> issues that we should solve in both SoRs:
>
> 1. ACS should be improved because now: (1) peers must lost chunks in order
> to receive less chunks and chunks never should be lost, and (2) those peers
> that send more chunks are going to erase from their list of peers to the
> peers that send less chunks if the "max_chunk_loss" parameter is not
> defined properly. The behaviour of the ACS peers should not depend on this
> parameter.
>
> 2. LRS only solves the chunks lost massively. Could we do something in the
> case a peer lost only some chunks? (Please, remember that we must minimize
> the latency of the protocol).
>
>
>> Anyway I will get back to you tonight, if I have any doubts while reading
>> the codebase ?.
>>
>
> Of course.
>
> Regards.
> Vi.
>
> --
> Vicente González Ruiz
> Depto de Informática
> Escuela Técnica Superior de Ingeniería
> Universidad de Almería
>
> Carretera Sacramento S/N
> 04120, La Cañada de San Urbano
> Almería, España
>
> e-mail: vruiz@xxxxxx
> http://www.ual.es/~vruiz
> tel: +34 950 015711
> fax: +34 950 015486
>