[freenet-dev] GSoC 2012

[Mohammad Hoda]

Thanks Michael and Matthew
I ll try doing tha. I ll probably make a new thread and seek help when
ever I am stuck

On 18/02/2012, Ian Clarke  wrote:
> Apparently I should have read the entire thread before replying, as Toad
> had already persuaded Nextgens, doh!
>
> Ian.
>
> On Fri, Feb 17, 2012 at 2:34 PM, Ian Clarke  wrote:
>
>> On Thu, Feb 16, 2012 at 5:26 PM, Florent Daigniere <
>> nextgens at freenetproject.org> wrote:
>>
>>> Different codebase
>>
>>
>> Same language, many architectural similarities (use of small-world
>> routing, UDP messaging, UDP-hole-punching, etc)
>>
>>
>>> arguably no active developpement (7months since last commit? -
>>> https://github.com/sanity/tahrir/commits/master), no release yet, no
>>> userbase, ...
>>
>>
>> Exactly why I want to get someone to work on it!
>>
>>
>>> different goals
>>>
>>
>> Different non-competitive approaches to achieving the same broad goal.
>>
>>
>>> I am not sure it's a good fit for GSoC tbh... but then again, it's up to
>>> Google to judge that.
>>>
>>
>> I don't see why not, with appropriate mentorship, which I'm willing to
>> provide, I think it could be a great way for a student to learn about P2P
>> architectures, and without having to learn a vast codebase.  They'd get
>> involved early enough that they could make a real difference, that is a
>> feature, not a bug.
>>
>>
>>> Why do you think Freenet would be the right umbrella?
>>
>>
>> Because it has a 501c3 and has a good track-record with GSoC.
>>
>>
>>> Is there any integration in between both planned on the roadmap?
>>>
>>
>> None planned, but it's possible.
>>
>>
>>> Why not presenting it as a mentoring organization on its own?
>>
>>
>> Because it would be a lot more work, with no advantage I can think of.
>>
>> What is the disadvantage of doing it under the Freenet umbrella?
>>
>> Ian.
>>
>> --
>> Ian Clarke
>> Personal blog: http://blog.locut.us/
>>
>
>
>
> --
> Ian Clarke
> Personal blog: http://blog.locut.us/
>


-- 
Mohammad Jamilish Shiyamul Hoda
Computer Science Engineer
Mobile:+91 829 624 0102

[freenet-dev] Heavy WebOfTrust optimization almost done

[xor]

Ok first benchmark result available for my heavily fragmented WOT main 
database:

The full score recomputation which it does at startup took 373 seconds with 
the old code and 95s with the new code. Thats down by 75%, or in other words 
the old code was 4x as slow.

Of course this is only a single measurement, but we are talking about minutes 
here so its unlikely that things like system load of the machine caused much 
bias.

Also, 95 seconds is still very much. I suspect it will go down by a huge 
factor when I implement defragmentation (tomorrow hopefully):
The database currently has a size of 312 MiB, which shows that it is really  
fragmented very much.