[Battlemesh] No results again

Pau pau at dabax.net
Mon May 9 16:34:29 UTC 2016



On 09/05/16 18:02, nemesis wrote:
> At the moment I have the impression that everybody is welcome to run
> tests as long as they use wibed, and if they don't, they get attacked.

I don't understand where did you get this idea. I think it is clear that
the team responsible for the testbed will choose which solution to use
every year. It has been always like this.


> So I do not really understand your email Christian, do you mean that if
> next year we want to participate to the battlemesh we have to use wibed?
> Or are we free to run our own tests without being forced to use a system
> which many people, who for some reason are not speaking out, find it
> uncomfortable using?
> 
> It is clear to me that now there are two groups of people: those who
> want to run wibed and those who would prefer trying something different.

I don't think we are talking about this. There have been some comments
in this thread blaming Wibed and putting the failure focus on it. Which
I think it is not fair, even more when the person has not been even part
of the event. No one is pushing the usage of Wibed for the next year.

There have been many efforts put on this firmware from many people
during several years. I'm one of those, however I'm quite calm even
after reading all the critics, because I know it is mainly coming from
people who do not know much about it and has not contributed a single
line of code to it.

Wibed is much more than a firmware for the battlemesh, so it is much
more complex than a one-time-usage firmware. If you like to know more
about what I'm talking about, please take a look on this document:

http://wiki.confine-project.eu/_media/wibed:wibed_2014_tfm_pau.pdf

So thanks to "this complexity" what we can do now, for instance, is to
create a small Wibed testbed in UPC (deploy it would not take more than
some hours now that the firmware is ready and installed in the routers).
And then leave the access to the central controller open to anyone who
would like to perform experiments. Let's see what people from UPC says
about this.

> We either let both participate and try different scenarios, hence
> getting even more results, because we can do more tests if we work well,
> or we should clarify that the tests need to be performed with a single
> system, like wibed, therefore those who don't find it comfortable to
> work with it will have to decide if coming just for the talks or not
> come at all.

Personally after reading some comments and reactions here, I don't care
at all if Wibed will be used or not in the next WBM. So from my side the
path is clean to anyone else who want to take the responsibility of
creating a new firmware. I've been doing this task in v6 and v7 (also
small contri butions in v9), and I'll step apart.

> Federico
> 
> 
> On Mon, 9 May 2016 17:53:08 +0200, Christian Huldt
> <christian at solvare.se> wrote:
>> +1
>>
>> Battle of the testbeds has to organize their own event IMHO.
>>
>> Battlemesh has always been a very welcoming environment, I think that is
>> one reason for the growth, please keep it that way...
>>
>> Den 2016-05-09 kl. 17:39, skrev Simon Wunderlich:
>>> Hi,
>>>
>>> on which authority is anyone expelling anyone here? I think nobody
>>> can and
>>> should get expelled on disagreements on the testbed! I think the
>>> discussion is
>>> getting a little unprofessional here, looking at Juliusz' statements,
>>> who
>>> wasn't participating WBM in Porto and various other events before.
>>> Please, let
>>> us keep to the facts, and don't mark anything "fundamentally flawed".
>>> I think
>>> that talking about problems and mistakes is very important to
>>> improve, but NOT
>>> in a way that anyone should feel expelled.
>>>
>>> Please, let us talk about the problems and solutions professionally
>>> here. Let
>>> us point out problems based on specific occurrences. And lets fix them.
>>>
>>> Personally I don't care which testbed system is used for next year,
>>> and its
>>> for the people who prepare and perform the tests to decide (which is
>>> not me).
>>> And personally I appreciate the effort of everyone helping to make
>>> the tests
>>> work, even if we don't have results every time, because even if we
>>> don't, we
>>> as the protocol developers get valuable results. For example, even
>>> this year
>>> we got various valuable bug reports which made us fix problems in
>>> BATMAN V. If
>>> I heard correctly, it was similar for other protocols as well.
>>>
>>> Lets stay positive here!
>>> Thanks,
>>>      Simon
>>>
>>>
>>> On Monday 09 May 2016 17:07:39 Roger Baig Viñas wrote:
>>>> Hi and goodbye,
>>>>
>>>> FORTUNATELY the evidences contradict the unfounded statements (the way
>>>> of doing science of some academics -together with overlooking the
>>>> facts which contradict their theories). In v6, inspired on v5's work,
>>>> we presented WiBed and, not only we  got test results, but we were
>>>> able to present them in figures systematically produced and a full
>>>> report was delivered afterwards [1].
>>>>
>>>> UNFORTUNATELY some individuals insist in expelling people from the WBM
>>>> while the community stays quite. Last year was Sven's turn, now it is
>>>> mine.
>>>>
>>>> [1] https://github.com/axn/wbm2pdf
>>>> [2] http://ml.ninux.org/pipermail/battlemesh/2015-August/003807.html
>>>>
>>>> My apologises to those who trust me again after last year's farce.
>>>>
>>>>
>>>> On 9 May 2016 at 15:48, Juliusz Chroboczek
>>>>
>>>> <jch at pps.univ-paris-diderot.fr> wrote:
>>>>>> You still hit the same problems as we had 10 years ago, it takes too
>>>>>> much time to get a decent testbed running.
>>>>> No, Benjamin, read Federico's mail again.
>>>>>
>>>>> Last year, most of the people working on the testbed tried to get
>>>>> wibed to
>>>>> work.  A small group of people (including Federico), some of which had
>>>>> never touched an OpenWRT router before, decided to work in parallel
>>>>> and
>>>>> build a simple testbed that we fully understood.  The small group
>>>>> got some
>>>>> very useful results; the wibed people got none.
>>>>>
>>>>> Let us please face it: the notion of a magical testbed that will solve
>>>>> world hunger is fundamentally flawed.  The basic idea behind wibed is
>>>>> fundamentally flawed.  Let's take this into account in the future
>>>>> -- let's
>>>>> limit ourselves to simple test frameworks that people actually
>>>>> understand.
> 
> _______________________________________________
> Battlemesh mailing list
> Battlemesh at ml.ninux.org
> http://ml.ninux.org/mailman/listinfo/battlemesh

-- 
./p4u

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 473 bytes
Desc: OpenPGP digital signature
URL: <http://ml.ninux.org/pipermail/battlemesh/attachments/20160509/fdd3d925/attachment.pgp>


More information about the Battlemesh mailing list