[ooni-dev] Cooperation between OONI & RIPE Atlas

[Vesna Manojlovic]

Hi Arturo, Naif, Virgil, all,

thanks for your interest, and for reaching out!

I'll reply inline.

On 12-sep.-15 12:55, Arturo Filastò wrote:
> Hi Virgil,
>
> Thanks for reaching out about this subject.
>
> I have spoken to Vesna (that I add to the cc of this thread in case
> she wants to chime in too) that runs the ripe atlas program in
> various occasions as to how we can in some way collaborate or join
> our efforts. I think the last consensus we reached is that our
> efforts are both very important, though complementary.

Indeed, and I try to promote OONI whenever someone asks "can we use RIPE 
Atlas for censorship measurements" -- I tell them to use OONI instead ;-)

> We are always open to seeing if there are some ways to collaborate
> more in the future, but from what I understand the policy and
> governance model of ripe does not allow for the sort of data
> collection we are currently mainly focussing on as OONI.

Until few weeks ago RIPE Atlas was not allowing regular users to do HTTP 
measurements. That's going to change, *but* probably not sufficient 
enough for you needs -- it is now only possible to do HTTP measurements 
towards RIPE Atlas anchors.

One of the reasons for this is ethical: protecting the safety of our 
hosts, but not allowing their probes to perform HTTP connections to 
random web sites requested by any other RIPE Atlas user.

Here's a paper that describes different use cases, benefits and 
limitations of multiple platforms, and addresses this ethical issue:
https://t.co/GKzQ0aZ1Te

However, what *is* possible with RIPE Atlas is SSL/TLS measurements. We 
have 8600+ probes that could be used for this, and if any of you find 
this interesting, I 'd like to have a discussion about the possible use 
case, or any details you need from me.

> That said we would be delighted to have them attend the OONI ADINA15
> hackathon at the Italian parliament:
> https://ooni.torproject.org/adina15/

I will send the info about this hackathon to my colleagues who are 
developers, and if they are interested they might apply -- and they cna 
bring the experience of our own platform & visualizations with them, 
plus they can learn from this hackathon about your solutions & challenges.

We also had a hackathon in March; all contributions are on GitHub:
https://github.com/RIPE-Atlas-Community/ripe-atlas-community-contrib/blob/master/README.md

& more tools by the community are in that repository too:
https://github.com/RIPE-Atlas-Community/ripe-atlas-community-contrib

We'll have another hackathon this year, I can let you know when&where.

 > Virgil: What particular direction of collaboration did you envision?

>> On 12 Sep 2015, at 09:18, Virgil Griffith <i at virgil.gr> wrote:
>>
>> http://atlas.ripe.net/
>>
>> The idea of putting their probes (or even anchors) next to Tor
>> relays and/or OONI probes seems an immensely good idea because
>> doing this will yoke the interests of generic network
>> infrastructure people with those of Tor and OONI.

That's certainly one way of "cooperating".

We can also somehow do it via MLabs,  alhou our cooperation with them is 
very slow -- two large bureaucracies add lots of inertia ;-)


>> I'd be happy to bring them to the Tor meeting in Berlin, but they
>> seem a better fit for the OONI meeting in Italy.

For the hackathon, you may want to make it easier and only combine 
existing RIPE Atlas data with OONI data.

We have APIs that are documented; our data is open, and historical.

We do constant, global pings, traceroutes & DNS lookups from _all_ the 
probes to _all_ root-namesservers; and "regional" measurements towards 
anchors. There are 143 anchors right now, and two of them are in Italy 
(17 in Germany, 5 of them in Frankfurt ;-)

Plus there are about 5000 measurements at any moment started-up by 
users, to the targets of their own choice; these measurements can be 
from 5 probes or form 500 probes, and can be one-time or periodic, 
lasting for several months.

>> Tell them to bring extra probes/anchors to give out to OONI
>> attendees.

If one of my colleagues comes, he can bring some probes.

However, hosting a probe is NOT the requirement for taking part in RIPE 
Atlas; it's a good start, and we could use more probes -- althou not in 
Italy ;-) We are currently trying to focus on "topological diversity", 
that is -- covering as many ASNs with probes.

On the other hand, we need people to *USE* that data!!

>> If for whatever reason they can't come to the Italy meeting I'll
>> invite them to the Berlin meeting.  Not as good a fit, but better
>> than nothing.

When is the "Berlin meeting"?

We also have a conference coming up - RIPE71 - in Bucharest, where one 
of you could come to talk to network operators about your needs & your 
results: http://ripe71.ripe.net

More things to share:

- training material about RIPE Atlas:
https://github.com/RIPE-Atlas-Community/syllabus

- I just saw the article about measuring censorship, that mentions OONI:
http://tma-2015.cba.upc.edu/images/TMA/Presentations/Monitoring_Internet_Censorship_with_UBICA.pdf 


- In 2013, there was a cross-cooperation mentioned with UC Louvain / 
tracebox & Tor / OONI: see this: http://www.multipath-tcp.org

They wanted to add this capability to RIPE Atlas probe; however, as Art 
diplomatically put it: "the policy and governance model of ripe does not 
allow for the sort of" addition. Here is more recent article about their 
work:
https://labs.ripe.net/Members/julien_colmonts/tracebox-a-nice-tool-to-detect-middleboxes

Let's keep talking - I'd love to come up with a successful model of 
cooperation!

Regards,
Vesna