18:01:23 <hellais> #startmeeting
18:01:23 <MeetBot> Meeting started Mon Feb 23 18:01:23 2015 UTC.  The chair is hellais. Information about MeetBot at http://wiki.debian.org/MeetBot.
18:01:23 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic.
18:01:28 <hellais> here we go
18:01:34 <hellais> so who is here?
18:01:54 <sbs> Hi
18:06:32 <hellais> anadahz, aagbsn: are you here?
18:07:24 <aagbsn> yes
18:08:18 <hellais> ok, I guess I will start by giving a summary of what has been done in the past couple of weeks
18:08:24 <anadahz> hellais: yes
18:08:27 <hellais> then we can move into discussing the topics
18:09:26 <hellais> so I have finished migrating the pipeline and the data has been put through all the steps of it
18:09:30 <hellais> it
18:09:54 <hellais> what it's still missing is setting up the cronjobs to keep everything in sync and re-run it automatically
18:10:02 <hellais> plus the website of the reports
18:10:30 <hellais> what has changed in the ooni-pipeline code is that it now supports multiprocessing
18:10:44 <aagbsn> ah that should speed it up rather a lot
18:10:46 <hellais> making it significantly faster
18:10:59 <hellais> yeah like 24x faster :P
18:11:20 <hellais> the data format in which it is imported in the DB has also changed
18:11:35 <hellais> if you recall some time ago we had two different tables: "reports" and "measurements"
18:12:14 <hellais> it turns out mongodb does not work very efficiently with JOINs so it's much better to just stick all the data inside of an array called "measurements" in every report
18:13:18 <hellais> there is also a new release candidate tagged (1.2.3-rc1) that I hope we will soon get to be merged into debian testing
18:13:52 <hellais> if would be great if anybody running debian or even other systems would test it to make sure that everything works properly on as many platforms as possible
18:14:11 <hellais> I have also been discussing a couple of censorship monitoring projects in two interesting countries
18:14:33 <anadahz> hellais: I can test the debian packages
18:14:39 <hellais> in particular we have been thinking of ways to check which censorship circumvention tools work
18:14:45 <hellais> anadahz: that would be awesome
18:15:13 <hellais> anadahz: though note that we don't have packages of 1.2.3-rc1 yet, so testing means also building them and/or installing it via pip
18:15:26 <anadahz> yep
18:15:59 <anadahz> I 'll have a look at the install.sh I see many bug reports about unsopperted OSes
18:16:09 <hellais> WRT censorship circumvention tool testing we have decided to opt for only testing open source ones, since doing it for the closed source ones automatically and reliably is going to be a pain
18:16:22 <hellais> anadahz: yes that would be very useful too
18:16:38 <hellais> and then there is the discussion on the future of OONI
18:17:05 <hellais> that more or less sums it up
18:18:46 <hellais> anybody else have something to share? questions to ask? comments to make?
18:18:56 <sbs> Well
18:19:25 <sbs> I did some testing and bug fixing of libight
18:19:42 <hellais> sbs: lol, dude why is your text black on white?
18:19:54 <sbs> No idea
18:20:18 <sbs> I'm texting from my mobile in a cafe
18:20:38 <hellais> ah ok, it's very retro
18:20:38 <sbs> Maybe some setting of the app?
18:20:42 <anadahz> fancy :P
18:20:49 <sbs> 😊
18:20:57 <hellais> anyways please continue :P
18:21:43 <sbs> I also added pull requests to simplify embedding, meaning that
18:22:48 <sbs> Now one can install libight its header and the dependency library and headers with make install
18:22:58 <sbs> This simplifies embedding
18:23:40 <hellais> ah on that topic I found this library that may be a good example of making a network library to be used on iOS and Android
18:23:44 <hellais> http://openpeer.org/
18:24:10 <sbs> Finally I used ooni pipeline to import Neubot and glasnost data in mongo for a paper to which I am working
18:24:14 <sbs> Cool!
18:24:56 <sbs> https://github.com/bassosimone/neubot-pipeline
18:25:21 <sbs> This uses a modified ooni pipeline
18:25:43 <sbs> I will send you a pull request with my changes
18:25:47 <hellais> on that we should think of a way of making pre-processors and post processors be something modular, so that we can have something like a plugin system in the ooni-pipeline
18:25:48 <sbs> That's all
18:26:08 <sbs> Yes, makes sense
18:26:45 <hellais> so we could have say the NDT plugin, the NeuBot plugin, the glasnost one, etc.
18:27:05 <hellais> anyways if this works for you, let's keep it like so at the moment
18:27:36 <hellais> anadahz: what's the news from the raspberry pi world?
18:28:02 <anadahz> have you tested the vmdebootstrap vagrant ?
18:28:24 <hellais> anadahz: no I haven't gotten around to doing that
18:28:41 <anadahz> hm.. I 'll figoure something else then
18:28:57 <hellais> did you try it with vagrant and it works?
18:28:59 <anadahz> It seems that the git version of vmdebootstrap works OK
18:29:13 <anadahz> I don't use vagrant..
18:29:56 <anadahz> but the step on the vagrant file are same with my tests on a clean debian vm
18:30:02 <hellais> then how do you know that it works ok with vagrant?
18:30:18 <anadahz> i don't :)
18:30:29 <hellais> ah ok
18:30:31 <anadahz> that's why i asked if you have tested the script :P
18:30:41 <hellais> i see i see
18:30:55 <hellais> I will start it in the background now even
18:31:10 <anadahz> tbh not too much news from my side. I'm still recovering from the hardware failure
18:31:26 <anadahz> I fried too many devices last week :(
18:31:44 <hellais> oh that sucks
18:32:04 <anadahz> all in all I 'll come with a way to build images on a VM or similar
18:32:36 <hellais> what hardware did you fry?
18:32:58 <anadahz> my main PC another laptop and an amplifier
18:33:06 <anadahz> it's crazy
18:33:10 <hellais> wow
18:33:24 <anadahz> so now I'm on an underpowered notebook until I find a better hardware
18:33:35 <anadahz> and I lost some data too..
18:33:42 <anadahz> but that's OK
18:34:21 <anadahz> I have ordered a Pi case with an LCD.. for furher tests.
18:34:46 <hellais> ah can you paste the link to it so we have it archived in the transcript?
18:36:02 <anadahz> http://www.banggood.com/3_5-Inch-TFT-LCD-Display-DIY-Acrylic-Case-For-Raspberry-Pi-BB-p-960194.html
18:36:09 <hellais> cool thanks
18:36:30 <anadahz> btw do we have a plan for circumvention tech?
18:36:59 <anadahz> Tasks, talks, etc.. ?
18:37:47 <hellais> well not really
18:38:17 <hellais> I am not too familiar with the format of the event
18:38:52 <hellais> but I bet it would be a great place where it would be good to have some raspberry pi's or at least SD cards with the image on it handy to give out to people\
18:39:35 <hellais> another fun thing to do there would be get some people interested in checking out the data and hacking away at that
18:41:16 <hellais> did any of you get a chance to take a look at the next steps in the ooni roadmap: https://lists.torproject.org/pipermail/ooni-dev/2015-February/000255.html?
18:41:43 <hellais> in the next days I will start transforming those various sections into some tickets
18:43:55 <sbs> Not yet, sorry
18:46:15 <hellais> comments on that are very appreciated
18:47:07 <anadahz> yes from me as well..
18:47:46 <anadahz> hellais: you mentioned about censorship circumvention tools testing..
18:48:40 <hellais> anadahz: yes I did, what in particular would you like to know about it?
18:51:17 <anadahz> is there a ticket, or sth to start reading about that?
18:51:29 <anadahz> I would like to take part on this
18:51:55 <hellais> anadahz: currently there is not, because I am not sure how I much I can disclose on the topic yet
18:53:43 <hellais> the main tools we will be looking at are Psiphon, Tor and it's pluggable transports, Scotty, popular VPNs
18:55:14 <sbs> So the idea is to write ooni tests and specs that mimic them
18:55:36 <hellais> sbs: correct
18:57:03 <sbs> Sounds interesting :-)
18:57:57 <hellais> is there anything else?
18:59:22 <sbs> Not from me
19:00:09 <hellais> ok well then thanks for attending the meeting
19:00:21 <anadahz> http://www.emma-hennig.de/Projekte/13731/oonireports.html
19:00:21 <hellais> I guess I will see a lot of you in valencia in less than a week
19:00:31 <hellais> so we will probably skip next weeks meeting
19:00:46 <anadahz> ^ this is a small project a student is working on
19:01:27 <sbs> anadahz: what is ZZ?
19:01:43 <hellais> anadahz: wow very cool
19:01:52 <anadahz> it's an undefined country
19:02:01 <hellais> sbs: ZZ is when we don't know the country
19:02:40 <hellais> anadahz: is the code for this available somewhere?
19:02:49 <anadahz> I think the student has imported the reports "manually" the code is the html files.
19:02:57 <hellais> I could possibly integrate the data processing stuff so that it can be updated in real time
19:03:06 <anadahz> I hope that it 'll be improved.
19:03:38 <hellais> oh my, I see...
19:03:58 <anadahz> but nice visualizations eh?
19:04:09 <hellais> I see they tried to use d3.js, but then just hardcoded it
19:04:43 <sbs> anadahz: ah, lol, so when the country is unknown "save and quit" :-P
19:04:45 <anadahz> I guess we can add a section on the website with this version of the reports as well
19:05:04 <sbs> Yes, nice viz
19:05:14 <hellais> anadahz: do the colors mean something?
19:05:39 <anadahz> no
19:06:27 <hellais> anadahz: I wonder how it would look if you set the background of the circles to be the flag of each country with this: http://lipis.github.io/flag-icon-css/
19:07:56 <anadahz> too small circles for flags :P
19:08:11 <sbs> I've to go because my battery is over, bye!
19:08:37 <anadahz> sbs: bye thanks for attending
19:09:35 <hellais> anadahz: I can still sort of tell the flag at 15x15 pixels
19:10:01 <hellais> anyways we shall see
19:10:31 <hellais> tell this person that if they need help with d3.js or with pre-processing the data they should join this channel and I would be happy to help
19:11:26 <anadahz> hellais: ah nice
19:12:30 <hellais> ok well thanks for attending
19:12:34 <hellais> see you soon!
19:13:15 <anadahz> see ya
19:13:34 <hellais> #endmeeting