18:00:36 #startmeeting 18:00:36 Meeting started Mon Jan 26 18:00:36 2015 UTC. The chair is hellais. Information about MeetBot at http://wiki.debian.org/MeetBot. 18:00:36 Useful Commands: #action #agreed #help #info #idea #link #topic. 18:00:46 we also have the meetbot now 18:00:51 cool! 18:00:56 #topic people present 18:01:07 #info 18:01:12 #help 18:01:17 are these macros? 18:02:20 unsure 18:02:39 I mean they are supposed to register some metadata associated to the meeting 18:02:49 https://wiki.debian.org/MeetBot 18:02:59 I see, they dump something into the output file 18:03:16 I hope 18:03:19 we will see 18:04:08 :) 18:04:23 graphiclunarkid said he will not make it 18:08:07 sbs: I guess this means it's just you and me :P 18:08:33 ah, lol 18:08:38 #topic activities of last week 18:09:20 ok, well: 18:09:50 since the update you sent on ooni's ml, which included the hackfest 18:10:22 I worked on a open pull request for libight 18:10:29 https://github.com/TheTorProject/libight/pull/73 18:10:40 I also worked on compiling for iOS and Android 18:11:16 I haved improved alemela's script for iOS but have not pushed that branch on github yet 18:11:28 while working on compiling for iOS and Android 18:11:38 I noticed a thing with the way in which we handle headers 18:11:49 but perhaps we could discuss this later? 18:11:54 that's it 18:12:16 yeah I have a couple of comments I can expand on PR #73 and on the iOS building of libight 18:12:24 ok sounds good 18:12:49 I worked mainly on fixing ooniprobe bugs for the pending new debian release 18:13:48 these are documented in: https://github.com/TheTorProject/ooni-probe/pull/376 18:14:04 I also worked on fixing some bugs in the backend and migrating the bouncer to a new machine\ 18:14:29 in doing so I wrote a bit of documentation and improved the ooni-sysadmin scripts 18:15:32 with graphiclunarkid we also discussed a bit about this OONI + ORG funding proposal 18:15:49 and made a list of what should be the main objectives of the proposal 18:16:27 that's my part 18:16:34 ok, cool 18:16:45 IIRC the proposal deals a lot with visualization, right? 18:17:57 (I think you mentioned me this in Turin, correct me if I'm wrong) 18:18:51 sbs: well that is what I am pushing for 18:19:04 but it's also a lot about integrating with other similar initiatives 18:19:23 #topic discussion 18:19:55 for example to merge some components of Blocked into OONI and vice versa 18:20:22 the core focus is re-creating something like blocked.co.uk for other countries in the world 18:21:55 ah, cool, I know very little about blocked 18:22:09 I have starred their repos and will look at them 18:22:43 sbs: you should look at the website: https://www.blocked.org.uk/ 18:23:10 Yes, that's the place from which I started 18:23:23 I was tried to figuring out how it works, and was attracted by their chrome extension 18:23:43 and firefox extension as well 18:24:59 I see 18:25:08 regarding the libight stuff 18:25:16 yes 18:25:35 do you have extra comments on pull request #73 18:25:37 ? 18:26:01 sbs: does the last comment I wrote there make sense? 18:26:18 I mean I would like to have just 1 settings that I use to visit .onion addresses and normal websites not via tor 18:27:00 Yes, it was clear: you want something affecting how httpo is handled only 18:27:31 or a setting that is just relative to tor 18:27:44 so if in the future we have a setting called "use_tor=yes' 18:27:50 then it will use the tor socks port 18:28:16 I think it's important to just distinguish a normal SOCKS proxy from a tor socks proxy 18:28:55 is it important to also add the address, or is 127.0.0.1 automatically implied? 18:29:30 given what we have now (socks5_proxy = "addr:port") I think it's least surprise to add and honor tor_socks5_proxy = "addr:port" 18:29:49 but if the address is automatically implied, I can simplify the setting 18:30:42 no the ip is not needed 18:30:44 it's just the port 18:31:24 ok 18:32:41 speaking of the headers of libight 18:33:02 I've noticed the following: when you want to integrate libight it is nice to `make install' it in some place 18:33:13 so you only have the compiled library and the required headers 18:33:28 sbs: May I PM you (later) about neubot? 18:34:04 however, typically reusers of a library include headers like this '#include ' 18:34:06 sbs: yeah that seems like a good idea 18:34:17 in our case libname = ight 18:34:38 however, if you do like this, headers referencing headers need to do '#include ' internally 18:34:53 yes I see 18:35:10 given this, I think it's more practical to move headers in $topdir/include/ight/... 18:35:23 which breaks the headers-under-src thing we discussed in turin 18:35:23 sounds like a good idea 18:35:33 but I don't think this is at the moment something of high priority 18:36:05 I think we should be the first users of the library and first get a working prototype that uses it, before making it usable by third parties 18:36:18 I see... I was raising this because it's annoying to write a compile script for iOS or whatever 18:37:09 because you need to move the headers yourself 18:37:51 sol: yes, I will be here until 20:30 CET, I think 18:38:22 sbs: Let me know when you're free to chat? 18:38:40 sol: yeah 18:38:59 I'm planning a large roll-out of neubot, and would like to have your input first. 18:39:09 sol: ah, cool! 18:39:28 "large" being 20-40 instances. 18:39:55 hum, I see, I guess it just depends how much effort it is to make this fix, if it's very little and it means making the iOS app will be much easier/cleaner, then by all means don't let me stop you! 18:40:23 sol: will you also be deploying ooniprobes? 18:40:32 hellais: ack ;-) 18:40:40 hellais: I'll be idling on this channel. Please do let me know if there's any movement on ooniprobes on the raspberry pi. 18:41:01 hellais: Yep. The idea is to deploy both neubot and ooniprobe. 18:41:07 sol: awesome! 18:41:16 I'll ping the developer working on that now 18:42:51 anyways I guess we should move onto what next 18:42:56 #topic next steps 18:43:13 how do I know that probe I run everyday positiely reports to the collectors? Yeah, I can check logs, but as a user I have not easy way to tell, would be nice to make it better 18:43:34 this week I plan on completing the migration of the pipeline and setting up all the glue to run all thje tasks automatically 18:44:00 tqx: if you run oonireport status you will get the list of reports that have not been submitted properly 18:44:23 if you don't see anything in there then it means that they have all been submitted properly 18:44:48 great, thanks 18:44:53 np 18:47:34 speaking of next week and other future stuff, I plan to finish the pending pull request 18:47:55 next, I'd like to streamline a bit compilation on iOS and Android 18:48:14 hellais: this reminds me, was nuke able to improve the iOS app? 18:48:43 and I will also spend time on integrating Tor and running it on a separate thread 18:48:48 sbs: well he started making it, but I'm not sure at what point he is with it atm 18:49:26 I don't think most of these things will happen this week, as I have to write a deliverable for an EU project 18:49:49 hellais: ah, ok... do we still want to integrate Tor and run it from a thread in any case, right? 18:49:56 yes I will also continue working on that OTF proposal 18:50:21 and depending on what vasilis says also update the raspberry pi image 18:50:39 sbs: yeah. I don't see how else we could do it 18:50:52 though he said that there is a better library than the onionbrowser 18:51:18 CPAProxy? 18:51:45 it's called onionkit: https://github.com/chatsecure/onionkit 18:52:16 hellais: thanks! 18:53:37 hellais: I will look at it... at least it could teach us ways to run Tor in a separate thread (the author of such library is also a contributor to CPAProxy btw) 18:55:23 yeah 18:55:40 are there any more points to discuss, questions, comments? 18:56:05 no, I think we can close the meeting 18:56:56 hellais: where the meeting records are going to be published? 18:58:20 sbs: check /msg 18:58:56 ok perfect 18:59:00 thanks for attending 18:59:14 #endmeeting