From 9868570a6744931aa209fa35326397e68272ced1 Mon Sep 17 00:00:00 2001 From: kytv Date: Wed, 8 Jan 2014 19:04:39 +0000 Subject: [PATCH] add meeting 229 --- i2p2www/meetings/logs/229.log | 234 ++++++++++++++++++++++++++++++++++ i2p2www/meetings/logs/229.rst | 19 +++ 2 files changed, 253 insertions(+) create mode 100644 i2p2www/meetings/logs/229.log create mode 100644 i2p2www/meetings/logs/229.rst diff --git a/i2p2www/meetings/logs/229.log b/i2p2www/meetings/logs/229.log new file mode 100644 index 00000000..033bb1c7 --- /dev/null +++ b/i2p2www/meetings/logs/229.log @@ -0,0 +1,234 @@ +20:02:10 http://zzz.i2p/topics/1539 +20:02:10 0) hi +20:02:10 1) jisko.i2p console home page submission http://zzz.i2p/topics/1539 +20:02:10 2) i2pjump.i2p addition http://zzz.i2p/topics/1539 +20:02:10 3) Host release files on i2p.no http://zzz.i2p/topics/1390 +20:02:10 4) OpenITP audit manager http://zzz.i2p/topics/1533 post #4 +20:02:11 5) anything else +20:02:13 6) baffer +20:02:19 0) hi +20:02:24 hi +20:02:40 1) jisko.i2p console home page submission http://zzz.i2p/topics/1539 +20:02:46 Is the jisko op here? +20:03:35 any comments on the jisko submission? +20:03:44 +1 +20:03:54 add it. +20:04:28 any objections? +20:07:05 hi +20:07:11 No objections here +20:07:13 great, even jisko's competitor does not object... +20:07:13 It's a good alternative to id3nt (which I need to spend some time on sometime) +20:07:13 Hey, the more the merrier +20:07:13 indeed. I'll add it to mtn +20:07:13 Then if I go down, there are alternatives +20:07:13 2) i2pjump.i2p addition http://zzz.i2p/topics/1539 +20:07:33 I see tuna is not here but he claims he got all the tests to pass +20:07:41 and he posted his code +20:07:46 any objections? +20:07:54 or comments? +20:08:22 no objections so far +20:09:17 there's this guy emailing grothoff saying he was "inspired" by the jump code +20:10:25 hearing no objections, I'll ask tuna to add it to mtn since he has privs +20:10:31 3) Host release files on i2p.no http://zzz.i2p/topics/1390 +20:10:39 Meeh, are you here? +20:11:06 we discussed at CCC, he said he almost certainly had the bandwidth +20:11:22 are there any reasonable alternatives? +20:11:28 google drive seems a little tacky +20:11:30 thoughts? +20:11:50 google drive is a bit ugly, right +20:12:08 but it is quite hard to block. +20:12:23 I would run a download host and a copy in google drive +20:12:36 or somewhere else like Mega, Sharehosts,... +20:12:56 str4d, what's the state of the pick-your-mirror code in the website? +20:13:51 zzz: mirrors get added in a backend file, and are used to generate the file-specific list. +20:15:16 oh, I forgot about the 'select alternate link' +20:15:19 Path urls can contain the version if desired (so files can be organized on the server by version) +20:15:31 The default is set in the backend code +20:16:05 would 'alternate download locations' be better than 'select alternate link' ? +20:16:07 or 'view mirror list' +20:16:27 thats really nice I forgot that we worked on that a lot +20:17:08 'Any mirror' pick a mirror at random from the list. +20:17:08 zzz: here now! +20:17:15 any objections to Meeh being the primary download location? +20:17:23 i2p.nu * +20:17:25 or i2p2.no +20:17:39 oh it's not i2p.no? +20:17:40 currently not, as long as it is not the only one +20:17:48 no sorry, I got i2p.nu, and i2p2.no +20:18:11 add "facebook" as a point on the meeting for me, got some nice news +20:18:41 ok FB will be 5) +20:19:01 the bad word... +20:19:12 I just need to cough... *sorry* +20:20:03 looks like you are already in the mirror list as download.i2p2.no, is that working? +20:20:07 if you want to add i2p.nu also, fine, more the merrier, as long as you are around at update time, or somebody else has the password +20:20:41 you can work with str4d on the urls? +20:21:19 yepp +20:21:34 any other comments? +20:22:02 I need to check where download.i2p2.no points, might need to change it to another host, need to check I/O. and I can provide more urls if needed. other than that; no :) +20:22:38 ok you have a couple weeks before the next release but better sooner than later to get everything right +20:22:48 No comments from me +20:22:50 I got the domains, I say we can use it as you guys wish, meaning, just give me instructions on how you want it :P +20:23:14 +1 for everything so far (and I'll be happy to continue to handle the uploading of the download files) +20:23:18 I'm tinkering with Makefile.gcj, and I've ran into a snag brought on by my relative naivety of java & gcj. I'm getting a java.lang.NullPointerException brought on by missing jar.so's like ant-1.8.2.jar.so, which exist in a dir outside of $JAVA_HOME, and I'm guessing that that's the problem +20:23:26 you and str4d can work the details if needed +20:23:42 DarkestMatter, we are in the middle of a meeting if you would please wait about half an hour +20:23:45 Where would I need to symlink the ACTUAL dir into under JAVA_HOME for the build process to find them? +20:24:06 next on the agenda: +20:24:06 4) OpenITP audit manager http://zzz.i2p/topics/1533 post #4 +20:24:29 background: OpenITP is about to open up their submission process +20:24:38 we need somebody in charge +20:24:56 They said we should apply even if we aren't ready +20:24:59 which we arent +20:25:28 I've posted a 3-phase process in the post above +20:25:45 where we submit, then get ready, then really submit and manage the audit +20:26:05 at CCC we agreed fixing up the threat model is the most important +20:26:31 Comments? Does anybody want to be in charge? +20:28:10 The general plan looks sound. +20:29:48 Suspecting that we would not have any volunteers, I discussed it with Brandon WIley, asking if we could pay him to do it. (He is currently consulting for zooko, whose company does audits). Thoughts? +20:30:05 how much? +20:31:03 I am happy to contribute to the audit, but will likely not have the time to manage it myself. +20:32:26 I assume several thousand dollars at least +20:32:26 much higher if we actually get audited although that probably wouldn't be until 2015 +20:32:29 it's probably better for an external party to do it even if it costs a few k +20:32:29 If we pay him, we want to ensure that it is useful. +20:32:53 that would incluse exactly what? +20:32:54 if he's doing this for a living he is (most probably) going to do a better job than anyone of us +20:33:49 ++1 external, and ++1 if it helps getting a better standing, review-wise +20:34:06 I would want him to actually help us fix stuff, not just be a traditional "consultant" or preauditor who just points out problems +20:34:27 Brandon is a busy guy but also a guy currently looking for money +20:34:56 and would he help you fix? +20:34:58 If you havent' seen it: http://www.kickstarter.com/projects/brandonwiley/operator-a-news-reader-that-circumvents-internet-c +20:35:52 I asked him to look at our openitp audit page on trac +20:36:07 http://trac.i2p2.i2p/wiki/OpenITPReview/Criteria +20:36:19 and await openitp's announcement and submission process details +20:36:26 I2P has never (to my knowledge) had a proper security expert. I think paying for one to get the groundwork for this right is a good idea (and probably long overdue). +20:36:27 my opnion as an outsieder / user only, is that a few thousand $ should be ok, as long as it's <10k and not just some makeup, but really thorough +20:36:42 until that's published it's not worth getting into negotiations with him +20:36:53 but Eleanor said any day... +20:37:51 eche|on, what's our balance? ~$500K? +20:38:00 If you want that I change topic of http://zzz.i2p/topics/1546? +20:38:31 wowa, we are in the middle of a meeting, please wait until we are done, thx +20:38:36 zzz: roughly 500 BTC and 50k € +20:39:37 sounds like people are positive about the idea. Sadly even if we hire him to be in charge, we need to find somebody to be in charge of him... +20:39:40 Moin +20:42:42 I don't think there's anything more to do until the OpenITP announcement, then we can ask if Brandon is interested, if so we will have to talk about again +20:42:49 any other comments on 4) +20:42:49 Did I miss the meeting? +20:42:49 EinMByte: in progress +20:42:49 Ein: Just the beginning +20:42:49 ok, 493 BTC and 48k euro +20:42:49 oh, okay +20:42:49 last call on 4) +20:42:49 No comments at this time. +20:42:49 Other than, planning should start on the wiki +20:42:49 str4d, let's discuss in NYC too +20:42:49 zzz: yes. +20:42:49 5) facebook Meeh go +20:42:49 * str4d will move the 3-step plan to /wiki/OpenITPReview/Plan +20:42:49 ok +20:42:49 As I said to zzz on CCC, I now got control over the https://www.facebook.com/I2P page +20:42:49 Good thing for publicity, I suppose +20:42:49 way cool :) +20:42:49 I personally avoid the thing, but obviously most people have facebook +20:42:49 and I used ~10$ and 15min of my time in between my dayjob, and likes rised from 150 to 300 +20:42:49 so far good, but I keep outof facebook +20:42:50 this was in less than 10 hours +20:42:56 so it's definitely a good thing to get more users +20:43:02 you don't need a account to see the page +20:43:22 however, I think we need to be public there too, to get "big", more users +20:43:22 Meeh: I know, was more referring to liking and the like +20:43:37 but, nobody here like facebook, nor got an account :P +20:44:04 so, I might have something from work, that I can reuse so we can have an i2p url for some selected which can login and post to facebook, via I2P +20:44:09 Hello, about 3) would it also be possible to host releases on github? +20:44:12 Meeh, is that it or are you asking for discussion or...? +20:44:55 mostly info, but I think it could be a idea to have an discussion about reaching our users via "social media" +20:44:57 TerraNullius, we are done with 3) for now, sorry, you can wait until after the meeting to discuss further +20:45:37 Meeh, please discuss with orion, he's our head of publicity, I assume he has some ideas +20:45:43 Meeh: obviously those new users will need to understand that using "social media" isn't without risks +20:45:54 we can also brainstorm with him in NYC and get back to you +20:46:07 we need more contributors, content providers, users, +++, and as we can see the "corporate" world collects a lot of users from SM +20:46:32 EinMByte: ofc. but someone needs to tell them ;) +20:46:39 that's where we comes in +20:46:41 very true +20:46:47 and underground I2P parties :) +20:46:47 Who would moderate i2p<->fb? That will be spam central. +20:46:49 you're paying people to like you? +20:46:56 (until 6 am) +20:47:03 yes a little now just to see how much I could boost it +20:47:05 zzz: yes, some folks do that +20:47:16 22 friends and the rest from ads +20:47:23 now all we need are firetrucks +20:48:11 Giant: read me again. "for some selected which can login" :) +20:48:16 selected is the keyworld +20:48:18 word* +20:48:24 good job Meeh , let's all feed him some content. There's a couple pix from ccc too that you can post +20:48:25 I have one from tuna but you'll have to crop off the people in the background +20:48:37 anything else on 5) ? +20:48:51 Awsome, if someone can give me a link for it I will upload a gallery, including some from me +20:49:08 ye, I'll fix that +20:49:11 ok just be sure to crop +20:49:26 anything else for the meeting? +20:49:35 yes +20:49:39 why crop? are you on them in the background? ;) +20:49:46 I'd like to propose a short discussion about GNS integration +20:50:16 Are we still looking into that? +20:50:16 * str4d was about to mention that +20:50:27 user: it wasn't allowed to take pictures if not all agreed on it, and we can't find all ppl to ask if it's ok to publish +20:50:35 GNS or others +20:50:35 ok it could go all day so let's try 10 minutes +20:50:37 6) GNS EinMByte go +20:51:07 Meeh: ah, ok. that makes sense +20:51:09 EinMByte: looking into it, yes. +20:51:22 Alright, so what approach are we going to take +20:51:25 reimplement? +20:51:28 No guarantees that it will happen, or be a direct integration (or even that it will be GNS). +20:51:40 is there still some possibility of not having to rewrite? +20:51:46 But what we have now is an ad-hoc "just works" mess. +20:52:16 What about support from their side? +20:52:31 topiltzin: want to be a facebook star again? :) (wrt I2P, pictures, CCC) +20:52:34 They seemed pretty "closed"... +20:53:02 sure thing Meeh go for it :) +20:53:34 spent hours and hours with Christian at CCC +20:54:19 zzz: no results whatsoever from that? +20:54:30 EinMByte: *if* we decide to use GNS as-is (and ignore the fact that by default its DHT is not anonymous), then we could hook it in for testing with ExecNamingService +20:54:33 There's a defined API. We just write a script that can query that API, and run it from ExecNamingService. +20:54:33 Simple. But not at all optimal. +20:54:42 could be anything from 'they solve all our problems' to 'lets use some of it' to 'those are some good ideas' to wow, no thanks +20:55:08 no results really. Other than we promised to look into it +20:55:37 I think we need to understand what they are doing. Beyond that, nothing is decided +20:56:15 atm I'm a skeptic on all things gnunet. Doesn't mean I couldn't be convinced. +20:56:31 zzz forwarded me an interesting exchange about I2P naming, it mentioned several other techs besides GNS +20:57:39 well, I suppose that (as you say zzz), there may be some problems with the ideas behind GNS +20:58:02 That is: will our users be able to adapt +20:58:09 DLV (DNSSEC Lookaside Validation), DNSCurve... +20:58:14 There are more technologies out there than GNS +20:58:14 I can see that many ideas in GNS have been developed from scratch, assuming nothing. +20:58:37 boiling the oceans +20:58:44 sure. It's been a long time since we discussed our naming system in the big context of the internet, the triangle, alternatives, etc, so that's a good thing +20:58:48 Yes. Since we need an alternative to hosts.txt for sure, we might as well look into those alternatives +20:59:00 Even things like how the local cache is stored encrypted, as a direct copy of the blocks obtained from the DHT +20:59:02 but he is indeed trying to boil +20:59:37 I ran a quick test, I can confirm that NamingServices can be distributed as plugins. +20:59:39 we talked about a layer that hides the GNS ugliness that we would have to design and implement. However that may also hide some of the benefits. +20:59:59 It only requires that the default NamingService is a MetaNamingService. +21:01:06 Are there any downsides to making the install-default NamingService a MetaNamingService, and the default added NamingService to it BlockfileNamingService? +21:01:08 let's wrap up 6), we aren't going to fix things here +21:01:08 anything else on 6) ? anything else for the meeting? +21:01:15 (Current default NS is BlockfileNamingService, and current default added NS to Meta is HostsTxt...) +21:01:26 one more thing on 6: +21:01:30 zzz: above proposal is I think part of 6 +21:01:32 str4d, you mention alternatives? +21:01:50 if GNS doesn't work, we can always check out other things +21:01:54 they might fit us better +21:02:25 EinMByte: yes. I have no knowledge of these alternatives, but all ideas are worth consideration. +21:04:12 zzz: I propose that we change to MNS in 0.9.10, there is no visible change to users, but then if people want to test a NS plugin they don't need to edit router.config +21:04:18 ok, everybody keep working on these ideas big and little, thats how we make things better +21:04:35 str4d: okay. I will try to find some time to look into them +21:04:48 I suppose this closes 6? +21:04:54 str4d, we can do it when we need to but no rush? needs more testing too. Maybe post a howto? +21:05:17 ok you all can keep yakking but I gotta run +21:05:19 I'll offer the contrarian view: our current naming system works just fine +21:05:27 same here +21:05:30 and has some benefits we want to keep +21:05:38 * str4d afk +21:06:06 * zzz *bafs* the meeting closed diff --git a/i2p2www/meetings/logs/229.rst b/i2p2www/meetings/logs/229.rst new file mode 100644 index 00000000..500f7e0e --- /dev/null +++ b/i2p2www/meetings/logs/229.rst @@ -0,0 +1,19 @@ +I2P dev meeting, January 07, 2014 @ 20:00 UTC +============================================= + +Quick recap +----------- + +* **Present:** + + eche|on2, + EinMByte, + Giant, + kytv, + Meeh, + str4d, + TerraNullius, + topiltzin, + user, + wowa, + zzz