16:01:51 <asn> #startmeeting SponsorR
16:01:51 <MeetBot> Meeting started Wed Feb 18 16:01:51 2015 UTC.  The chair is asn. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:01:51 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic.
16:01:54 <asn> that's the right way to do it.
16:02:05 <asn> OK, so who is around now?
16:02:08 <asn> Is this even the right time?
16:02:17 * karsten is around
16:02:21 <asn> great karsten
16:02:27 <ohmygodel> syverson should coming
16:03:13 <ohmygodel> … not that we should wait for him
16:03:16 <asn> ok
16:03:20 <asn> so let's start with status reports
16:03:25 <asn> over the past week
16:03:32 <asn> for sponsorr
16:03:35 <asn> i fixed up the blog post.
16:03:40 <asn> now it's on blog.toprproject.org but unpublished
16:03:48 <asn> i'm still waiting for kate to look at it.
16:03:49 <ohmygodel> when will it be published ?
16:03:51 <ohmygodel> ok
16:04:12 <asn> and i also look a bit more at AnonStats1 and AnonStats3
16:04:15 <asn> i just posted some thoughts on [tor-dev]
16:04:20 <asn> but we can discuss more later.
16:04:25 <asn> and that's that for SponsorR.
16:04:31 <asn> whos next?
16:04:34 <ohmygodel> ok i can be quick
16:04:37 <asn> yes ohmygodel
16:04:38 <ohmygodel> i worked on peerflow
16:04:51 <ohmygodel> deadline next monday
16:04:54 <ohmygodel> thats pretty much it
16:04:59 <asn> makes sense
16:05:03 <asn> good luck with the deadline
16:05:06 <asn> karsten: ?
16:05:09 * ohmygodel need sit
16:05:14 <karsten> sure
16:05:26 <karsten> I worked on teor's #13192 patch together with dgoulet.
16:05:39 <karsten> interesting edge cases with floats and ints..
16:05:41 <asn> yes thx for that. still on my TOREVIEW list.
16:05:43 <karsten> "interesting"
16:05:49 <asn> on the current code?
16:06:16 <asn> there are triggerabe edge cases on the current code?
16:06:19 <karsten> yes. though I think it's mostly relevant for certain compilers, not all.
16:06:30 <asn> ok
16:06:35 <karsten> I didn't follow the latest changes on the ticket.
16:06:36 <karsten> too crazy.
16:06:42 <Yawning> it's fine
16:06:56 <karsten> ah, cool. thanks for working on that Yawning.
16:06:58 <Yawning> now, after the fixes
16:07:11 <Yawning> np, floats, are fun, let's go shopping
16:07:14 <karsten> :)
16:07:17 <asn> ok
16:07:33 <karsten> so, I also worked on improving the extrapolation code towards running it every hour on the metrics server.
16:07:33 <asn> and you also posted that code about calculating the graphs every one hour
16:07:38 <karsten> yep.
16:07:49 <asn> that's also on my TOREVIEW list. but this is looking worse.
16:07:50 <karsten> it's running right now, and it's producing new data every hour.
16:08:02 <asn> because it's java, which I don't speak very well.
16:08:03 <karsten> yeah, it's not tiny.
16:08:06 <karsten> and it's java.
16:08:11 <asn> and also it's big, and not much documentation.
16:08:14 <asn> i would suggest
16:08:18 <asn> you point out specific areas that you want review
16:08:19 <karsten> what documentation would you want to see?
16:08:32 <asn> what the functions do in each step
16:08:43 <karsten> like an overview doc?
16:08:45 <asn> i looked a bit at the Extrpolate.ava or something
16:08:47 <karsten> or inline comments
16:08:58 <asn> but i didn't have enough time to reverse engineer the code
16:09:08 <asn> inline comments I think
16:09:11 <karsten> ok.
16:09:14 <karsten> I'll write some.
16:09:20 <asn> or maybe suggest me a specific area to look at.
16:09:26 <asn> the one that is more prone to failure.
16:09:30 <asn> or more fiddly
16:09:35 <karsten> ok.
16:09:40 <karsten> will do both.
16:09:44 <asn> ok
16:09:45 <karsten> that's all from me.
16:09:54 <asn> otherwise this review would have been a week-long project.
16:10:04 <karsten> yep.
16:10:11 <asn> ok thanks for that karsten
16:10:13 <asn> anyone else around?
16:10:14 <asn> dgoulet: ?
16:10:23 <asn> maybe not.
16:10:26 * syverson can go.
16:10:39 <asn> syverson: ah hello
16:10:40 <asn> yes!
16:11:03 <syverson> Completed and submitted Memex monthly report.
16:11:14 <syverson> Also worked on PeerFlow.
16:11:17 <syverson> That's it.
16:11:41 <asn> great thx
16:11:55 <asn> ok
16:11:59 <asn> so status reports are done
16:12:02 <asn> let's go to discussion
16:12:06 <asn> what would you like to talk about?
16:12:22 <asn> Unfortunately, I'm still unsure on what to do about the stats part over the next month.
16:12:25 <asn> I will have to consult nick or something.
16:12:39 <asn> or roger...
16:12:48 <ohmygodel> you are trying to decide on what to work on, right?
16:12:51 <asn> yeah...
16:13:15 <ohmygodel> so this last week you looked at some of the AnonStats suggestions i had made
16:13:16 <asn> I still believe that the stats aggregation thing should have priority, but it looks like a research project.
16:13:20 <asn> yes
16:13:47 <asn> I'm not very happy with the AnonStats privacy guarantees right now.
16:14:00 <ohmygodel> and then you were going to think about either doing that or some of the stats that dont need anonymization
16:14:03 <ohmygodel> yeah asn i read your email
16:14:25 <asn> yeah
16:14:27 <asn> that was the dillema
16:14:50 <asn> my plan is to do something useful next week, and discuss the dillema it in person in Valencia.
16:15:21 <asn> fwiw, I think AnonStats is a nice idea.
16:15:27 <asn> it's just that more research needs to be done I think
16:15:35 <ohmygodel> btw my response to your concern is that yes, you may learn the popularity of the pseudonymous most-popular HS, but you can probably learn that anyway from the total popularity. pseudonymous popularity is not a privacy concern we had raised
16:16:09 <asn> it's not only most-popular though.
16:16:20 <asn> maybe over a big enough period, you are also able to get the popularity of other popular HSes too.
16:16:45 <ohmygodel> yes that is a possibility as well
16:16:48 <asn> yes, in a pseudonymous fashion.
16:17:13 <asn> i don't know.
16:17:23 <asn> i just think that this is not quite ready to write a proposal of right now.
16:17:40 <asn> it's more like a paper that needs to be written or something.
16:17:41 <ohmygodel> if pseudonymity isnt enough, then it wont work
16:18:32 <ohmygodel> that paper would get in nowhere. maybe hotpets haha
16:18:48 <ohmygodel> because its trivial and uninteresting
16:19:06 <asn> not trivial. but not very interesting yeah.
16:19:22 <syverson> I'm guessing the chair wouldn't accept that paper.
16:19:25 <ohmygodel> although im excited by the possibility of applying the secure multiparty computation stuff that research people tell me has become practical
16:19:40 <syverson> FYI all ohmygodel is the next HotPETs chair.
16:19:41 <ohmygodel> so if thats the direction were heading, :-D
16:20:02 <asn> :)
16:20:36 <asn> ohmygodel: do you have any suggestions here? my current intuition is to wait till Valencia and discuss these things in person.
16:20:45 <asn> but then we won't have enough time to produce any big outcome by April.
16:21:13 <ohmygodel> hm
16:21:13 <asn> again I think I am a bad person to take decisions here.
16:21:24 <asn> maybe I'm not YOLO enough.
16:21:30 <ohmygodel> it would be helpful to get rogers input
16:21:41 <asn> i don't think we can get that before valencia ;)
16:21:55 <asn> without bugging him considerably.
16:21:56 <ohmygodel> he tends to be decisive. not sure about yolo though
16:21:59 <asn> but nick could help.
16:22:12 <ohmygodel> roger knows sponsor r better
16:22:17 <asn> yes ture
16:22:18 <asn> true
16:22:28 <asn> OK, maybe I can shoot a mail to Roger and ask for directions?
16:22:44 <ohmygodel> yeah that seems like a good idea
16:22:55 <ohmygodel> if hes happy with fewer hard results this next quarter, then great
16:22:59 <nickm> asn: if you do, be clear that if he needs to think about the answer, he should instead put you in contact with some person XYZ who can give you a faster answer :)
16:23:09 <nickm> that's what I do
16:23:14 <asn> ack
16:23:22 <asn> OK, then I can send that email.
16:23:29 <asn> And also do some work on #8243.
16:23:36 <asn> so that I get a bit of change.
16:24:11 <ohmygodel> sounds good
16:24:13 <asn> Anything else we should discuss?
16:24:45 <ohmygodel> i have nothing to discuss
16:24:50 <asn> (I also need to review #13192 and karsten's java. so not sure how much time there will be for #8243)
16:24:53 <asn> ohmygodel: ok no problem.
16:25:06 <karsten> asn: feel free to prioritize #8243.
16:25:21 <karsten> I'll need some time to comment things better anyway.
16:25:41 <karsten> agreed that reviewing #13192 is worthwhile.
16:26:04 <asn> OK.
16:26:19 <asn> another thing we should discuss durign valencia
16:26:24 <asn> is the "enable stats by default" thing.
16:26:28 <karsten> yep.
16:26:43 <asn> I feel that real life discussion will be better for this.
16:27:07 <ohmygodel> thats a good idea
16:27:12 <ohmygodel> we should keep a list of these things
16:27:13 <asn> syverson: also, I read your email about changing the HS terminology in steps, and not all at once and officially.
16:27:20 <asn> syverson: i think that might be reasonable.
16:28:06 <asn> but not for this blog post. this one is long done.
16:28:26 <asn> OK, so we can call this meeting done for now.
16:28:35 <syverson> OK. Maybe a little more chat in Valencia on that too.
16:28:36 <asn> I'm also quite tired of computer screens today.
16:28:43 <asn> syverson: absolutely
16:28:48 <ohmygodel> im adding valencia discussion topics to tho sponsor r page
16:28:49 * asn notes it down
16:28:54 <ohmygodel> under the “meeting notes” section
16:28:59 <karsten> is that on the first two days or the last?
16:29:00 <asn> ohmygodel: maybe add it to the valencia page?
16:29:01 <ohmygodel> please add ideas
16:29:06 <ohmygodel> whats the valencia page ?
16:29:07 <asn> i have added some
16:29:10 <karsten> because I'm only going to be there on the first two days.
16:29:13 <asn> https://trac.torproject.org/projects/tor/wiki/org/meetings/2015WinterDevMeeting#SessionsNotes
16:29:23 <asn> karsten: we will do it whenever you are there
16:29:31 <ohmygodel> ok ill add there instead. thanks!
16:29:32 <karsten> cool :)
16:29:48 <asn> OK great.
16:29:57 <asn> Thanks for the meeting guys :)
16:30:08 <karsten> next meeting on tue?
16:30:11 <asn> yes
16:30:16 <asn> that's my understanding.
16:30:19 <karsten> great. thanks for running this meeting. as always. :)
16:30:19 <ohmygodel> yeah well have a paper draft to share
16:30:22 <asn> do you think I should send mail to clarify?
16:30:28 <karsten> ohmygodel: good luck with that!
16:30:35 <asn> ohmygodel: syverson yes good luck!
16:30:39 <ohmygodel> thanks !
16:30:47 <syverson> Same bat time same bat channel. (Says the old man.)
16:30:54 <asn> :)
16:30:56 <asn> #endmeeting