1 2015-10-12 08:00:42 <jmcn> would running the 32bit version of bitcoind use significantly less memory?
  2 2015-10-12 08:01:23 <jmcn> (I have a dedicated machine (Intel Atom) with 2Gb memory and it's really not coping)
  3 2015-10-12 08:13:29 <Luke-Jr> jmcn: #bitcoin
  4 2015-10-12 09:46:14 <jmcn> Luke-Jr> thanks
  5 2015-10-12 13:20:49 <harding> Hey, wumpus created an alert for Bitcoin.org about the upnp vulnerability.  Is anyone able to do a quick review before we go live?  https://github.com/bitcoin-dot-org/bitcoin.org/pull/1086
  6 2015-10-12 13:21:52 <harding> Oh, wait, maybe he was preparing that in advanced for when the 0.11.1/etc binaries go up.
  7 2015-10-12 13:52:41 <wumpus> well review is good; but yes we should have at least rc binaries up before it can go live
  8 2015-10-12 14:02:45 <harding> Ok.  Sorry, this took me by surprise and I thought it was more urgent.
  9 2015-10-12 15:58:42 <OxADADA> did the new email list get created yet?
 10 2015-10-12 16:02:40 <bsm1175321> There's supposed to be a meeting about it today.
 11 2015-10-12 16:03:00 <maaku> did the old list ever get shut down? you could just point people there ;)
 12 2015-10-12 16:03:11 <jamesob> is the new list for devs, or general discussion/soapboxing?
 13 2015-10-12 16:03:26 <maaku> general discussion, no soapboxing
 14 2015-10-12 16:03:37 <jamesob> amen
 15 2015-10-12 16:30:05 <cjcj> Question about python-bitcoinlib: I can sign a transaction and pass a check in VerifyScript just fine, but when I try to broadcast with sendrawtransaction I get the following error: mandatory-script-verify-flag-failed (Non-canonical DER signature)'  code: -26. Does someone know what I am doing wrong?
 16 2015-10-12 17:05:28 <warren> Meeting regarding dev/discuss list policy in 1 hour 55 minutes.
 17 2015-10-12 17:08:12 <Texploit> @warren When is the HK dev shops? I need to know so i can get back from china in time. i know the main shop is dec. 6-7 but todd told me that there is something just before
 18 2015-10-12 17:19:39 <floopflorg> how immediate is the "walletconflict" property filled on wallet transactions? I see that the "time" property is the same, but how reliable is it to be there when the transaction(s) first enter the mempool?
 19 2015-10-12 17:20:31 <floopflorg> (also, I am not sure if this an appropriate room to ask questions about tangeltal bitcoin dev such as this, but the main #bitcoin channel has generally not been help[ful on matters like this)
 20 2015-10-12 17:28:59 <maaku> cjcj: what is going wrong is you have a non-canonical DER signature
 21 2015-10-12 17:29:09 <maaku> sign it with something else
 22 2015-10-12 17:29:34 <maaku> bitcoind can sign for you, with private keys passed in by JSON-RPC
 23 2015-10-12 17:35:26 <floopflorg> So I am seeing that it gets the -1 conf and the conflict array when it is neither in the blockchain, nor the mempool. How long would that generally take from the time it is initially seen?
 24 2015-10-12 17:40:01 <cjcj> maaku: Yes, I understand that, but is there something wrong with python-bitcoinlib then? I can see that it uses openssl for signing, but I am not savvy enough to locate where the issue might lie.
 25 2015-10-12 17:40:59 <cjcj> maaku: I'll try using the rpc for now, but am curious what is wrong with python-bitcoinlib
 26 2015-10-12 17:41:03 <cjcj> maaku: thanks
 27 2015-10-12 17:42:13 <maaku> cjcj: maybe. this isn't really a python-bitcoinlib channel
 28 2015-10-12 17:44:01 <cjcj> maaku: Ok, sorry
 29 2015-10-12 17:52:24 <instagibbs> what's the best way to check current mempool size?
 30 2015-10-12 17:53:18 <Apocalyptic> instagibbs, which mempool ?
 31 2015-10-12 17:53:29 <instagibbs> local bitcoind instance?
 32 2015-10-12 17:54:08 <Apocalyptic> the "getmempoolinfo" rpc command
 33 2015-10-12 17:54:34 <Apocalyptic> (also this is #bitcoin material)
 34 2015-10-12 17:55:00 <instagibbs> huh, must have been looking at old list, and sorry
 35 2015-10-12 17:55:17 <instagibbs> It's for PR testing, but agreed
 36 2015-10-12 18:57:58 <btcdrak> ping warren:
 37 2015-10-12 18:59:51 <warren> btcdrak: let's see who is here.  I forgot this is a national holiday of 2+ countries when this date was set...
 38 2015-10-12 19:00:00 <warren> if we don't have enough people we need to reschedule
 39 2015-10-12 19:00:37 <btcdrak> ping jcorgan: jgarzik: wumpus:
 40 2015-10-12 19:01:24 <btcdrak> warren: what is to be discussed?
 41 2015-10-12 19:01:38 <warren> btcdrak: set policy for dev/discuss lists
 42 2015-10-12 19:02:27 <btcdrak> might have been better if we'd sent a heads up reminder earlier today
 43 2015-10-12 19:02:40 <wumpus> warren actually did
 44 2015-10-12 19:03:03 <jcorgan> here
 45 2015-10-12 19:03:45 <btcdrak> oops
 46 2015-10-12 19:05:40 <warren> I move that we reschedule this meeting.  How about Thursday, one hour before the other meeting?
 47 2015-10-12 19:05:58 <wumpus> who do we need that is missing?
 48 2015-10-12 19:06:25 <btcdrak> I thought we had this pretty clear about list definitions already tbh
 49 2015-10-12 19:07:01 <morcos> i wasn't really planning on joining this meeting, but in my mind you haven't made it absolutely clear who the moderators of each list are.  are they the same?
 50 2015-10-12 19:07:01 <warren> jgarzik? morcos? petertodd? other people who were involved in list policy discussion.  I don't know, it just feels like we have too few people
 51 2015-10-12 19:07:23 <Luke-Jr> after the other meeting may be better
 52 2015-10-12 19:07:35 <warren> Luke-Jr: I personally have a meeting at that time
 53 2015-10-12 19:07:58 <jcorgan> you sent a meeting notice by email and i presume people accepted
 54 2015-10-12 19:08:02 <Luke-Jr> k, before should be fine too as long as someone notifies the ML of the time
 55 2015-10-12 19:08:09 <wumpus> it's usuallly impossible to schedule a meeting where everyone can be present, just plan a time and stick with it :)
 56 2015-10-12 19:08:19 <morcos> i think jgarzik is an important person to take the lead on this, he's a committer, he's viewed relatively neutrally with respect to this block size mess, and he's willing to put his foot down.  i'm hoping he'll take a bit of the lead here, at least on the bitcoin-dev list
 57 2015-10-12 19:08:24 <warren> I forgot today was a holiday
 58 2015-10-12 19:08:26 <warren> people are mising
 59 2015-10-12 19:08:30 <cfields> here
 60 2015-10-12 19:08:47 <btcdrak> cfields: here to talk about the new list policy
 61 2015-10-12 19:08:54 <cfields> roger
 62 2015-10-12 19:09:08 <warren> ok, who else other than jgarzik do folks think should be here?
 63 2015-10-12 19:09:38 <morcos> CodeShark had offered to moderate as well i think, but we have enough people if we have enough volunteers.
 64 2015-10-12 19:09:41 <btcdrak> warren: are we talking about the administrators of bitcoin-dev@ ?
 65 2015-10-12 19:10:16 <warren> ok, who will be the group of moderators?  how about we decide now who they are, and they decide on policy.
 66 2015-10-12 19:10:38 <morcos> to that point i'd say we start by splitting this in two.  if we have a couple of people who will take responsibility for one list and a couple for the other, then yes, agreed warren
 67 2015-10-12 19:10:52 <morcos> but is it the same moderators for each list, i'm still not clear on that.
 68 2015-10-12 19:11:04 <Luke-Jr> warren: eh, not sure policy should be decided by the mods..
 69 2015-10-12 19:11:34 <jcorgan> i'll be happy to mod the -dev list, but not the -discuss one
 70 2015-10-12 19:11:42 <btcdrak> warren: I think what is more important is to set what is on topic for  each list, much moire important than moderators.
 71 2015-10-12 19:12:08 <morcos> btcdrak: part of my point is many of us might not care what the topic is for the -discuss list
 72 2015-10-12 19:12:26 <morcos> i think the topic for the -dev list is relatively clear right?
 73 2015-10-12 19:12:34 <morcos> _technical_ development
 74 2015-10-12 19:12:39 <Luke-Jr> same as this channel, IMO
 75 2015-10-12 19:12:42 <btcdrak> morcos: I am of the view that anything goes on -discuss, but -dev should be quite specific
 76 2015-10-12 19:12:45 <Luke-Jr> except more strictly on-topic
 77 2015-10-12 19:12:54 <btcdrak> morcos: yes
 78 2015-10-12 19:13:29 <jcorgan> also, "anything goes" is not appropriate for -discuss, there is proper list etiquette and decorum for any list
 79 2015-10-12 19:13:58 <warren> OK, we don't need to decide on exactly who now
 80 2015-10-12 19:14:01 <Luke-Jr> jcorgan: well, "anything goes" in terms of topics accepted is not an invitation to violate proper etiquette
 81 2015-10-12 19:14:02 <btcdrak> So bitcoin-dev, technical discussion relating to bitcoin development and the protocol. bitcoin-discuss for general discussions about bitcoin
 82 2015-10-12 19:14:06 <warren> and it was a little short sighted to let them set policy
 83 2015-10-12 19:14:27 <warren> How do we do this better than last time?
 84 2015-10-12 19:14:29 <morcos> we need to try to decide on who
 85 2015-10-12 19:14:45 <warren> bitcoin and bistoin-test list existed at sourceforge.  nobody used it.
 86 2015-10-12 19:15:08 <Luke-Jr> IMO whoever is willing to try to be unbiased and enforce the policy
 87 2015-10-12 19:15:15 <btcdrak> usual "dont be a dick" rules apply for etiquette surely: i.e. no personal attacks on either lists, no doxxing, etc.
 88 2015-10-12 19:15:15 <morcos> agreed
 89 2015-10-12 19:15:27 <Luke-Jr> (preferably not people who have better things to spend time on)
 90 2015-10-12 19:16:13 <warren> https://lwn.net/Articles/659221/
 91 2015-10-12 19:16:30 <jcorgan> i do think that the -dev list serves to increase the productivity of actual developers, so anything that detracts from that goal should be frowned upon
 92 2015-10-12 19:17:07 <morcos> yikes
 93 2015-10-12 19:17:20 <warren> TLDR: Linux kernel mailing list has been known for being a brutal place.  Sarah Sharp @ Intel tried to improve the expected standard of respect between developers and ended up quitting.  Should we attempt to set a higher standard for our own dev list?
 94 2015-10-12 19:17:48 <morcos> i don't think we're at that problem right now.
 95 2015-10-12 19:17:50 <GreenIsMyPepper> wait is the meeting for -discuss now or is it moved to thurs?
 96 2015-10-12 19:18:06 <Luke-Jr> warren: there is such a thing as taking it too far also
 97 2015-10-12 19:18:06 <morcos> i think our problems are more basic
 98 2015-10-12 19:18:17 <btcdrak> yes. people like sipa have unsubscribed exactly because the list became a rabble.
 99 2015-10-12 19:18:23 <warren> GreenIsMyPepper: it's possible we're making substantive progress on deciding what we want, in addition to meeting to planning the next meeting
100 2015-10-12 19:18:28 <GreenIsMyPepper> ok
101 2015-10-12 19:18:41 <morcos> ok, so jcorgan volunteered for -dev, do you want to connect with jgarzik, and the two of you find a 3rd moderator if you want and then take it from there
102 2015-10-12 19:18:56 <jcorgan> it's not up to me
103 2015-10-12 19:19:01 <morcos> we can always un annoint you moderator if you suck
104 2015-10-12 19:19:05 <morcos> :)
105 2015-10-12 19:19:25 <morcos> i'm just trying to move us towards having a usable -dev list again
106 2015-10-12 19:19:33 <jcorgan> agree entirely.
107 2015-10-12 19:19:46 <warren> morcos: we do have more basic problems like on-topicness enforcement not happening yet, but I wonder if we should also use this opportunity to set expectations
108 2015-10-12 19:19:52 <warren> of professional conduct
109 2015-10-12 19:19:54 <morcos> once -discuss exists, and someone is willing to boot people off -dev if they don't willingly move to -discuss, thats the critical first step
110 2015-10-12 19:20:00 <morcos> we can improve the rest as we go
111 2015-10-12 19:20:02 <wumpus> warren: linux mailing list is different problem, uncomparable imo
112 2015-10-12 19:20:05 <jcorgan> my only issue is timeliness, and we should have multiple moderators to provide coverage
113 2015-10-12 19:20:11 <Luke-Jr> morcos: boot threads, not people
114 2015-10-12 19:20:24 <wumpus> problem with bitcon-dev is not so much that it's a 'brutal place', but that it's wildly off topic for development
115 2015-10-12 19:20:25 <stonecoldpat> the biggest problem with the list today is just personality attacks or steering off-topic
116 2015-10-12 19:20:30 <morcos> well thats a good question, do we boot people if they repeatedly violate the topic
117 2015-10-12 19:20:31 <stonecoldpat> (from my perspective)
118 2015-10-12 19:21:00 <jcorgan> i'm in favor of giving people exactly one opportunity to demonstrate they can change their behavior
119 2015-10-12 19:21:07 <Luke-Jr> morcos: if they ever contribute anything on-topic, we shouldn't boot them
120 2015-10-12 19:21:24 <btcdrak> warren: it's important we preserve the way the dev list used to be. A place not only for discussion about the bitcoin protocol, of BIPs but also a place where people would publish their technical papers. The list was, until recently a pretty reliable, and low traffic place others in the ecosystem could keep up to date of technical developments, and progress.
121 2015-10-12 19:21:33 <Luke-Jr> morcos: also, people should never be blocked from reading
122 2015-10-12 19:21:51 <wumpus> everyone can read the archives
123 2015-10-12 19:21:55 <morcos> Luke-Jr: there has to be some kind of consequnce. i think people should be warned (i'd have maybe said more than one strike) but then they are tempoarily banned if they can't change behavior.  its to protect the rest of us
124 2015-10-12 19:21:55 <warren> rusty mentioned that he uses a different approach for expectation setting on lightning-dev list, perhaps we should think about it for -dev list.
125 2015-10-12 19:22:31 <btcdrak> warren: what is rusty's method?
126 2015-10-12 19:22:39 <warren> rusty sets the moderation bit enabled by default for new subscribers.  Their first post gets stuck, he sends a personal message about expectation setting (on-topic rules, etc) and approves the message and unsets moderation bit.
127 2015-10-12 19:23:18 <jcorgan> that works well ime but is labor intensive
128 2015-10-12 19:23:22 <warren> That sounds OK-ish for a new list, might have drawbacks though.  Unpopular on-topic posts might be silenced entirely.
129 2015-10-12 19:24:18 <btcdrak> warren: I would not like that on bitcoin-dev. Rather we should do it the other way round. If someone does not conform to the wishes of the moderators, then they get the moderation bit set on their account
130 2015-10-12 19:24:25 <jcorgan> popularity/unpopularity of a topic should never be a criteria, only on-topicness and professional behavior
131 2015-10-12 19:24:27 <morcos> btcdrak: +1
132 2015-10-12 19:24:32 <morcos> jcorgan: +!
133 2015-10-12 19:25:27 <morcos> warren: what does it take to make -discuss live?
134 2015-10-12 19:25:32 <jcorgan> agree with btcdrak
135 2015-10-12 19:25:41 <morcos> even without a moderator, we can all yell at people to move there, once its live
136 2015-10-12 19:26:19 <btcdrak> as far as policy enforcement goes it should be clear we are only trying to raise the focus and quality of discussion on bitcoin-dev and provide a spillover for more general discussion and philosophy, but that good behaviour is expected on both lists.
137 2015-10-12 19:26:26 <bsm1175321> FWIW, bitcoin-dev has been fairly civil for the last week. It seems a certain combative poster has stopped or been banned. Open discussion forums can never be 100% signal...perhaps we don't need a new list. If some people learned to ignore bad behavior, is it enough?
138 2015-10-12 19:26:33 <btcdrak> morcos: -discuss is actually live already
139 2015-10-12 19:26:41 <GreenIsMyPepper> it might be necessary to have reasonable scope for -discuss though, egregiously bad topics should be discouraged entirely not have -discuss completely as a dumping ground
140 2015-10-12 19:27:02 <stonecoldpat> GreenIsMyPepper: +1, i wouldnt want -discuss becoming another reddit
141 2015-10-12 19:27:18 <jcorgan> yeah, i don't think -discuss should be view as spillover, but as a professional discussion list in its own right
142 2015-10-12 19:27:26 <btcdrak> GreenIsMyPepper: bitcoin discussion, not a support forum for sure, no advertising.
143 2015-10-12 19:27:38 <btcdrak> jcorgan: +1
144 2015-10-12 19:27:47 <btcdrak> but maybe not a place for technical discussion
145 2015-10-12 19:28:00 <morcos> yes and i think we also need moderators to have a strategy for "technical" discussion on -dev which just goes on and on in a back and forth between two people repeating the same thing.
146 2015-10-12 19:28:02 <warren> jcorgan: I'd like to make professional behavior required, but that's more subjective than on-topic
147 2015-10-12 19:28:05 <warren> ness
148 2015-10-12 19:28:19 <warren> morcos: -discuss list already exists
149 2015-10-12 19:28:56 <btcdrak> warren: we can certainly specify what is considered unprofessional (I already gave some examples above)
150 2015-10-12 19:29:21 <morcos> i'd think it makes sense for moderators to issue private reminders to people, that although their thread may be on-topic'ish they may want to consider taking it off-list if it doesn't seem like something everyone wants to read.  they'll have to use their judgement of when things have degenerated
151 2015-10-12 19:29:57 <jcorgan> there is also the point morcos brought up, which is how to terminate a thread that has exhausted its potential and is just going round and round with out progress
152 2015-10-12 19:30:07 <warren> Another problem: Mailman 2 currently has a shared admin and shared moderator password.  Mailman 3 will have separate accounts for individual admins/mods, but it will be ~6 months before they are ready to upgrade it.
153 2015-10-12 19:30:12 <jcorgan> that's a much more subjective call
154 2015-10-12 19:30:22 <Luke-Jr> morcos: blocking the offending messages seems sufficient to "protect" most people
155 2015-10-12 19:31:34 <stonecoldpat> is it possible to post blocked posts to an archive site? so at least people can see what is being blocked?
156 2015-10-12 19:31:37 <jcorgan> warren: the password situation isn't ideal, but do you see a real issue?
157 2015-10-12 19:32:02 <stonecoldpat> that way the list isnt censored as such, just access to our inboxes is restriced
158 2015-10-12 19:33:08 <warren> jcorgan: need to be careful about who is a mod/admin as if someone behaves poorly it could be hard to figure out who did it?
159 2015-10-12 19:33:11 <jcorgan> stonecoldpat: in the long past that was what killfiles did
160 2015-10-12 19:34:10 <jcorgan> warren: hmm.  "if put into a position of trust by the community and allowed to be a list moderator, i hereby promise to never take any actions without identifying myself as the actor, if asked"
161 2015-10-12 19:34:44 <btcdrak> warren: we can start small and see how things go
162 2015-10-12 19:34:57 <stonecoldpat> jcorgan: im not familiar with killfiles (googled and it appears to be a virus? i doubt it), but how did that approach go generally?
163 2015-10-12 19:35:08 <btcdrak> I expect the worst of the problems are over anyhow.
164 2015-10-12 19:35:28 <jcorgan> it was a filter you could use that would redirect mail from certain recipients to /dev/null
165 2015-10-12 19:36:03 <warren> Other problem: While it would be a major improvement to shunt non-dev discussion from -dev list, I don't think we're being honest about -discuss list.
166 2015-10-12 19:36:04 <jcorgan> same purpose and semantics as /ignore here on irc
167 2015-10-12 19:36:28 <btcdrak> stonecoldpat: we have the limitations of mailman. If people get moderated it will be because they have behaved badly and refused to a bait
168 2015-10-12 19:36:34 <warren> Let's be real, do smart people really want to educate people on -discuss list?
169 2015-10-12 19:36:36 <warren> Will you do it?
170 2015-10-12 19:36:42 <warren> If not, this isn't honest, IMHO.
171 2015-10-12 19:36:57 <GreenIsMyPepper> i think it'd be more interesting to frame it as a discussion about meta bitcoin topics. i.e. not reddit. perhaps this can be reframed by calling the list bitcoin-meta, names are important..
172 2015-10-12 19:37:21 <warren> A better approach might be to collaboratively work on a FAQ so the educative function is better automated.
173 2015-10-12 19:38:08 <GreenIsMyPepper> (but i concede that's a little bit aggressive in limiting the scope of the list)
174 2015-10-12 19:38:14 <jcorgan> that might be true, but -discuss should serve more than just "educating people"
175 2015-10-12 19:38:30 <btcdrak> warren: -discuss should not be a place to educate others or give support. it's definitely not the purpose of the list. The purpose is to be a place for the non-technical/protocol discussions/philosophy debates.
176 2015-10-12 19:39:31 <warren> btcdrak: good luck with that
177 2015-10-12 19:39:39 <warren> you can state it
178 2015-10-12 19:39:56 <jcorgan> i see mailing lists as a tool to increase productivity.  for -dev, it is to increase the productivity of developers with development related discussions.  for bitcoin users -discuss should help them be more productive with using bitcoin, whether that be help or debating philosophy
179 2015-10-12 19:40:40 <jcorgan> #bitcoin used to be like that
180 2015-10-12 19:40:43 <warren> Who is willing to put time into making the -discuss list actually used?
181 2015-10-12 19:40:59 <warren> I don't see this as particularly honest if the smart people ignore it entirely
182 2015-10-12 19:41:04 <jcorgan> what does that actually mean?
183 2015-10-12 19:41:18 <jcorgan> of course there are smart people who aren't developers?
184 2015-10-12 19:41:31 <warren> telling people to "go there" where all the other garbage that is ignored is posted?
185 2015-10-12 19:41:44 <wumpus> smart people can pay attention to it if they like, no way to force them
186 2015-10-12 19:41:50 <warren> These non-dev lists have existed before, nobody used it.
187 2015-10-12 19:42:07 <warren> that being said, we don't have a better option, so we do it.
188 2015-10-12 19:42:27 <morcos> warren: why is it the responsibility of people that want a highly focused technical list to address the unknown needs of other people who may want to discuss something else
189 2015-10-12 19:42:37 <wumpus> and people can talk with each other there... I don't see it as realistic that everything would be ignored
190 2015-10-12 19:42:54 <stonecoldpat> warren: i dont think you need the smartest people in a -discuss, and id be surprised if a subset of -dev did not follow it anyway
191 2015-10-12 19:42:59 <warren> morcos: technically it isn't, I'm just pointing out that trying to state a purpose for the other list is kind of over optimistic
192 2015-10-12 19:43:34 <warren> Let's go back to the action items.
193 2015-10-12 19:43:41 <jcorgan> i don't think you make technical discussions off-topic in bitcoin, people should be free to engage that way there
194 2015-10-12 19:43:59 <warren> jcorgan: huh?  nobody suggested that
195 2015-10-12 19:44:15 <jcorgan> that didn't come out right
196 2015-10-12 19:44:29 <jcorgan> i mean, -discuss shouldn't be assumed to be "less technical"
197 2015-10-12 19:44:40 <warren> Who were the moderators that we agreed upon for -dev list?
198 2015-10-12 19:44:51 <stonecoldpat> for the -dev, it may be good to propose ways to post to the list, i.e. professional, quoting parts of the proposal that you want to discuss, etc. Some of the e-mails that make the list aren't always clear what exactly they are discussing in the proposal
199 2015-10-12 19:45:23 <GreenIsMyPepper> jcorgan: i agree, that's a very important point, there especially shouldn't be a "move this to -dev"
200 2015-10-12 19:46:04 <GreenIsMyPepper> unless it was proposing changes to bitcoin/code etc
201 2015-10-12 19:46:44 <jcorgan> i'm willing to help -discuss become what we're suggesting here, i just may not have the timeliness to deal with an expected high volume of traffic, and would expect there to be multiple mods
202 2015-10-12 19:47:06 <jcorgan> surely i'm not the only masochist here
203 2015-10-12 19:47:20 <stonecoldpat> jcorgan: i'm willing to help either as well, so you won't be alone :)
204 2015-10-12 19:47:22 <GreenIsMyPepper> i'm also willing to help, but with the caveat that i only check email twice daily
205 2015-10-12 19:47:34 <warren> who is stonecoldpat, btw?
206 2015-10-12 19:47:46 <morcos> warren: i proposed jcorgan, who volunteered, jgarzik, who isn't heare, but volunteered earlier, and some unnamed 3rd that the two of them found.  for -dev mods
207 2015-10-12 19:47:54 <jcorgan> a diversity of trusted moderators with overlapping styles and time zones would be a good thing
208 2015-10-12 19:48:01 <stonecoldpat> warren: i'm a phd student @ newcastle uni
209 2015-10-12 19:48:42 <warren> I nominated GreenIsMyPepper as mod, mainly because he was there at that moment and seemed like a good idea in my mind.
210 2015-10-12 19:50:32 <jcorgan> i'm not wedded to the idea, happy to help out if i can, but it's really up to you guys
211 2015-10-12 19:51:01 <warren> wumpus: would it be reasonable to ask the project leader to make a decision here?
212 2015-10-12 19:53:27 <btcdrak> I think some input from wumpus is essential.
213 2015-10-12 19:54:17 <warren> OK, the REAL meeting is Thursday, one hour prior to the other meeting.
214 2015-10-12 19:54:35 <btcdrak> golly, a whole hour before?
215 2015-10-12 19:55:01 <wumpus> what do you need me to say here?
216 2015-10-12 19:55:18 <jcorgan> i might miss the beginning, but said mostly what i wanted to say here already
217 2015-10-12 19:55:27 <wumpus> I like the bitcoin-discuss idea, have been in favor of splitting up the mailing list from the beginning
218 2015-10-12 19:56:25 <warren> wumpus: 1) decide on who are the -dev moderators 2) decide on list policy, or delegate who drafts the policy 3) decide on who approves the list policy
219 2015-10-12 19:56:35 <wumpus> I have no time or inclination to play babysitter though
220 2015-10-12 19:57:18 <warren> I proposed to LF that they host a simple static website for the lists, which will 1) list all the lists 2) host the policy 3) allow github pull requests to update it
221 2015-10-12 19:57:31 <wumpus> I think we should just pick some people for moderators that are interested in this and bother to be here
222 2015-10-12 19:57:45 <cfields> warren: i don't think those burdens fall on a particular implementation's maintainer
223 2015-10-12 19:58:32 <warren> cfields: then somebody make a decision, or whoever bothers to show up on Thursday votes on ratifying that decision
224 2015-10-12 19:59:17 <wumpus> as for list policy, that's not really my thing, would prefer if someone else writes that
225 2015-10-12 20:00:05 <warren> I need to move on to <another meeting>, we talk about this again 1 hour prior to the Thursday meeting.
226 2015-10-12 20:00:14 <warren> I don't know that time in UTC.
227 2015-10-12 20:01:30 <morcos> sigh... jcorgan and jgarzik as initial -dev mods.  GreenIsMyPepper is welcome to join if he wants
228 2015-10-12 20:01:46 <btcdrak> yeah fine with me
229 2015-10-12 20:01:47 <morcos> all in favor.  aye
230 2015-10-12 20:01:51 <btcdrak> aye
231 2015-10-12 20:02:15 <wumpus> yeah
232 2015-10-12 20:02:26 <GreenIsMyPepper> warren: 18:00 UTC / 11:00 Pacific
233 2015-10-12 20:02:56 <warren> ad patches to several mailing lists was met with courtesy and helpful advice, in spite of serially repeating the behaviour."
234 2015-10-12 20:02:56 <warren> https://lwn.net/Articles/659884/  SCSI subsystem maintainer James Bottomley has posted a different view on the issue of civility on the kernel's mailing lists. "So, by and large, I’m proud of the achievements we’ve made in civility and the way we have improved over the years. Are we perfect? by no means (but then perfection in such a large community isn’t a realistic goal). However, we have passed our stress test: that an individual with b
235 2015-10-12 20:03:38 <warren> Hmm, OK, I guess if we did focus the topic on -dev list, we aren't nearly as brutal as lkml.
236 2015-10-12 20:04:23 <GreenIsMyPepper> morcos: sure / aye
237 2015-10-12 20:05:04 <jcorgan> heh, i don't think lkml gets much in the way of "the purpose of linux is <my specific ideological hobby horse> and you must be evil to want to change it".
238 2015-10-12 20:05:08 <jcorgan> or maybe it does :-)
239 2015-10-12 20:05:19 <btcdrak> warren: you know it might be a good idea to ask sipa what kind of conditions would make him more willing to rejoin the -dev list again. Frankly the list have been severely neutered with key people no longer reading it.
240 2015-10-12 20:05:22 <kanzure> jcorgan: before linux there was already pre-existing understanding of what kernels were
241 2015-10-12 20:06:11 <warren> btcdrak: ok, I'll ask
242 2015-10-12 20:06:13 <morcos> btcdrak: i agree we want people like sipa to rejoin, but i think we all know what it takes. lets not put more responsibility on their plates.  if he doesn't rejoin after these guys clean up the list
243 2015-10-12 20:06:22 <morcos> then we can say, what haven't we done yet
244 2015-10-12 20:06:37 <GreenIsMyPepper> jcorgan: you mean comp.os.minix? :^)
245 2015-10-12 20:06:55 <btcdrak> morcos: for me personally, there's no point in the dev list if key people dont read it. My priority is to try fix that.
246 2015-10-12 20:07:27 <warren> jcorgan: btcdrak: are your GPG keys well connected?  need to know where/how to send the passwords securely
247 2015-10-12 20:07:29 <kanzure> for the non-dev mailing list, if you want it to stick and hold, then i recommend getting a few non-development-related emails from the recent archives, and queue them up to seed the other mailing list
248 2015-10-12 20:07:38 <jcorgan> not really
249 2015-10-12 20:07:40 <kanzure> this way there is some amount of activity that you can point to, showing tha the mailing list is not dead
250 2015-10-12 20:07:42 <warren> I'll separately send the list admin passwords to wumpus
251 2015-10-12 20:07:46 <btcdrak> warren: mine is signed by petertodd
252 2015-10-12 20:07:56 <morcos> btcdrak: mine too!  thats why i'm trying to move this towards action instead of talking.  sipa doesn't want to be responsible for determining the topic and mods of the list (at least thats my guess).  nobody wants to make decisions for everyone else
253 2015-10-12 20:07:59 <bsm1175321> Sipa has little patience. That's his problem. I don't think engineering a list to his standards is a reasonable goal. We all have to learn to ignore a certain amount of noise, and not engage the trolls.
254 2015-10-12 20:08:28 <morcos> bsm1175321: that's unfair, but we've gotten off topic
255 2015-10-12 20:08:29 <kanzure> part of the problem is that none of the developers are extremely interested in a non-dev mailing list, and they tend to be the one writing the most emails anyway, so... anyway, taking action like i proposed would be necessary to convince others to actually use another mailing list (unless you want to use the banhammer a lot, which is fine, but exhausting)
256 2015-10-12 20:08:33 <btcdrak> morcos: right, but I mean if we understand his criterion (that make him comfortably being on the list) it helps us make decisions.
257 2015-10-12 20:08:33 <wumpus> it's not about him specifically
258 2015-10-12 20:08:51 <CodeShark> Just got here and caught up with the convo...
259 2015-10-12 20:08:54 <bsm1175321> +1 kanzure.
260 2015-10-12 20:09:08 <bsm1175321> I don't want any part of bitcoin-discuss.
261 2015-10-12 20:09:30 <kanzure> you can use pre-existing email from bitcoin-dev mailing list recent history to seed the other mailing list
262 2015-10-12 20:09:39 <kanzure> you can either replay the emails or just put them in the archive (i dunno which would be better)
263 2015-10-12 20:10:13 <wumpus> but telling people to just 'we have to learn to ignore noise' is unreasonable, not everyone has time to track a high-traffic mailing list with lots of non dev related talk to find e.g. BIP proposals
264 2015-10-12 20:10:32 <CodeShark> the idea of -discuss arose in a very specific context...one of discord and acrimony between core devs
265 2015-10-12 20:10:42 <bsm1175321> How many times has lkml forked and reformed because of the noise there?
266 2015-10-12 20:10:47 <warren> it's usually very rude to auto-subscribe people to a discussion list
267 2015-10-12 20:10:58 <btcdrak> wumpus: I agree with you. And the -dev list never used to be this rowdy. it only started this the blitz
268 2015-10-12 20:10:59 <wumpus> lkml doesn't have that much noise at all, it's all about linux kernel development
269 2015-10-12 20:11:11 <kanzure> in the future we will have far more developers that are completely unfamiliar with how communication in open-source projects tends to work
270 2015-10-12 20:11:19 <CodeShark> If not for that, there'd be no pushbash to just having little tolerance to nontech talk in -dev
271 2015-10-12 20:11:30 <bsm1175321> wumpus: in the early days it had a lot more noise.  But nose makers were ignored, and the list became more technical.
272 2015-10-12 20:11:33 <wumpus> if you start a topic there about the e.g. economics of linux you're going to have trouble very soon
273 2015-10-12 20:11:44 <warren> bsm1175321: lkml has sub-lists for focused discussion of certain things prior to it being ready for lkml
274 2015-10-12 20:11:54 <kanzure> you should nominate not just a moderator for the -discuss mailing list, but someone who will actively source contributions and tend to that garden (in a productive way). nobody wants to subscribe to a "send cypherpunk spam here" mailing list.
275 2015-10-12 20:12:12 <morcos> what actually has to happen to give jcorgan, jgarzik and GreenIsMyPepper mod powers for -dev?
276 2015-10-12 20:12:18 <btcdrak> bsm1175321: ignoring noise is just not a reasonable option, sorry.
277 2015-10-12 20:12:25 <warren> kanzure: that was my point earlier ...
278 2015-10-12 20:12:40 <wumpus> btcdrak: problem is also that noise amplifies noise, people that want to make noise reply to other people that want to make noise
279 2015-10-12 20:12:53 <wumpus> not all problems go away by ignoring them
280 2015-10-12 20:12:59 <warren> morcos: sending them the password, and someone needs to draft the policy, we need to ratify it and post it
281 2015-10-12 20:13:32 <btcdrak> wumpus: and then the noise buries the on topic replies in a thread. Becomes impossible, I agree.
282 2015-10-12 20:13:53 <morcos> can we send them the password?  lets not let the perfect be enemy of the good here.  once those 3 people feel like its their responsibility to clean up the list, we'll be on our way
283 2015-10-12 20:14:13 <wumpus> agree morcos
284 2015-10-12 20:14:16 <bsm1175321> +1 morcos
285 2015-10-12 20:14:28 <kanzure> how does lkml handle large quantities of content? surely there's >1k emails/day generated about kernel development.
286 2015-10-12 20:14:40 <kanzure> i don't think they all get sent to lkml. so where do they go?
287 2015-10-12 20:14:47 <btcdrak> morcos: yes. especially because the new list is already active to
288 2015-10-12 20:16:16 <kanzure> btcdrak: the number of complaints about signal/noise is only going to increase in the future. some individuals are completely opposed to any noise whatsoever, even if they would miss high signal items by accidently classifying those contributions as noise. not sure what to do about this long-term.
289 2015-10-12 20:17:20 <btcdrak> kanzure: we can improve the quality of the list very easily.
290 2015-10-12 20:17:36 <CodeShark> the situation would be a lot less delicate were it not for the bitterness that arose between some core developers...and all this ultimately is a proxy battle, imho
291 2015-10-12 20:18:02 <kanzure> i have noticed that many complaints about noise have been entirely unrelated to xt stuff
292 2015-10-12 20:18:15 <btcdrak> CodeShark: there is no bitterness between core developers. There is bitterness from protagonists against people who do not share their views.
293 2015-10-12 20:18:44 <helo> bitcoin is prime bikeshed land
294 2015-10-12 20:19:23 <wumpus> helo: yes, it's as if no one ever agrees on anything
295 2015-10-12 20:19:25 <bsm1175321> We are never going to agree on an appropriate signal/noise ratio, nor what constitutes signal or noise. Once upon a time this was handled personally using a kill file. The less time spent arguing over this, the better. So let's have those 3 share a password for now, and move on. I'll be happy to point people to procmail tutorials so they can dump offensive posters.
296 2015-10-12 20:19:26 <kanzure> wumpus: some people seem to be better at reading high-volume traffic. i suggested a while back the concept of emailing summaries to the development list. individuals who can't read the noise would instead prefer to read from people summarizing recent developments.
297 2015-10-12 20:20:04 <kanzure> killfiles are a good way to block totally unreliable sources of noise, but not a good way when you just want to ignore random "lol" comments from an otherwise good contributor
298 2015-10-12 20:20:36 <wumpus> helo: difficult proposals sometimes go unheeded, but every little detail needs to be discussed to death
299 2015-10-12 20:20:38 <btcdrak> warren: are we done for this meeting?
300 2015-10-12 20:20:42 <kanzure> one of the tests that i use in my head is "will this solution work if there were 20,000 developers? what about 50,000?" and the answer is usually no.
301 2015-10-12 20:20:54 <bsm1175321> kanzure: I've always relied on lwn.net for summaries of lkml development.  I love it. I wish someone would start a similar site for bitcoin.
302 2015-10-12 20:21:11 <kanzure> bsm1175321: actually i was thinking of summaries slightly shorter than that, but point taken.
303 2015-10-12 20:21:27 <btcdrak> bsm1175321: someone started doing summaries of the weekly IRC meetings, they are pretty good.
304 2015-10-12 20:21:34 <GreenIsMyPepper> so far the summaries of the #bitcoin-dev IRC meeting minutes have been excellent
305 2015-10-12 20:21:58 <bsm1175321> They are good.  Here in NYC we do a biweekly meetup (BitDevsNYC) where we go over recent developments, but it doesn't make it onto the intertubes.
306 2015-10-12 20:23:05 <kanzure> is there any particular individual that anyone could recommend for proactively contributing to -discuss? e.g. like congress whip.
307 2015-10-12 20:23:25 <kanzure> oh wait, no, wrong analogy
308 2015-10-12 20:25:20 <bsm1175321> For reference, this is what Jonathan Corbet compiles from lkml on a weekly basis.  Would anybody here be willing to blog or summarize -dev and post to -discuss? http://lwn.net/Articles/658288/
309 2015-10-12 20:25:34 <kanzure> summaries of -dev should be sent to -dev
310 2015-10-12 20:26:07 <bsm1175321> kanzure: the point is to move some discussion away from -dev.  ;-)
311 2015-10-12 20:26:27 <kanzure> that's slightly more thorough than i was aiming for, i was actually thinking just two-four sentences per topic, then a link to where the contribution is from. but a few paragraphs per topic is fine with me, i don't care.
312 2015-10-12 20:26:38 <kanzure> bsm1175321: i think the idea is to keep development discussion on -dev.
313 2015-10-12 20:27:55 <bsm1175321> There's a set of people who don't/won't/can't contribute to development who still want to read about it. Having a reasonable place for them to discuss, without the temptation of the "reply" button would be nice.
314 2015-10-12 20:28:33 <kanzure> i don't understand
315 2015-10-12 20:29:05 <bsm1175321> ?
316 2015-10-12 20:34:44 <helo> summarizing the list would be quite an undertaking, but i bet it would help the quality of the list (particularly if comments were enabled) in the long run
317 2015-10-12 20:38:22 <kanzure> yes i should clarify that i don't mean comprehensive summaries, but rather summaries of whatever you happen to be interested in
318 2015-10-12 20:39:15 <kanzure> high quality buckets of related content, some that might have been from a week or two ago, is much more valuable than a list of "here are all of the emails that were received this week" :-)
319 2015-10-12 20:40:08 <kanzure> and also, doesn't have to be weekly; progress in privacy, anonymity, cryptography, etc. doesn't often happen on a regular schedule.
320 2015-10-12 20:41:56 <kanzure> for example, https://lists.linuxfoundation.org/pipermail/bitcoin-dev/2015-August/010759.html
321 2015-10-12 20:47:10 <jcorgan> there's someone doing something like that now for the weekly meeting, i imagine it is useful
322 2015-10-12 20:51:41 <jgarzik> btcdrak, sorry, couldn't physically be in IRC anytime sooner :(
323 2015-10-12 20:51:53 <jgarzik> warren, ^
324 2015-10-12 20:51:58 <btcdrak> jgarzik: np. check the logs
325 2015-10-12 21:06:55 <morcos> jgarzik: i volunteered you.  :)