1 2015-09-21 01:09:04 <warren> I am disappointed by the discourse on bitcoin-dev list straying so far from technical discussion.  Perhaps we should again discuss enforcement of an on-topic policy.
  2 2015-09-21 01:38:36 <jgarzik> warren, It is sufficient for a list admin to jump in and say "keep it technical" etc.
  3 2015-09-21 04:52:17 <warren> slas
  4 2015-09-21 06:36:17 <jonasschnelli> warren: I really would like a list admin step up and stop such discussions like we see in thread "[bitcoin-dev] Scaling Bitcoin conference micro-report"
  5 2015-09-21 06:36:47 <jonasschnelli> warren: Would it be to harsh to warn people serval times, if they won't keep it technical, unsubscribe them?
  6 2015-09-21 06:48:02 <CodeShark> jonasschnelli: we need a forum in which to discuss these matters - so if bitcoin-dev isn't the place, we should create a separate mailing list for it. banning people for discussing these things is detrimental to PR
  7 2015-09-21 06:48:51 <CodeShark> I'd also like to see bitcoin-dev be more technical...however, I don't think the fundamental disagreements right now are really technical
  8 2015-09-21 06:49:50 <CodeShark> a mistake we've made is treating the disagreement as technical...which has just led to more arguments over numbers pulled out of a hat using models that are not properly specified and assumptions over design goals that are often glossed over or ignored
  9 2015-09-21 06:49:53 <jonasschnelli> CodeShark: Agreed. Maybe we should keep bitcoin-dev as it is (more political) and introduce new mailing lists for serval tech topics (wallets, core-dev, etc.)?
 10 2015-09-21 06:51:05 <CodeShark> perhaps that would be the path of least resistance
 11 2015-09-21 06:52:00 <CodeShark> it's far easier to get the more technically-inclined people to sign up to a new mailing list than vice versa
 12 2015-09-21 06:52:20 <CodeShark> and we could set new rules from the get go
 13 2015-09-21 06:59:05 <CodeShark> regardless, it's clear we should separate these concerns and address them as such - we can't let these discussions slow down technical developments that are entirely orthogonal to the political disagreements
 14 2015-09-21 07:07:08 <CodeShark> it's also clear that if we don't provide a good forum for the political discussion that's well moderated yet inclusive, people are likely to end up trolling reddit and such
 15 2015-09-21 07:16:33 <CodeShark> I really want to see a good-faith effort to resolve these disagreements in an amicable, civilized way...and ultimately get good buy-in from the entire community for whatever vision we ultimately put forth
 16 2015-09-21 07:18:15 <CodeShark> arguing over metrics and numbers is unlikely to achieve this right now :)
 17 2015-09-21 07:19:11 <CodeShark> such arguments are generally only productive once we have agreed upon basic goals
 18 2015-09-21 07:27:58 <moa> agreement on basic goals could be difficult given the diverse user groups
 19 2015-09-21 07:29:07 <CodeShark> there will have to be some give and take...some compromise, etc... But the compromises shouldn't be over technical parameters
 20 2015-09-21 07:30:57 <CodeShark> the technical parameters should be based on scientific study of the problem given specific design goals
 21 2015-09-21 07:35:19 <CodeShark> i.e. you can decide to build a passenger jet or build a fighter jet - but it would be ridiculous to compromise by, say, going half-way on wingspan or fuel tank capacity
 22 2015-09-21 07:37:09 <moa> so we need some design specs?
 23 2015-09-21 07:37:35 <CodeShark> we need to resolve the issue of what it is we're really trying to build
 24 2015-09-21 07:37:47 <moa> where's the architect?
 25 2015-09-21 07:39:06 <moa> gotta go, bbl
 26 2015-09-21 07:39:44 <CodeShark> i.e. is it more important to grow userbase short-term even at the expense of long-term infrastructure? or is it more imporant to fix infrastructure issues even if it temporarily complicates usability?
 27 2015-09-21 07:39:52 <phantomcircuit> CodeShark, there seems to be a genuine widespread failure to understand bitcoins security model
 28 2015-09-21 07:39:55 <phantomcircuit> which is sad
 29 2015-09-21 07:40:51 <CodeShark> then we've done a poor job of communicating it
 30 2015-09-21 07:41:23 <phantomcircuit> i agree
 31 2015-09-21 07:43:47 <CodeShark> I'd also add that the way we communicate it to the technical community isn't necessarily the best way we communicate it to the nontechnical community
 32 2015-09-21 07:44:24 <CodeShark> the nontechnical community isn't really interested in big-O notation and other such crap (although to us engineers it is crucial)
 33 2015-09-21 07:45:46 <CodeShark> the argument with the nontechnical community has a lot more to do with the kinds of real-world problems this technology can solve
 34 2015-09-21 07:47:18 <phantomcircuit> CodeShark, comically i've had conversations with bank type people who have a much better understanding of the security model than most bitcoin space people do
 35 2015-09-21 07:47:29 <dcousens> CodeShark: wrong channel imho
 36 2015-09-21 07:47:44 <CodeShark> what's the right channel?
 37 2015-09-21 07:48:16 <dcousens> #bitcoin , this channel is typically for discussions based around the code
 38 2015-09-21 07:50:11 <phantomcircuit> dcousens, conversation is an extension of the questions around moderating the mailing list
 39 2015-09-21 07:50:27 <phantomcircuit> it's probably ot now but it took a windy path from on topic to off
 40 2015-09-21 07:50:54 <phantomcircuit> also ever since montreal im tired at reasonable hours
 41 2015-09-21 07:50:58 <phantomcircuit> so bye
 42 2015-09-21 07:51:00 <dcousens> phantomcircuit: just parotting what I've seen moderated in here before. :)
 43 2015-09-21 07:52:03 <CodeShark> I disagree, dcousens...this forum consists of probably the people who best understand what this technology is and can do...and communicating these things to the public is at least as important as writing code
 44 2015-09-21 07:52:52 <CodeShark> perhaps we should have a separate channel for this...but right now I'm not sure we do
 45 2015-09-21 07:53:02 <CodeShark> #bitcoin is not that channel
 46 2015-09-21 07:56:31 <dcousens> CodeShark: I'll have to defer that judgement to midnightmagic or the likes,  but it was my understanding this channel was mostly for technical discussion in regards to the software,  not necessarily how the developers should 'be better' at PR
 47 2015-09-21 07:57:30 <CodeShark> I'm fine with moving it elsewhere - but what I'm not fine with is avoiding this and then falling back on technical discussion over numbers we just invent in the guise of being technical
 48 2015-09-21 08:00:44 <CodeShark> given our privileged position in our understanding of this tech I believe it is our responsibility to do a better job of communicating it to the public. if we shirk this responsibility we'll just continue suffering intrusions
 49 2015-09-21 08:02:19 <jcorgan> i think this discussion thread is both insightful and important, and would like to see more of these qualities in #bitcoin.  as a mod, though, i'd defer to the core devs that use this channel the most as to whether it is on topic here.
 50 2015-09-21 08:03:17 <dcousens> CodeShark: you may be right,  but,  typically that has very rarely been the case in open source software (in the sense the public will formulate such a strong opinion),  so,  you can't blame devs who just want to.. dev
 51 2015-09-21 08:05:39 <dcousens> typically software satisfaction (from a public perspective) directly relates to them moving with their feet,  in bitcoin's case,  that isn't so simple.  It becomes even more difficult since the only real communication is between developers and the public,  often times with no translation of context or domain knowledge
 52 2015-09-21 08:07:07 <CodeShark> indeed, dcousens - Bitcoin is in many ways exceptional. to complicate matters, it directly involves money
 53 2015-09-21 08:09:14 <CodeShark> other people's money...and people tend to have strong opinions when it comes to their money
 54 2015-09-21 08:10:13 <dcousens> CodeShark: again,  I think this is the wrong channel for this,  I think any proposed actions of this discussion would be relevant here,  but the discussion of the problem and how to solve it is probably better suited for #bitcoin.  Anyway,  I need to return to work :)
 55 2015-09-21 11:24:20 <Yoghur114> does a timetable or any sort of release schedule for the deployment of OP_CHECKLOCKTIMEVERIFY / BIP65 and OP_CHECKSEQUENCEVERIFY / BIP112 (along with BIP68 and BIP113) exist?
 56 2015-09-21 13:05:24 <kanzure> wumpus: in https://github.com/bitcoin/bitcoin/pull/6695 do you mean "libevent-based" instead of "libhttp-based"?
 57 2015-09-21 13:08:05 <wumpus> kanzure: yes, should be libevhttp
 58 2015-09-21 13:09:04 <wumpus> no clue what libhttp is, but we're not using it
 59 2015-09-21 16:08:30 <wumpus> BlueMatt: ./contrib/verify-commits/verify-commits.sh complains about two commits by you, which I don't understand as they're not merge commits
 60 2015-09-21 17:22:58 <warren> jgarzik: jonasschnelli: CodeShark: btcdrak: Regarding the suggestion to create a discuss list, sourceforge had one.  Nobody used it.  We could do it again, but we need a better plan.
 61 2015-09-21 17:47:54 <wumpus> there is a chance it may work better now, apparently there is a lot more interest in discussing various things. And it'd help to be able to tell people to discuss things on another list instead of just telling them to go away.
 62 2015-09-21 17:49:35 <jcorgan> wumpus: did you mean tue instead of thu in your email about meeting time
 63 2015-09-21 17:50:07 <wumpus> jcorgan: oh did I get it wrong?
 64 2015-09-21 17:51:14 <jcorgan> it looks like the doodle is tue 10/6 1900UTC
 65 2015-09-21 17:51:20 <warren> Who would be the mod/admin of the discuss list?
 66 2015-09-21 17:53:34 <wumpus> may be better to ask in #bitcoin, doesn't have to be a developer
 67 2015-09-21 17:54:27 <warren> I personally don't know who the long-time helpful #bitcoin persons are, do you?
 68 2015-09-21 17:56:46 <warren> LF just informed me that they are maybe weeks away from fixing bugs that would allow transitioning from Mailman 2 to 3.  The latter would be superior in several ways including the ability to have separate accounts (and passwords) for mods/admins.
 69 2015-09-21 17:58:14 <wumpus> jcorgan: looks like both Thu and Tue 19:00-20:00 are fully green
 70 2015-09-21 17:58:57 <wumpus> jcorgan: so I was not entirely crazy :)
 71 2015-09-21 17:59:13 <jcorgan> oic
 72 2015-09-21 17:59:56 <jcorgan> not that i have much to say about it, but i suggest you go ahead and declare a time
 73 2015-09-21 18:02:15 <wumpus> I posted the doodle again in case some other people still wanted to add their 'vote', if it's still the same tomorrow or so I'll declare a time
 74 2015-09-21 18:05:09 <morcos> wumpus: warren: i'm strongly in favor of making the technical list the new list.  i think it has much greater chance of working that way.  we're the ones motivated to have a more focused discussion.  who cares if the naming isn't ideal.  "development" isn't necesarily completely technical anyway
 75 2015-09-21 18:05:35 <jonasschnelli> bitcoin-dev-non-political@?
 76 2015-09-21 18:05:44 <jonasschnelli> (or similar)
 77 2015-09-21 18:06:15 <wumpus> yes, that sounds more realistic
 78 2015-09-21 18:06:48 <morcos> bitcoin-codereview or something boring, and you're required to review 4 PR's a month to stay on the list. :)
 79 2015-09-21 18:06:57 <jonasschnelli> haha
 80 2015-09-21 18:07:04 <jonasschnelli> PoW
 81 2015-09-21 18:07:09 <morcos> jonasschnelli has lifetime membership i guess
 82 2015-09-21 18:07:34 <Lightsword> wondering if anyone has any thoughts on if I should modify my stratum server to support multiple active GBT connections? would it make any significant improvements on block changeovers?
 83 2015-09-21 18:09:24 <moa> make it technically difficult to access, like a deliberately raw or archaic tech. messaging system
 84 2015-09-21 18:09:33 <moa> cli gpg
 85 2015-09-21 18:10:20 <moa> requires a special client that needs building from source ...
 86 2015-09-21 18:24:01 <btcdrak> list as much as they should, and that even includes people like gmaxwell. If we create a bitcoin-discuss list at linuxfoundation then we have an sanctioned go to place for the more general and philosophical bitcoin discussion. Thats a big improvement especially when we can shepherd non dev discussions on to it.
 87 2015-09-21 18:24:01 <btcdrak> the dev list however, all we can do is ask people to stop, and they have nowhere else to go. That's not very satisfactory, but then allowing those conversations to range on the dev list means the list becomes unusable for those who are only interested in dev based discussions. We cant continue this way - I know of several people who are just not reading the
 88 2015-09-21 18:24:01 <btcdrak> warren: wumpus: jgarzik: jonasschnelli: CodeShark: its really simple. The past does not matter, whether the old sf bitcoin list was used or not, what is important is the situation right now. In IRC we are able to segregate -dev specific and non specific bitcoin discussions which improves the experience for all participants. When conversations get offtopic on
 89 2015-09-21 18:24:26 <CodeShark> [11:10] (moa) requires a special client that needs building from source ... <--- hah
 90 2015-09-21 18:24:32 <btcdrak> bitcoin-discuss is nice and general
 91 2015-09-21 18:32:17 <CodeShark> btcdrak: I'm generally for such an approach if people actually go there. Quite honestly, I think we should probably stop trying to argue positions on technical grounds over political disagreements...and those interested in resolving these disputes should be working on better communicating stuff to the general public.
 92 2015-09-21 18:34:07 <CodeShark> so I don't want this to be seen as a place we just dump discussion "we're not interested in" because it is of crucial importance and we need highly competent people doing it
 93 2015-09-21 18:34:39 <wumpus> the problem is that highly competent people also have zillions of other things to do but argue
 94 2015-09-21 18:35:29 <CodeShark> Yes, but this is an important job function, too
 95 2015-09-21 18:35:40 <CodeShark> Not arguing so much
 96 2015-09-21 18:35:47 <CodeShark> Communicating
 97 2015-09-21 18:36:08 <wumpus> other things shouldn't grind to a halt because of political issues, and if key people are unsubscribing from the mailing list that's not good
 98 2015-09-21 18:37:03 <warren> btcdrak: post a proposal on the list, including a plan where experts would actually educate on -discuss list.  It is a wholly dishonest measure to shunt less technical discussion onto a separate list if there is nobody there to educate.
 99 2015-09-21 18:37:58 <richardkiss> is there a #bitcoin-lightning-dev or something similar?
100 2015-09-21 18:38:02 <btcdrak> CodeShark: you're worrying too much. the point is to bring back focus. Techncial people are not reading the -dev list much these days or skimming and thats resulted if a few lost conversations already for important stuff.
101 2015-09-21 18:38:04 <wumpus> you can disagree with them, but that doesn't change it - it's no one's job to be on that mailing list, it's a tool, if people feel it's no longer useful for them they'll just stop following it
102 2015-09-21 18:40:52 <CodeShark> I used to have that attitude at earlier jobs - focus on real dev work...avoid distractions with stupid politics...and what ended up usually happening is I would end up having to do stuff in ways I knew was crappy to satisfy expectations from people who were not particularly that bright. This is why now I'm so adamant that setting expectations (even with nontechnical people) is so important to having the freedom to do dev the way
103 2015-09-21 18:41:17 <btcdrak> warren: there is really nothing to discuss, the -dev list has always been about academic and technical posts relating to bitcoin, not general disucssions or philosophising. Over the last several months the signal to noise ratio has increase exponentially. The list was never meant to be once size fits all. You've seen, if you follow all discussions on the ML
104 2015-09-21 18:41:17 <btcdrak> (which is probably unlikely given the overly numerous posting) that every now and again a thread gets unfocused, OT and people are asked to take it elsewhere, but there isnt anywhere for them to go.
105 2015-09-21 18:41:40 <wumpus> CodeShark: no need to get defensive, I'm not trying to prescribe what you should do
106 2015-09-21 18:41:49 <warren> richardkiss: #lightning-dev
107 2015-09-21 18:41:53 <warren> also #sidechains-dev
108 2015-09-21 18:42:18 <richardkiss> thanks
109 2015-09-21 18:43:08 <warren> btcdrak: random people have asked for more lists to be created, I think we should write proposals on list and come to some sort of consensus of the purpose and policy of how they are to be managed.
110 2015-09-21 18:43:11 <btcdrak> warren: the dev list is a precious resource and communication vehicle for people in the technical community, including other implementors, downstream projects, stakeholders etc. They can no longer use the list to keep up on relevent topics or discussions because of the high signal to noise ratio.
111 2015-09-21 18:43:16 <wumpus> btcdrak: a similar thing happened to the bitcoin talk forums
112 2015-09-21 18:43:23 <CodeShark> perhaps we should stadt by moving this discussion we're having somewhere else - I'd like a channel where we can discuss how to communicate with and educate others
113 2015-09-21 18:43:23 <wumpus> (long time ago)
114 2015-09-21 18:43:40 <wumpus> CodeShark: #bitcoin?
115 2015-09-21 18:43:57 <CodeShark> No, that's a channel *to* educate others
116 2015-09-21 18:44:01 <jcorgan> +1 for making the new list bitcoin-discuss and redirecting off-topic -dev traffic there
117 2015-09-21 18:44:11 <CodeShark> Not to disucuss how to educate others
118 2015-09-21 18:44:21 <btcdrak> I'll post something to the list, but to be this is a no-brainer. If we do teh same for #bitcoin-dev and #bitcoin I cant see the reason not to do it for the ML
119 2015-09-21 18:45:01 <warren> btcdrak: write a multi-paragraph proposal
120 2015-09-21 18:45:29 <warren> btcdrak: look at Alex Morcos' post from a few weeks ago pertaining to guidelines of the purpose of bitcoin-dev list
121 2015-09-21 18:45:31 <btcdrak> warren: I'm not even saying the list would be that high usage, but it can whatever doesnt fit on -dev
122 2015-09-21 18:46:15 <btcdrak> warren: lol, no white paper coming from me, I'll say what I think. If we cannot agree on a separate mailing list I think we really do have a problem.
123 2015-09-21 18:46:37 <btcdrak> because it's so obvious as to be, well, obvious
124 2015-09-21 18:48:05 <btcdrak> warren: the purpose of the -dev mailing list is already defined and a ton of discussions over the last several months do not fit within that remit.
125 2015-09-21 18:48:15 <wumpus> btcdrak: I don't think everyone will ever agree :) but if someone were to take initiative and make it I'm sure people will follow
126 2015-09-21 18:48:57 <wumpus> could also make more specific lists, like bitcoin-bips, which is only for proposing and discussing new bip proposals
127 2015-09-21 18:49:21 <warren> It's overoptimistic to believe that simply having another list will solve the problem.
128 2015-09-21 18:49:57 <btcdrak> wumpus: if you have too many lists then it also doesnt work as people dont subscribe to all lists, so people miss discussions.
129 2015-09-21 18:50:23 <wumpus> 'bitcoin-dev' is too general anyway, people were never sure whether it was about bitcoin core development or alternative clients or bip proposals or all of those...
130 2015-09-21 18:50:30 <wumpus> it's just athat a lot more people joined
131 2015-09-21 18:50:38 <warren> We need a full package of something like: 1) The on-topic policy needs to be defined for bitcoin-dev. 2)  moderators tasked to enforce it, and 3) separate lists for less technical persons staffed with people to educate, as it would be wholly dishonest to shunt people onto a list where nobody technical answers questions.
132 2015-09-21 18:50:42 <btcdrak> need one place to discuss development proposals on techncial and academic merits and so on like has always happened. I guess the dev ML also served a -wizards like role too.
133 2015-09-21 18:50:53 <btcdrak> warren: yes. that's clear
134 2015-09-21 18:51:02 <jcorgan> well, if the core devs don't have a place to do their business, they'll just create their own medium for it, and i'd rather be able to listen in and possibly participate
135 2015-09-21 18:51:13 <wumpus> btcdrak: at least people can check the archives - certainly if the lists themselves are low traffic
136 2015-09-21 18:52:03 <btcdrak> wumpus: right, but I would say we shoudl not fix problems that dont exist. as far as I can tell, the dev list used to be ok before the-war-on-blocks started
137 2015-09-21 18:52:19 <wumpus> well, bitcoin core discussion mostly happens here and on github. It's the network-wide proposals that need a ML.
138 2015-09-21 18:54:14 <btcdrak> warren: the problem afaict has not been people coming for technical assistance on -dev-ML, but conversations diverging from technical and academic issues to OT discussions. There's one raging on the ML right now actually, it's a good example
139 2015-09-21 18:55:34 <jcorgan> i've been bitbucketing the ML for weeks now, except for a whitelist of people who actually discuss dev stuff
140 2015-09-21 18:55:51 <wumpus> btcdrak: agreed - if people actually were to come for, serious, technical assistance it's not that much of a problem
141 2015-09-21 18:57:04 <warren> I cna't find Alex Morcos' post, can someone find the URL?
142 2015-09-21 18:58:14 <wumpus> it's the people that don't ask, but make wrong assumptions and argue based on them, time after time, even though they've been explained the issue many times, that wear people out
143 2015-09-21 19:01:20 <CodeShark> perhaps we're framing the issue incorrectly, then
144 2015-09-21 19:03:11 <CodeShark> we shouldn't even be arguing with these people in the first place
145 2015-09-21 19:03:40 <wumpus> most of use don't. But they also argue with each other :)
146 2015-09-21 19:08:19 <moa> ... and then one of them rage-quits
147 2015-09-21 19:10:15 <Ylbam> on est juste repassé devant Nxt et Bytecoin
148 2015-09-21 19:10:20 <Ylbam> sorry wrong chan
149 2015-09-21 19:13:37 <morcos> warren is this the email you're referring to: https://lists.linuxfoundation.org/pipermail/bitcoin-dev/2015-August/010463.html
150 2015-09-21 19:15:00 <morcos> we're trying to hard to find perfect solutions.  lets do the simple expedient thing, start a new highly focused list for strictly technical development.  if someone wants to take the effort to split up bitcoin-dev later, they can.
151 2015-09-21 19:15:37 <CodeShark> Yes, agreed
152 2015-09-21 19:16:17 <warren> morcos: hmm, this is the opposite of what others have proposed, define in more detail?
153 2015-09-21 19:17:16 <morcos> i said same thing in the scrollback.  its going to be easier to move the smaller group of people who are motivated to have a more isolated technical discussion, than to move the whole community.
154 2015-09-21 19:17:46 <morcos> also it serves the purpose that many of us might not actually unsubscribe from bitcoin-dev so woudl still be there to as CodeShark says educate and communicate when we have time and inclination
155 2015-09-21 19:18:26 <btcdrak> moros: that's backwards, we have a technical list, and the readership. We need to introduce moderation to shepherd the OT discussions to a general, sanctioned place (not a dumping ground, but something "official")
156 2015-09-21 19:18:28 <wumpus> agree morcos
157 2015-09-21 19:18:37 <wumpus> it's backwards, but it may be more practical
158 2015-09-21 19:18:51 <wumpus> less 'shepherding' necessary
159 2015-09-21 19:18:52 <btcdrak> remember the dev ML is not just for us, it's for the technical community at large to keep abreast of things
160 2015-09-21 19:19:04 <moa> a bitcoin-wizards ML?
161 2015-09-21 19:19:07 <wumpus> (at least I don't feel much like doing that)
162 2015-09-21 19:19:09 <btcdrak> wumpus: in my experience people follow rules
163 2015-09-21 19:19:14 <jgarzik> I can't believe we're still wasting time on this :)
164 2015-09-21 19:19:14 <morcos> btcdrak: yes thats what makes more theoretical sense, but think about what will work practically
165 2015-09-21 19:19:26 <jgarzik> Spend a week where list admin grumps at people off topic
166 2015-09-21 19:19:26 <morcos> jgarzik, can you bring your neck a little closer
167 2015-09-21 19:19:37 <btcdrak> wumpus: the problem is there's not been a separate place at all... that's the issue, nothing else.
168 2015-09-21 19:19:52 <jgarzik> Have a standard "go here if off topic" answer
169 2015-09-21 19:20:11 <jgarzik> Exile for 1 day -> 1 week -> escalate if person can't calm down
170 2015-09-21 19:20:14 <btcdrak> jgarzik: I know right? This is slightly frustrating. I tried to create the list already, but didnt find the right listadmins lol
171 2015-09-21 19:20:20 <btcdrak> jgarzik: exactly
172 2015-09-21 19:20:32 <jgarzik> This is a Solved Problem in open source land.
173 2015-09-21 19:20:53 <wumpus> that's very easy to say, it may be less easy with bitcoin
174 2015-09-21 19:20:58 <CodeShark> it's too late to start imposing strict shepherding on -dev without serious pushback, unfortunately...but if we create a new ml with these rules from the start we can avoid the pushback.
175 2015-09-21 19:20:58 <morcos> btcdrak: jgarzik: agreed that will help.  but will work better if we move.  right now there is too much overlap between what is acceptable and what is not.
176 2015-09-21 19:21:10 <jgarzik> oh good grief
177 2015-09-21 19:21:17 <jgarzik> it's a brand new list.  this is not rocket science.
178 2015-09-21 19:21:22 <gmaxwell> Don't underestimate the power of "this is offtopic here, go there"
179 2015-09-21 19:21:25 <jgarzik> "if off topic, go here"
180 2015-09-21 19:21:43 <btcdrak> yay gmaxwell.
181 2015-09-21 19:21:59 <morcos> sigh... ok, lets assume your way will work..  why is it better than my suggestion?
182 2015-09-21 19:22:05 <CodeShark> The problem is some people are already subscribed :p
183 2015-09-21 19:22:08 <jgarzik> All lists need an asshole policy
184 2015-09-21 19:22:26 <btcdrak> What's happening here is people are worrying too much. We just create  anew list. I'll quite happily step up
185 2015-09-21 19:22:35 <jgarzik> agreed
186 2015-09-21 19:22:40 <btcdrak> <---- prime asshole
187 2015-09-21 19:22:47 <CodeShark> +1 for morcos on pragmatism over ideological purity!
188 2015-09-21 19:22:59 <gmaxwell> To be clear, I also think a more focused list would be effective too. Doing so might make scoping easier... or harder. I'm not sure.
189 2015-09-21 19:23:05 <jgarzik> tempest in a teakettle.  make the list what devs need it to be.  the end goal is productive technical discussion on that list.
190 2015-09-21 19:23:14 <wumpus> the problem is that bitcoin-dev has always been virtually unmoderated, so if you start trying now you'll likely get all kinds of allegations of censorship, I think morcos' solution is more practical
191 2015-09-21 19:23:18 <jgarzik> boot those
192 2015-09-21 19:23:25 <CodeShark> Precisely, wumpus
193 2015-09-21 19:23:29 <jgarzik> wumpus, list moderation is normal
194 2015-09-21 19:23:35 <wumpus> but I'll shut up about this, indeed, it seems a waste of time, even here we get into endless arguments
195 2015-09-21 19:24:06 <btcdrak> Who has the contact details of the Linux Foundation mailman admins?
196 2015-09-21 19:24:13 <gmaxwell> wumpus: for some reason "this is offtopic here, go there" is very nearly a fully general answer to that kind of respose. (for better or worse)
197 2015-09-21 19:24:20 <gmaxwell> "
198 2015-09-21 19:24:48 <wumpus> gmaxwell: it may work if there's another list to redirect them to, there may be a bootstrapping problem though
199 2015-09-21 19:25:11 <wumpus> boostrapping with the people that are interested in a better list is likely easier
200 2015-09-21 19:25:37 <jgarzik> wumpus, Come on - every decision does not need to be dragged out, overly engineered!
201 2015-09-21 19:25:50 <wumpus> jgarzik: it seems to be
202 2015-09-21 19:26:01 <gmaxwell> wumpus: yes, that fix requires another place to redirect people.
203 2015-09-21 19:26:07 <jgarzik> I'm happy to post to the list "this list is for focused technical discussions.  all else will get booted"
204 2015-09-21 19:26:18 <btcdrak> jgarzik: ditto
205 2015-09-21 19:26:34 <btcdrak> there we've got two peolpe willing to get their hands dirty.
206 2015-09-21 19:26:38 <wumpus> ok, you have my support then
207 2015-09-21 19:26:40 <gmaxwell> That said, our problems on Bitcoin-dev are not limited to a lack of "focused technical discussions", though if that much were solved it would be a step forward.
208 2015-09-21 19:27:04 <morcos> ok, last thing i'll say, please go read the response to my email linked above
209 2015-09-21 19:27:28 <morcos> agreed i don't have quite the same "authority" as you guys, but it was a flat out failure
210 2015-09-21 19:27:37 <CodeShark> "bitcoin-dev" is just a name, afterall - what's the most expedient way to have the forum we want? Create one and set the policy we want from the get go
211 2015-09-21 19:27:44 <wumpus> morcos: that may be an advantage in this case
212 2015-09-21 19:28:49 <jgarzik> CodeShark, the list was created here by IRC denizens for focused technical discussion </recent history>
213 2015-09-21 19:28:51 <CodeShark> Otherwise we'll just end ip arguing on bitcoin-dev over what the policy in bitcoin-dev shoild be :p
214 2015-09-21 19:28:59 <wumpus> morcos: I know at least two people who would completely jump to the roof if I was to propose something changing about bitcoin-dev
215 2015-09-21 19:29:53 <jgarzik> wumpus, There are always some that will disagree no matter what.  At the end of the day, the list needs to be productive or its useless to its namesake.
216 2015-09-21 19:30:10 <wumpus> jgarzik: I agree
217 2015-09-21 19:30:25 <jgarzik> 1. Have a bitcoin-dev-chat list, a target spillover list
218 2015-09-21 19:30:40 <jgarzik> 2. announce focused policy for bitcoin-dev
219 2015-09-21 19:30:42 <jgarzik> 3. done
220 2015-09-21 19:31:23 <btcdrak> 4. we all go out for beers
221 2015-09-21 19:31:53 <jgarzik> bitcoin-dev is a brand new list, so now's the best time to tweak things.  A second mass move of users is pointless pain.
222 2015-09-21 19:32:01 <btcdrak> ^
223 2015-09-21 19:37:43 <CodeShark> hmmm...so now we're even forking the -dev list :p
224 2015-09-21 19:38:33 <btcdrak> I'm going to post this to the ML https://gist.github.com/btcdrak/97518dd923e8e2ec13a5
225 2015-09-21 19:39:12 <btcdrak> any comments or tweaks?
226 2015-09-21 19:39:17 <gmaxwell> I think any solution will require a strong commitment from people less politically vulnerable to enforce the rules.  Currently a reasonable belief is that a fair amount (but not all, by far) of the noise is from people who are largely outside of the technical community, who don't have any feeling for the norms of it, and whom don't care about preserving their reputation in it.
227 2015-09-21 19:39:46 <gmaxwell> If told not to do whatever the hell they want, some of them will fight back by attacking the reputation of whomever ask them to change their behavior.
228 2015-09-21 19:40:15 <morcos> Also I see no point in making such a suggestion before bitcoin-discuss or equivalent is created
229 2015-09-21 19:40:27 <morcos> We got in a big enough discussion here about the best way to move forward
230 2015-09-21 19:40:38 <morcos> no reason to open it up to the mailing list for further discussion
231 2015-09-21 19:40:43 <morcos> lets just do and tell them what we did
232 2015-09-21 19:40:50 <jcorgan> ^^^^^
233 2015-09-21 19:41:27 <jgarzik> +1
234 2015-09-21 19:41:39 <jgarzik> see above two-step plan :)
235 2015-09-21 19:42:12 <btcdrak> jgarzik: morcos: right, that's what I suggested right off the bat but got asked to write to the ML
236 2015-09-21 19:42:25 <jgarzik> I'm happy to be a list admin censoring *hole
237 2015-09-21 19:42:38 <btcdrak> I already tried to contact LF but didnt manage to find the right person to request the list
238 2015-09-21 19:42:50 <jgarzik> btcdrak, it should go through warren
239 2015-09-21 19:43:02 <btcdrak> jgarzik: but warren asked me to write to the ML
240 2015-09-21 19:43:24 <btcdrak> so, we need to pressgang warren.
241 2015-09-21 19:43:52 <gmaxwell> If there are people, ideally including people who aren't heavy contributors to Bitcoin Core (who are more politically vulnerable) who are willing to step up and horse whip people, then I think it would be fine to create a -discuss and whip people over to it.
242 2015-09-21 19:44:42 <btcdrak> warren: I think we're getting consensus here to create bitcoin-discuss first, with jgarzik as list admin (and I'm willing to take on role if it helps), then we can talk about it on the list.
243 2015-09-21 19:47:21 <jgarzik> per standard policy need more than just one of anything
244 2015-09-21 19:47:35 <jgarzik> but otherwise that's fine
245 2015-09-21 19:50:03 <jcorgan> i'm willing to be backup asshole if needed, but timeliness would be an issue for me.
246 2015-09-21 19:51:08 <gmaxwell> +1 jcorgan but there should be a couple others, but not the most critical of decisions right now.
247 2015-09-21 19:52:36 <CodeShark> I'd be willing to help moderate a discussion list if it can successfully pull off the philosophical/political stuff from -dev
248 2015-09-21 19:53:05 <wumpus> btcdrak: yes, good idea
249 2015-09-21 19:53:26 <kanzure> there is something to be said for having a moderator that works much more slowly (like a few weeks delay) to catch things that blew by really fast (but not as the only moderator)
250 2015-09-21 19:54:59 <CodeShark> I think such a list is serously needed - a place where people can have a civil discussion on philosophical/political issues surrounding bitcoin...
251 2015-09-21 19:55:16 <kanzure> CodeShark: so, cypherpunks
252 2015-09-21 19:55:46 <kanzure> ("send everything to cypherpunks instead of bitcoin-dev" was a proposal i gave to adam3us once, although i don't think he liked it)
253 2015-09-21 19:57:28 <CodeShark> furthermore, I think some important issues need to be moved to that thread instead of -dev (i.e. what should we do about blocks getting full) :p
254 2015-09-21 19:59:15 <gmaxwell> Datapoint: -pricetalk was basically 100% effective at removing the target communication from #bitcoin,  as well as -wizards has almost been highly (though not 100%) effective removing the target conversation from #bitcoin-dev.
255 2015-09-21 20:28:27 <warren> gmaxwell: morcos: jgarzik: I request that the full plan be summarized into one document before any list is created.  I heard two different plans in opposite directions in the above discussion.
256 2015-09-21 20:29:20 <gmaxwell> warren: This response is not very constructive.
257 2015-09-21 20:29:42 <warren> It just isn't clear to me what the goal is.
258 2015-09-21 20:29:57 <gmaxwell> warren: if you have a question, ask it. Do not demand people spin their wheels on some complex specification or whatnot.
259 2015-09-21 20:30:20 <warren> <morcos> [19:15:00] we're trying to hard to find perfect solutions.  lets do the simple expedient thing, start a new highly focused list for strictly technical development.  if someone wants to take the effort to split up bitcoin-dev later, they can.
260 2015-09-21 20:31:09 <gmaxwell> right, so your ask is do we want a bitcoin-dev-discuss or a bitcoin-dev-for-real, right?
261 2015-09-21 20:31:24 <warren> If the general consensus is to spin up a bitcoin-discuss list, do we have a plan for the on-topic guidelines and process for bitcoin-dev?
262 2015-09-21 20:31:56 <warren> <warren> [18:50:38] We need a full package of something like: 1) The on-topic policy needs to be defined for bitcoin-dev. 2)  moderators tasked to enforce it, and 3) separate lists for less technical persons staffed with people to educate, as it would be wholly dishonest to shunt people onto a list where nobody technical answers questions.
263 2015-09-21 20:32:50 <warren> A related problem is mailman currently has a shared password for all moderators and all admins.  Soon LF will upgrade to Mailman 3 where finer-grained access is possible.
264 2015-09-21 20:33:14 <morcos> warren: i think for now we're pursing bitcoin-discuss as the new list, i appear to have been out chatted
265 2015-09-21 20:33:48 <warren> morcos: in that case, I want to hear a rough consensus plan for on-topic policy and how it would be enforced
266 2015-09-21 20:34:12 <gmaxwell> warren: Perhaps I didn't say this strongly enough before.  If you put lots of demands on this it just will not happen.
267 2015-09-21 20:34:28 <gmaxwell> Is that your goal? If not you need to adopt a different approach.
268 2015-09-21 20:34:44 <warren> I'm saying it would be reasonable to have a plan for the less-technical list to be viable.
269 2015-09-21 20:35:07 <warren> Otherwise you're forcing people to move there, where nobody answers questions or takes the time to educate them.
270 2015-09-21 20:35:16 <gmaxwell> So?
271 2015-09-21 20:35:36 <gmaxwell> I gain _nothing_ in the short term for this. The best approach for me is to simply unsubscribe. Demanding I take design costs and reputational exposure for writing rules for this grand plan you're demanding it's an argument that pans out.
272 2015-09-21 20:36:06 <gmaxwell> (gah multitasking... it's an argument without a payoff for me)
273 2015-09-21 20:36:18 <morcos> i was about to say i'm not sure you know what pans out means
274 2015-09-21 20:36:32 <gmaxwell> Now, if you want to go ask btcdrak and jcorgan, etc. for a plan okay dokie.
275 2015-09-21 20:36:48 <warren> btcdrak seems to be demanding it happen without a plan.
276 2015-09-21 20:37:12 <gmaxwell> warren: At the end of the day you cannot force people to have conversations they don't want to spend their time on.
277 2015-09-21 20:37:47 <morcos> warren, the problem is here is lack of action, not lack of thinking these things through, so lets not derail action, we can always improve things later as we go...
278 2015-09-21 20:37:50 <warren> I'm willing to go forward without a plan formed beforehand.  Who do we give moderator access on bitcoin-dev though?
279 2015-09-21 20:38:06 <btcdrak> https://www.irccloud.com/pastebin/pznglMIb/
280 2015-09-21 20:38:12 <btcdrak> warren:
281 2015-09-21 20:38:39 <warren> OK, let's do it, my new question is who moderates bitcoin-dev.
282 2015-09-21 20:44:26 <btcdrak> so people to moderate were "assholes" were jgarzik, jcorgan, CodeShark, myself.
283 2015-09-21 20:44:43 <btcdrak> omg, that was a horrible muddle
284 2015-09-21 20:44:50 <warren> I didn't understand it.
285 2015-09-21 20:45:10 <btcdrak> I'll try again, "people willing to moderate were jgarzik, jcorgan, CodeShark, myself"
286 2015-09-21 20:45:28 <CodeShark> Lol
287 2015-09-21 20:47:27 <jgarzik> IMO moderate is too strong a word.  It is after-the-fact moderation not post approval...
288 2015-09-21 20:47:35 <jgarzik> "we grumble at you" moderation
289 2015-09-21 20:47:52 <jgarzik> (which can escalate to list exile but probably won't)
290 2015-09-21 20:48:11 <warren> well, moderation bit comes before list exile
291 2015-09-21 20:48:33 <warren> I think the approach on lightning-dev was the opposite
292 2015-09-21 20:49:05 <warren> moderation bit by default for new users, their first post is judged for on-topicness, they receive a personal note about setting expectations, and moderation bit is removed.
293 2015-09-21 20:50:20 <warren> That sounds heavy-handed, except the moderation bit is only in that case used as an opportunity to set expectations, they are allowed in no matter what.
294 2015-09-21 20:50:28 <warren> Not sure if we want to go that far?
295 2015-09-21 20:51:02 <btcdrak> warren: I think for the time being we just want to see how well telling people works (it usually does).
296 2015-09-21 20:51:11 <warren> ok
297 2015-09-21 20:51:24 <btcdrak> warren: those kind of more heavy handed tactics are for when lists get spammed by pill pushers.
298 2015-09-21 20:51:36 <morcos> warren: since the more restricted list is -dev, and already has all the subscribers, it seems we can't really follow that approach..  (ok i'm too slow)
299 2015-09-21 20:59:29 <warren> To clarify, we want one additional list right now named bitcoin-discuss ?
300 2015-09-21 20:59:59 <phantomcircuit> if we receive an invalid block from a peer do we set Misbehaving?
301 2015-09-21 21:04:33 <CodeShark> warren, unless you can think of a better name
302 2015-09-21 21:05:08 <btcdrak> warren: yes
303 2015-09-21 21:55:40 <BlueMatt> wumpus: check the return code? its not complaining - its telling you the full list of commits that touch contrib/verify-commits, for your reference
304 2015-09-21 23:14:42 <CodeShark_> for an implementation of versionbits, it would be really nice to make it as stateless as possible...but meh...we need to persist the state once we turn off the bit
305 2015-09-21 23:17:18 <CodeShark_> so how best to persist it? stick it into chainstate?
306 2015-09-21 23:22:25 <CodeShark> or stick it into ~/.bitcoin/blocks ?