#ossasepia Logs for 19 Mar 2020



April 21st, 2020 by Diana Coman
dorion: hello enthdegree [00:55]
dorion: enthdegree, better to chat in the channel than pm. [00:58]
lobbes: I think he needs voice from ChanServ anyways [01:04]
lobbes: http://logs.ericbenevides.com/log/ossasepia/2020-03-17#1022160 << thanks. It is bittersweet of course (I can't help but think "if I didn't suck maybe it would've extended the shelf-life of tmsr a bit longer"), but it'd have been worse if I didn't even manage to get a log dump to him [01:04]
ericbot: Logged on 2020-03-17 02:57:01 dorion: lobbes congrats on getting the job done. [01:04]
ericbot: (trilema) 2019-10-09 mp_en_viaje: you saved them. [01:04]
lobbes: http://logs.ericbenevides.com/log/ossasepia/2020-03-18#1022231 << March 28, 2016 is as far back as I have for #t logs (I think this is the case for all three 'stan loggers' currently live) [01:07]
ericbot: Logged on 2020-03-18 06:21:47 jfw: lobbes: in that log import project you ever get a hold of the pre-2016 ones (ie. #b-a) in a raw form? It's the kinda thing I'd like to have around and afaik phf never shared what MP had given him to seed btcbase. [01:07]
lobbes: though I believe alf may have the older phf logs. [01:08]
lobbes: jfw: actually, phf still has those old logs up it seems [01:10]
ericbot: (trilema) 2019-08-09 phf: there's no need to znc, the entire archive of logs is available here: http://btcbase.org/log-raw/ with 2016-03.txt being the last kako file. the only outlier is tmsr-logs-apr2012-oct2013.txt which is the dump mircea_popescu gave me of the prehistoric logs, which i have a custom reader for. [01:10]
dorion: http://ossasepia.com/2020/04/21/ossasepia-logs-for-19-Mar-2020#1022284 - yeah, forgot about that. enthdegree, you'll have to pm diana_coman when she's available, 19 UTC is her regular time here. make sure you have a key registered though. [01:19]
ossabot: Logged on 2020-03-19 01:04:50 lobbes: I think he needs voice from ChanServ anyways [01:19]
ossabot: Logged on 2020-02-09 20:13:49 diana_coman: and for any newcomers and log readers: please register first a key with deedbot (see the guide: http://www.eulorum.org/Account_Setup) as otherwise it's unlikely you'll get voiced in here. [01:19]
diana_coman: hello enthdegree , what brings you here? [04:47]
diana_coman: re deedbot I suppose I'll have to roll my own too since I'll be damned if after all of it I'll just go about re-doing all the ritualic dances too. [05:02]
diana_coman: so meanwhile consider the requirements slightly relaxed and take advantage of the situation while it lasts. [05:02]
diana_coman: ritualistic* [05:03]
diana_coman: (for the log, the ritualistic dance in this case goes like this: something is promised; naive person awaits for it; nothing happens; naive person asks about it and promise is reiterated; still nothing happens; on it goes) [05:05]
diana_coman: will be back later. [05:16]
trinque: diana_coman: see, over here it looks like "ex-soviet bellows demands, offers nothing in return" as the prime cultural offering of once-tmsr. [09:47]
trinque: I told you what, two days ago, that deedbot needed to be modified a bit to support multiple channels. [09:48]
trinque: in *all of you* from the soviet system there's "narcissism for me; altruism for you" [09:57]
spyked: heh. I'll be pedantic (and besides the point) and say that in particular the Ro system looked closer to dprk than the soviets, but... yeah. [10:00]
diana_coman: trinque: you know, if I meant to target you with that , I'd have pinged you; no, I mean it way more general as an approach and it rests on my incoming-review (that will still take some time because myeah, loads of work) [10:03]
trinque: at any rate, the changes to deedbot are minimal, certainly not of the "design an OS" scale. [10:05]
diana_coman: trinque: where exactly did I bellow any demand from you? because the way it reads from here is that you gave in to your own brand of hubris and grand-words above with ex-soviets and all that. [10:05]
trinque: I read above something I offered turned into homework on which I'm late. [10:06]
diana_coman: trinque: no, I don't do implicit like that, aka "it's obvious", so no. [10:07]
trinque: anyhow. deedbot is running in both #deedbot and #trinque, so that part works. [10:08]
trinque: next step is the wot calculations, and seeing that e.g. voicing in one channel does not in the other, etc [10:08]
diana_coman: trinque: what's the reason you are doing this deedbot update & roll out anyway? [10:10]
trinque: multi-channel deedbot is the next best approximation of a distributed http://wot.deedbot.org/ [10:14]
diana_coman: ok, but what's your plan with that or you know, why sink time into it? [10:15]
diana_coman: http://ossasepia.com/2020/04/21/ossasepia-logs-for-19-Mar-2020#1022304 - btw, esp in the light of all my recent re-reading of past-tmsr stuff, this is particularly infuriating in its blindness and informed-by-own-feelings. [10:17]
ossabot: Logged on 2020-03-19 09:57:03 trinque: in *all of you* from the soviet system there's "narcissism for me; altruism for you" [10:17]
trinque: "everyone is a secret wrecker but me" also extremely infurating. [10:17]
trinque: stan also does this. [10:17]
diana_coman: huh? [10:18]
diana_coman: what's a secret wrecker? [10:18]
trinque: I just went on a dive for the thread, didn't find. [10:18]
diana_coman: and what the fuck does stan have to do in this conversation? [10:18]
trinque: I really have no idea how to read what you said as other than "trinque ain't bringing deedbot" [10:20]
trinque: but whatever. [10:20]
diana_coman: http://ossasepia.com/2020/04/21/ossasepia-logs-for-19-Mar-2020#1022318 - I can see this being infuriating (working with some guess as to "secret wrecker"); I have no idea though how does it relate to me; does it relate to me? [10:20]
ossabot: Logged on 2020-03-19 10:17:50 trinque: "everyone is a secret wrecker but me" also extremely infurating. [10:20]
diana_coman: trinque: so why not *ask*, man? how the fuck am I supposed to guess what you read in it/can't read in it? I meant specifically that I won't ask people for stuff. [10:21]
trinque: to answer your earlier question, I am working on deedbot because I wish to help keep some cohesion in former-tmsr as it devolves. [10:22]
trinque: deedbot has long been a bad prototype of a gossipd node, making it an incrementally less-bad prototype [10:23]
trinque: diana_coman: I guess there are things we're all furious about, then. thanks for explaining. [10:24]
diana_coman: thank you for the clear answer; and indeed re prototype of gossipd node, I can see it. [10:24]
trinque: nonzero chance of another great depression after all this money printing and panic. [10:26]
trinque: which is sort of ironic. kinda reminds me of the biblical revelations thing, where by the time christ comes back, nobody believes in him, or w/e. [10:27]
trinque: point being, the use-case for bitcoin might become abundantly clear real soon. [10:27]
diana_coman: sure; still too late. [10:27]
trinque: we're certain to find out either way. [10:40]
trinque: https://www.opentable.com/state-of-industry << I was struck recently by how uniform the reaction looks in this dataset, came up on w/e news orifice. [10:42]
diana_coman: http://ossasepia.com/2020/04/21/ossasepia-logs-for-19-Mar-2020#1022305 - lolz, I missed this but I have to say I disagree on both counts really. [10:43]
ossabot: Logged on 2020-03-19 10:00:51 spyked: heh. I'll be pedantic (and besides the point) and say that in particular the Ro system looked closer to dprk than the soviets, but... yeah. [10:43]
trinque: it seems like when there's something real to fear, human behavior is quite rational. [10:44]
trinque: even among the derps. [10:44]
trinque: it'd have to be for there to any derps in the first place, I'd claim. [10:44]
diana_coman: trinque: not sure how you link "uniform response" to "rational". [10:49]
diana_coman: spyked: now I wonder if you imagine #trilema and #ossasepia as exercises in narcissism of the two respective owners; is that what you're saying? [10:50]
spyked: diana_coman, not at all. I was discussing the soviet/dprk thing besides trinque's point. but I was just thinking, where does one draw the line between "personality cult" and "protecting the elite among one's group"? I can certainly see the narcissism angle, although I don't personally abide by it. [10:53]
diana_coman: I am even quite curious so if anyone cares to make a coherent case for #ossasepia being about my narcissism, I'd certainly read it; as already stated, I won't ask even for this but I'd certainly read and comment on it. [10:53]
diana_coman: spyked: do you realise that the "personality cult" is not the doing of the person at the centre of it? ie if everyone strived to *immitate* MP, it's *their doing*, not his, wtf. [10:54]
diana_coman: this is pretty much the old "they whistle after her -> she's a whore!" [10:55]
spyked: no argument here [10:55]
trinque: diana_coman: I do not wish to trade insults. I took you as taking a shot at me, and I accept that you weren't. [10:56]
diana_coman: re soviet/dprk, on one hand the reality in Ro was quite different depending on whether industrial town or not (talk to some from each type of town and you'll notice - if no heavy industry, it was nowhere near as horrible in a day to day sense) [10:56]
diana_coman: and otoh, if you talk to people from the proper soviets, it's pretty much a matter of timing. [10:57]
trinque: I don't buy the idea that the leadership has nothing to do with a personality cult, though. [10:57]
trinque: leadership authors culture, doesn't it? [10:57]
diana_coman: trinque: I don't quite follow your full reasoning there and atm I need to go so - do you mind if we get back to this around 7pm UTC? [10:58]
trinque: sure, I'll be around all day. [10:59]
diana_coman: cool; laters then. [10:59]
spyked: diana_coman, agreed re. reality. I was thinking more about the ideology (as outlined in the july theses, for example) [10:59]
diana_coman: spyked: I confess I'm quite ignorant regarding dprk's ideology [16:16]
diana_coman: I suppose it can be though. [16:16]
diana_coman: trinque: assuming you referenced there the what is art (which I don't dispute), I still don't see how you get from there to the personality cult [16:19]
jfw: http://ossasepia.com/2020/04/21/ossasepia-logs-for-19-Mar-2020#1022291 - tyvm lobbes. That spot didn't have pre-2016 when I last checked (during the "trilema goes dark" incident), and I missed where phf announced that. [16:21]
ossabot: Logged on 2020-03-19 01:10:33 lobbes: jfw: actually, phf still has those old logs up it seems [16:21]
ericbot: (trilema) 2019-08-09 phf: there's no need to znc, the entire archive of logs is available here: http://btcbase.org/log-raw/ with 2016-03.txt being the last kako file. the only outlier is tmsr-logs-apr2012-oct2013.txt which is the dump mircea_popescu gave me of the prehistoric logs, which i have a custom reader for. [16:21]
dorion: hi diana_coman, making progress over here. thrown out a couple drafts that read a bit to much as before. cutting to the chase more on my latest iteration. [16:22]
diana_coman: dorion: focus always and even start with the things you actually do/did, yes; even if it ends up as the middle in your article/draft, there's no problem and it helps to start with what you did because that's surely what you also know best, obviously. [16:23]
diana_coman: lobbes: is that apr2012-oct2013 the only part currently missing from your logger's webpage? [16:25]
diana_coman: iirc mine is also missing some part in the beginning and I guess it's that one (haven't looked at it in ages and I don't recall now) [16:26]
dorion: thanks. right now I'm starting with the problem. the biggest problem we're aiming to solve computer security. this allows me to incorporate mp's prior advice : http://logs.ossasepia.com/log/trilema/2020-01-24#1957236 , http://logs.ossasepia.com/log/trilema/2020-01-24#1957237 [16:27]
ossabot: (trilema) 2020-01-24 mircea_popescu: so my proposal is rather to look at the matter not as much as you're in the business of TEACHING LINUX (while getting together), but in the business of GETTING TOGETHER (while for instance teaching linux, or gales, or bitcoin, or whatever is needed) [16:27]
ossabot: (trilema) 2020-01-24 mircea_popescu: build it around the relationship, their building, their lifecycle, their needs. do not build it around the imaginary "point", that isn't. [16:27]
dorion: I'm still detailing the hardware/gales/training package, but making it clear that's not all we do. [16:28]
diana_coman: dorion: from what you say as such, it sounds fine indeed; and sure, what problem you solve is part of what you do, obviously. [16:29]
dorion: http://ossasepia.com/2020/04/21/ossasepia-logs-for-19-Mar-2020#1022347 - I see it more as you're interested in creating relationships and helping pe [16:40]
ossabot: Logged on 2020-03-19 10:53:16 diana_coman: I am even quite curious so if anyone cares to make a coherent case for #ossasepia being about my narcissism, I'd certainly read it; as already stated, I won't ask even for this but I'd certainly read and comment on it. [16:40]
dorion: ople become more effective in achieving their goals. Pretty much all you've demanded so far is they commit to consistently improving the [16:40]
dorion: themselves* [16:40]
lobbes: http://logs.ericbenevides.com/log/ossasepia/2020-03-19#1022353 << I think March 28th 2016 is the starting point for at least your logger and mine. (I haven't confirmed this on alf's, but I think his too starts at this date). [16:42]
ericbot: Logged on 2020-03-19 19:15:31 diana_coman: lobbes: is that apr2012-oct2013 the only part currently missing from your logger's webpage? [16:42]
diana_coman: dorion: thanks for stating it. [16:43]
diana_coman: lobbes: ah; somehow I recall the raw archive and I'm pretty sure I have it; possibly there was some trouble at import and I never got around to sort it, nor really added it to the todo list. [16:45]
diana_coman: lobbes: do I take it that your bot imports everything from 2012 on? [16:45]
lobbes: diana_coman, no, only from 2016 on [16:46]
diana_coman: (because I'd really not "solve" stuff if there's no need for it; there's plenty to solve without adding to it things already solved) [16:46]
lobbes: mp imported those old logs from his personal logs [16:46]
trinque: diana_coman: I don't think mp was curating a personality cult. [16:46]
trinque: I made the claim that leadership authors culture, is all. [16:47]
lobbes: but, if we move to use the new blog-logger then he technically has all the data we'd need (just a dump from his mysql) [16:47]
jfw: diana_coman: the trouble was precisely that the raw-log archive only started at 2016-04.txt (actually the tail end of March) when asciilifeform built his logger and nobody online at the time had earlier except via lossy scraping. [16:47]
trinque: I also totally agree with your "catcall doesn't make her a whore" comment [16:47]
trinque: so I'll think on this more [16:47]
diana_coman: trinque: ok. [16:48]
diana_coman: lobbes: so what was with that custom reader you said you made/had? I'm confused now. [16:48]
lobbes: diana_coman: I have the reader, but MP has the full dataset. Since the original plan was that once he gets everything imported to trilema.com, I'd then be running a server with a copy of his data [16:49]
diana_coman: oh, I see; ok, I'll just ask him for a dump and sync with that when everything else is in place, not a problem. [16:50]
lobbes: nice :) [16:50]
lobbes: I'll be happy if this thing actually turns into.. something at least [16:51]
lobbes: bbl in a few [16:52]
jfw: I've been working through bvt's v.sh part 1. Not sure if I'd class it as a luxury or necessity in this case but either way the inline commentary is appreciated. mod6's v.pl may have great usage docs, but has zero code-level comments; rather a regression from the original v.py it seems to me. [17:11]
jfw: Got some questions queuing up, some of which have resolved themselves on further reading so I'll keep at that for now. [17:11]
jfw: brb in 10 mins [17:12]
diana_coman: jfw: that's how it usually goes with questions in my experience too so yeah, possibly worth finishing first one read and then asking whatever questions are still standing. [17:19]
jfw: right [17:22]
diana_coman: will be back tomorrow. [17:45]
whaack: I am attempting to produce the list of channels I've successfully logged for (at least) 2 weeks without a disconnect lasting longer than N minutes. I have logs that contain the bots' JOIN / DISCONNECT timestamps for various channels. I need to run a postgres command with the psuedocode "list all channels whose first JOIN is > 2 weeks ago, excluding all channels that have a DISCONNECT without a [18:00]
whaack: following JOIN that occured within N minutes. [18:00]
jfw: whaack: something's missing from that pseudocode [18:03]
jfw: also dunno if it's a terminology difference between your code or cl-irc and irc protocol, but in the latter you PART or are KICKed from a channel, disconnect isn't an irc message [18:04]
whaack: is thinking about what's missing. [18:05]
whaack: jfw: It's a terminology difference. I considered myself connected when I receive a "JOIN" message from irc, and so I used the same term "JOIN" to denote connection to a channel. I use the term "DISCONNECT" because I can be disconnected from a channel for various reasons such as being KICK'd or losing connection to the network. [18:07]
jfw: what do you consider yourself when you've connected to the server but not yet joined any channels then? :P but suit yourself I suppose [18:08]
whaack: Here is the data http://paste.deedbot.org/?id=0dXY (and I realize I actually use "JOINED" rather than "JOIN") [18:08]
whaack: Right, better names could have been used. I also did not log the event of connecting to the server. [18:09]
jfw: re the pseudocode, just compare it to your stated goal in the same line [18:09]
whaack: k [18:09]
whaack: argh sorry, I actually do specify KICKs with KICKED [18:10]
jfw: uh, you do distinguish KICKED from DISCONNECTED though! -yep. [18:11]
whaack: (so first correction is excluding all channels that have a DISCONNECT _or KICKED_ without a following JOIN...) [18:13]
dorion: hey jfw check pm real quick [18:13]
whaack: jfw: Another problem with my pseudocode is that if I have a channel I joined 3 weeks ago that the bot successfully logged for 2 weeks, but then disconnected during the 3rd week, I would erroneously exclude that channel. [18:16]
jfw: right. [18:16]
jfw: whaack: I'll add that this sounds like a pretty tricky thing to do purely in SQL. You're looking at differences between subsequent events, which seems much better fit to a ...list-processing language (even if just awk) than a set-processing one :) [18:28]
jfw: but maybe I'm just a sql-noob and there's a nice way, would love to see it. [18:29]
whaack: jfw: I was planning on taking the intersection of the channels that matched the above criteria with the channels I am *currently* connected to. There is a case where I don't properly log the DISCONNECT message - when the bot crashes. I am concerned that there is another case where I may stop logging a channel without receiving a KICK/DISCONNECT message that I have not considered. [18:32]
jfw: whaack: do you think the intersection solves that problem? [18:37]
jfw: if considering all disconnect cases is really too hard, maybe look for where the time since last message from a given server (irrespective of channel) exceeds a given timeout [18:41]
jfw: that'll be prone to false positives of course. [18:42]
whaack: jfw: No the intersection is not really a solution. Perhaps taking that intersection and seeing if it even reduces the original set at all will give some evidence of a problem, but it won't do more than that. [18:46]
jfw: I'd have thought the bot would automatically move on to the next channel in the queue after 2 weeks thus it'd be more concerning if there *wasn't* a reduction there (dunno how long it's been going though) [18:50]
jfw: anyway I'm off for now. [18:50]
feedbot: http://ossasepia.com/2020/03/19/the-soap-rush-and-the-toilet-roll-bubble/ << Ossa Sepia -- The Soap Rush And the Toilet Roll Bubble [19:40]
whaack: jfw: thanks. atm there is no functionality to automatically rotate through channels. [23:44]

Comments feed: RSS 2.0

Leave a Reply