#trilema | 2016-11-19

( see also horse's mouth, https://archive.is/zfAli )

00:13

heh

00:14

and the obvious response "we have no fucking idea" isn't on the table because coinbase ISNT a bitcoin company.

00:14

Strength of the gabriel strategy, what will Feds take? All of one's nothing?

00:14

indigence is not a strategy.

00:14

there are always organs to sell.

00:16

<+mircea_popescu> in all deadpan honesty, this. << haha.

00:16

yea!

00:18

<asciilifeform> there are always organs to sell. << For solution to this see BingoBoingo strategy. Use substances just enough to erase market value from organs, unless grinding for sausage.

~ 1 hours 49 minutes ~

02:07

well now the remaining jo

02:08

harumph.

02:13

pete_dushenski: did you ever apply asciilifeform's 'banhammer'?

~ 4 hours 35 minutes ~

06:49

http://btcbase.org/log/2016-11-18#1570377 << cool!

06:49

Logged on 2016-11-18 22:58 mircea_popescu: Framedragger hey, i paid a grand via paypal for like 60 btc back in 2012. back then bitcoin was still enjoying the benefit of not having been popularized.

06:49

(mandatory framedragger sob about reckless spending of btc back in 2012... :( )

06:52

!~later tell gabriel_laddel fd at mkj dot lt , gpg fingerprint E2DF 986D 58A0 D387 6BA1 65FA CC05 10AA FD8A F4B7

06:52

jhvh1

Framedragger: The operation succeeded.

~ 32 minutes ~

07:25

Framedragger see, but there's a larger point here. i dunno if you followed the recent discussion re nirvana, but in any case : it is specificlaly NOT the intent of the republic to become popular.

07:25

the idea is for the republic to be imposed, preferably at the point of a sword, and painfully, VERY painfully, depersonalizingly painfully, to the common man.

07:26

we absolutely do not want for the common man to "see", in his own, common man terms, the "benefits" of the republic, and then "become part of it" and in the process start selling tmsr keychains at hot topic.

07:26

mircea_popescu: hmm. okay, i guess i get that this implies a wholly different approach.

07:26

the battle is for the elites and for the elites strictly. the common man is unwelcome in any capacity outside his physicality.

07:26

attempts to get the common man involved is how the various socialisms, be they hitler's or stalin's fail.

07:26

the http://btcbase.org/log/2016-11-18#1570169 will need to be addressed in due time then, i guess.

07:26

Logged on 2016-11-18 21:57 mircea_popescu: anyway. the republic isn't poor. but it is very fucking difficult to get proper leverage applied.

07:27

because the common man consists principally of shit, and as something becomes "popular" it becomes shitty.

07:27

so no, i don't want joe q mschmucky to "understand" why if he fucks up his payment we keep the txn. or to accept it, or to think it's a good idea.

07:28

no, i want him to be as upset as he can muster about it ; and then still not actually manage to do anything about it.

07:28

that is the proper place of the common man : angry powerlessness.

07:28

(i wonder if there is a retribution/psychological component to this. but maybe not.)

07:28

there is - the common man is lazy. the only thing that can support his transition to actual human is the experience of furious impotence.

07:29

it does not work for all, of course. but if it didn't work it means there was nothing there.

07:29

strangely i can't object to this. ("spent too much time here"). yeah, ok.

07:29

right. because this isn't another lollapalooza or w/e.

07:29

"the republic isn't optional."

07:30

for the derps, it is mandatory because they are powerless ; and for the elite it is the only viable alternative (and thus also mandatory).

07:30

because hey, if you can... you must.

07:30

(also, if there was one tv series you may want to watch, it'd be "black mirror". i watched their "christmas special" ("white christmas") yesterday, and somehow on an emotional level it made me feel easier about not having much hope in the "populace". the smarter the technological tools that the populace gets to play with, the more they fuck themselves and others up.)

07:31

*

mircea_popescu adds to list.

07:31

i don't knot about the "*only*" viable alternative. you yourself mentioned some time ago that it's perfectly normal for other intelligent peeps to have their own WoT networks (which are not connected to tmsr WoT).

07:32

yes, and they are part of the republic. per definition.

07:32

(https://en.wikipedia.org/wiki/White_Christmas_(Black_Mirror) - episodes are disjoint in terms of narrative and actors, and are stand-alone.)

07:32

just because i'm influential doesn't mean i'm universal. a grass eating quadripede in south america is just as much a herbivore as the common horse, even if they never met.

07:32

hm. i guess i see that.

07:38

by and large, the notion that the common man may have a say in his usage / fate is coming to a close. it's some weird shit some dudes pulled out of their ass and then argued persuasively a few centuries ago, it was tried in a massive social experiment, and showed to not work in any conceivable implementation. to call this a trend reversal is to entirely miss that the alternative dominates millenia by the hundreds ; "humanism"

07:38

is nary a spec in this history and soon forgotten like bell bottoms and whatever other fashionable nonsense. 80s hair and 1790s sociopolitics.

07:40

as this realisation grips the masses, watch for the populist politicians display a turn of heart about the term "future".

07:40

the same dorks who were all happy with us electoral system up until nov 8th.

07:41

similarly "it's the way of the future" isn't going to make shiny happy people out of the populist party that used nothing but this all through 1800.

07:43

and NOW we shall come back to your http://btcbase.org/log/2016-11-19#1570510 : it's easy to misperceive that whatever, the republic, take it or leave it, in 2016. it's easy to think that hey, maybne i;ll make myself a computer that works, or maybe i won't, who cares. in 2016.

07:43

Logged on 2016-11-19 11:49 Framedragger: (mandatory framedragger sob about reckless spending of btc back in 2012... :( )

07:43

but the battle is on, and it's the battle for souls. once the door closes there's no redemption.

07:45

$1k was nothing in 2012 like it is today ; and would have done nothing more for me then than now. what i bought for it was pretty stupid, according to the experts of the time, and doomed to failure. a working computer is pretty stupid, according to the experts of 2016, and obviously doomed to "it can't work".

07:45

the common man is made out of bad choices, informed by his laziness.

07:47

a working computer *may* just not be available in the future, i do believe that fully. not to go all cliche dystopian, but neural implants based on Secure Microsoft Quantum Encryption(R) and the likes may be what people use to "compute" in the future

07:48

and if that happens, we will have but our laziness to blame.

07:48

so, yeah.

07:49

(and the common man will just have found more tools to assfuck himself with.)

07:49

(and i'm scared, and i should be scared; etc.)

07:54

lol.

07:54

was in some 2013 article iirc, "history usually flows in the direction of most fucking common man"

07:55

in lighter news, found delicious armenian restaurant yest. manned by actual to god armenian. the difference from the local cows is shocking and immediate.

07:55

if danielpbarron ever visits ima feed him a sarma there :)

08:02

Framedragger the whole "oh, microsoft secure implants" is really just so much narcissism. as the raped girl angrily if unsuccessfully points out ( http://trilema.com/2013/the-dead-jew-and-the-raped-girl/ ), the peasants could at any point not have been peasants. "could". the pretense that "couldn't" is exactly what kept them in line them, and it's the source of this "secure implants bla bla" wet dream.

08:02

reality never worked that way to date.

08:03

no, it simply will be "nobody knows how to turn it on without the talking paperclip, and they can't muster the energy to try and find out". that's it. what, you think argentines fail because they're not connected to the same internet we are ? they're connected, i'm living proof. but to them, it's all netflix.

08:05

a country of farmers with "vida de noche" which consists of http://btcbase.org/log/2016-11-19#1570491

08:05

Logged on 2016-11-19 04:41 mircea_popescu: "no, i found it in a plastic bag in the alley behind the hotel where i used to work."

08:18

mircea_popescu: yeah, ok - i guess my point was that it'd be very horrible, but i suppose the chief horror is in the learned-unthinking-ness of masses. the rest is just aftereffects

08:18

nobody forces them to carry the smartphones hm ?

08:18

yeah.

~ 17 minutes ~

08:36

mircea_popescu: i plan to set up a "phuctor results investigation dashboard". (iirc trinque or someone planned to do sth of the kind, but since this overlaps with the "check distinct banner amounts etc", i may as well do it properly). nothing fancy so as to keep it actually deliverable, but i have a plan.

08:36

i was wondering, would S.SNA (or whatever entity) be able to pay for a VPS? 9.6 EUR / mo., payment monthly, my VPS accepts bitcoin.

08:37

how about you apply for a grant to the foundation and i lean on ben_vulpes / mod6 to approve it ?

08:37

(if not, i still want to do it, so will prolly still do it.)

08:37

nice, sounds like a proper process.

08:37

mircea_popescu: mkay will do.

08:37

you need what, .15 btc for a year ?

08:38

also : mkay is resistive okay. not what you mean here amirite ?

08:38

0.18 to be safe, i think.

08:38

oh shit :D

08:38

mircea_popescu: okay.

08:38

not at all.

08:38

alrighty, you write the application, make it short and to the point.

08:38

okay

08:39

then once approved someone can write a qntra piece about the whole thing and lo and behold, crony republicanism!

08:40

elite insiders 1%!!!

08:41

(also, my initial plan was to ask for a month as i can't predict with absolute certainty actual resource requirements. but i think i have a decent picture of what's needed.)

08:44

won't be worth anyone's time to dick around with monthlies

08:45

sure.

08:45

also, i take it S.BISP is not currently offering services? :)

08:46

kinda deals in whole boxes, you know.

08:46

yeah, that i know. well, it's not needed for now, so, ok

08:47

*

mircea_popescu doesn't believe in this whole virtualization thing as implemented by cloud/vps/etc. the only virtualization that makes sense to me is via teh uci.

08:47

but that infrastructure's not yet ready

08:48

it's nice to be able to have a virtual box for cheapsies - can be migrated to bare metal when the time comes.

08:48

!~later tell BingoBoingo http://wotpaste.cascadianhacker.com/pastes/dt7dr/?raw=true

08:48

jhvh1

shinohai: The operation succeeded.

08:48

i see that http://thebitcoin.foundation/tickets/UCI_tickets.html doesn't have an 'owner' field - can sometimes be a useful thing, no?

08:49

no argument.

08:50

Framedragger could be, and mod6 is also working on making private ticket trees and other things

08:50

cool stuff.

08:50

it is quite possible mod6 's ticket infrastructure eventually turns into the hiring interface we were discussing yest.

08:51

wouldn't that be something :)

09:00

mircea_popescu: asking for advice - should i include a note that should additional hosting resources be needed during the 12 month period, an additional grant (it'd include supporting data of course) may be sought? just to be transparent?

09:00

(transparent about the fact that hosting resource expectations are preliminary.)

09:01

not really the place to discuss contingencies i dun think.

09:01

sure, ok.

09:01

slippery slope, "and iof the internet runs away i'll try and run after it, and if it rains i;ll wear umbrellas and..."

09:02

grantwriting is in desperate need of reform.

09:02

should be judged on terseness not on bureaucratic notions of value ("it's long! and full of platitudes! let's count the platitudes!")

~ 15 minutes ~

09:17

mircea_popescu: plox to review: http://wotpaste.cascadianhacker.com/pastes/nz8wn/?raw=true

09:18

lol k

09:20

not terrible, has all the needed parts except "which shall provide a platform for those interested in analysing RSA keys and the surrounding metadata collected and provided by Phuctor" could benefit from (such as : by x, and y, and z) as it's currently not even vaguely clear what i could do on that platform.

09:21

okay - i wanted not to over-commit, but this is indeed pretty damn vague. will think and amend.

09:21

for starters, explain it to me. so what do i do there ?

09:21

Framedragger: let me know what, if anything, you need on my end. but i must warn, querying the db is ruinously slow.

09:22

asciilifeform ideally he gets the phucked page once and then follows rss.

09:22

much to ben_vulpes 's chagrin.

09:22

the rss skipped 100s of keys not long ago.

09:22

www is shit soup.

09:22

*

mircea_popescu is amused how the collection of crap pisses off someone particularly. dns, rss, if it ends in s it's shit.

09:23

asciilifeform suppose he reloads the phucked page every time he gets a rss feed which doesn't include items he already has.

09:23

then.

09:24

there would be, as per my current plan, two 'portals'. one would expose the db (postgres). current plan: ths would use phppgadmin. it's maintained and stable. user would make use of a read-only db role. so you could run sql queries on the whole thing. "whole thing" = sql schema which

09:24

hosts the phuctor data - p, q, e - and metadata - including country codes from geoip where applicable.

09:24

it isn't whole bag though - i had to make manual sql queries, and wait quite a while, to get the stats for the recent qntra piece

09:25

the info re ~unphuctored~ moduli, or relationship b/w moduli and keys, is not in the 'phuctored' page naturally

09:25

Framedragger what db ?

09:25

second 'portal' would be more simplistic and would, first off, be a simple way for me to present some non-gimmicky visualizations, e.g. what asciilifeform suggested some months ago.

09:26

Framedragger: if i put even slightly more load on db, site will grind to a halt

09:26

so hm

09:26

mircea_popescu: by db i mean a postgresql populated with phuctor data.

09:26

ah - i wasn't clear. db would live on this VPS.

09:26

it wouldn't proxy to phuctor server

09:26

so you want to pipe data from phuctor to a vps, and display it there ?

09:27

Framedragger: where do you intend to get the data though

09:27

i'm unclear what to do with 'live data feed', but i don't think it's a problem. it could just sync with phuctor at timed intervals.

09:27

mircea_popescu: yes, if by pipe you mean, bulk-import into VPS, not torture phuctor's own db 24/7.

09:27

the part i don't follow is where isn't this a waste of your time ? you seem to be trying to do something akin to "i will create a gui atop this command line". ok... how will you guess aforehand what buttons people want to push ?

09:28

mircea_popescu: the first 'portal' would allow users to run read-only sql queries.

09:28

but grep already allows me to do this

09:28

so there would be no decision making needed as regards user friendliness / abstract buttons

09:28

mhm.

09:28

so if grep already allows me to do this... what are you doing ?

09:29

*

asciilifeform mostly uses grep, sed, to read phucked keys

09:29

me2.

09:29

well awk but w/e.

09:29

well, you wanted to run, say, DISTINCT on banners. it sure would be great to do it in a non-hacky way, and for others to allow to do the same, no?

09:29

but banners aren't on phuctor page are they ?

09:29

again i screwed up re. phrasing. i'd import the banners from the data i have.

09:30

i got a thing that curls http://.....phucked and goes , nmap, fetches ssl certs if 445, etc

09:30

ok so you aim to take some data from phuctor, integrate it with some data you already have, and perhaps with other data you can obtain, and present it conveniently ?

09:30

i suppose a more simplistic thing to do would be for me to fish out those banners, convert into decent format if needed, and to give to whoever wants :)

09:30

this, as an evolutionary superset of "i'll just dump a db of banners and you're more than welcome to match yourselves" ?

09:30

mircea_popescu: yes

09:30

so like this it makes sense

09:30

really the banners ought to be in phuctor db

09:30

i dun wanna have to resolve record matching every time i want to query the joint set.

09:31

see ? gotta make sense!

09:31

asciilifeform: i know - i hope to have time next Wed to look into this

09:31

asciilifeform if you want to integrate it there. and next week someone observes really Y thing should also be and you're unbound.

09:31

dja feel mentally flexible enough to deal with that ?

09:32

i mean, tbh i should give asciilifeform those banners. VPS DB idea can wait.

09:32

*

mircea_popescu is not against either approach.

09:32

mircea_popescu: not really. nothing stops phuctor from displaying '3.1.3.7 on date X' and '3.1.3.7 on date Y' as separate keys etc

09:32

well up to you, it's your shitsoup.

09:33

and 'Notes' field can include whatever.

09:33

yeah but by now it includes so much we actually wnat to structure it ;/

09:33

which is why i realized that i am doomed to rewrite it yearly...

09:34

heh

09:34

that's how it starts you know.

09:34

i guess another thing is, it would be nice for #trilema folx to be able to test their own phuctor-related hypotheses without having to download all the data. but, it's not as if it's exabytes of data or anything...

09:35

Framedragger: probably thing to do will be periodic dumps

09:35

Framedragger ok but if one actually wants to sql on it, dumping it in a table isn't much work.

09:35

often though i'm happy with just a bash. i think like once i actually wanted sql syntax

09:36

and then was too lazy and bashed it anyway.

09:37

that's true! this wouldn't be anything amazeballs.

09:37

Framedragger if you wanna do data visualisations, the thing to do is the wot explorer. that i actually miss.

09:37

hmm, there is that, too.

09:37

when i reopen the thing ( which is definitely not now, i am sweating now over a long-behindschedule unrelated item coauthored with mircea_popescu ) it will be to speed up the key eater

09:38

*

mircea_popescu subscribes to this view.

09:38

okay, i'll sit on it, meanwhile have $afk stuff to do, and i'll update next week re. what i'd like to work on, if anything.

09:38

enjoy.

09:40

($afk item #1 is "coffee grinder is a bit shit and requires force, i don't want to use force in my mornings.") :)

~ 24 minutes ~

10:04

(also, rereading the convo, yes http://btcbase.org/log/2016-11-19#1570648 is what i meant, and sorry for you having to spell it out for me!)

10:04

Logged on 2016-11-19 14:30 mircea_popescu: ok so you aim to take some data from phuctor, integrate it with some data you already have, and perhaps with other data you can obtain, and present it conveniently ?

10:04

/me afk for nao

10:15

https://www.tesla.com/en_GB/videos/autopilot-self-driving-hardware-neighborhood-short

~ 20 minutes ~

10:36

is it just me or does nosuchlabs.com not ping ??

10:38

mircea_popescu: prod the boxmeister plox

10:39

looks like it's been dead for at least 3 hours.

10:45

central points of failure - suck.

10:47

and the days of 'there are servers, and they serve all-comers' -- are numbered.

10:50

now it pings, but won't take a tcp socket.

10:50

i suspect epic ddos, or blackholing.

10:52

the folx whose chumpnet we blew open, with the debian boxes, i suspect -- trying to make their displeasure known.

10:56

ben_vulpes: after laocoon was getting connect/disconnect blackholed about a week ago i threw down the banhammer as well as manually banning some decently large ip ranges that were acting the fool. now very slowly catching up to full height but down from ~70 connex to ~15.

10:56

asciilifeform: seems that i can make it open 80/tcp, but it won't give me any data

10:56

shinohai: weechat is easier than znc in my exp.

10:58

Indeed pete_dushenski ... though I use a combo of both xD

10:58

Framedragger: my guess is -- syn flood

10:58

*

Framedragger hopes balticservers provides out-of-band kvm

10:58

seems quite likely...

10:59

Framedragger: it does not, afaik.

11:00

shinohai: o neat

11:03

*

asciilifeform getting quite sour on the whole idea of 'public services'.

11:03

the 'public' needs exactly 1 'service' - an infinitely-long, red hot poker thrust up its collective arse, forever.

11:05

'providing public services' does not necessarily imply 'provide them on the very box which does the important stuff'.. (though i hear you re. your 'ability to provide caching to live feed' concerns..)

11:06

still, the permalink-able key pages can be cached and/or served from somewhere else, no?

11:07

mats: do you see the appeal in the whole autonomous driving thing ? it strikes me as appealing to the same neophilic and inconsiderate mind that wanted to vote shillary in simply because she's a woman and WOULDN'T THAT BE GREAT, not because the consequences had any bearing on the decision making process. frankly the idea that sv-shitware was in total control of my vehicle is frightening. that many millions

11:07

of lines of code and that many chinese sensors ~will~ fail at some point. and it will be unexpectedly. and catastrophically.

11:09

Framedragger: black mirror is pretty good -- im halfway through the latest series.

11:10

word! :)

11:11

Framedragger: fuck lot of good caching does if the links break.

11:11

(from, e.g., qntra.)

11:11

enemy's whole objective was 'make 0 come up when folx click the link'

11:11

it was achieved.

11:12

usg.dns does not give this 'serve from somewhere else.' deliberately.

11:13

qntra incidentally also loads very, very slowly from here.

11:13

(but -- loads.)

11:13

asciilifeform: by links break you mean that they are unreachable for extended periods of time? but boxes with cached static content can be armoured against ddos more, no? i guess unless you maintain that this is indeed syn flood or equivalent, which is agnostic to whether content up the stack is dynamic or sttaic...

11:13

it's a syn flood.

11:13

the trb node just as dead.

11:13

ah.

11:14

the 24/7 ssh pipe i had to the box on dedicated display - also dead.

11:14

ah right, whole box unreachable, not just http.

11:15

Framedragger: it is unreachable WHEN IT COUNTS

11:16

i wrote, in the qntra piece, 'examine debianized boxes for nsaware'. now 'owner' will have a chance to clean up before any mass 'examination' takes place.

11:17

i hear you, examining them ourselves (in some automated fashion or w/e) would have been prudent. "trust the public to do it", uh :/

11:17

('but,' , said my internal mircea_popescutron, 'nobody had the slightest inclination to examine anything!' -- but this is besides the point)

11:31

<+mircea_popescu> Framedragger if you wanna do data visualisations, the thing to do is the wot explorer. that i actually miss. << me too.

11:35

Framedragger: point taken about 'owner' on tickets. I considered adding a 'assignee' or 'owner' for a given ticket(s), but with the narrow view of the first project (trb), i didn't want to discourage people from thinking about solutions for a given problem just because it had my name on it or something.

11:36

Anyway, now that it seems that private a more rich ticketing system is wanted, these things can be considered for sure. Salud!

11:38

pete_dushenski: i'm mostly interested in semiautonomous hacks

11:38

luxury models commonly ship with park assist and collision avoidance

11:39

mats: familiar with idea of 'risk homeostasis' ?

11:39

as well as stability and adaptive cruise control

11:39

mod6: sure! maybe "assignee" would have the desired (lesser) connotation, i don't know. coming from some trac feature/bug tracking in distributed teams experience, 'owner' is there interpreted as simply 'person who is ultimately responsible for implementing/fixing this', with other collaborators invited and acknowledged

11:39

but yeah, sounds good :)

11:39

so, remote steering + acceleration

11:40

asciilifeform: first i've heard of it

11:41

Framedragger: aha. yeah, in my experience with ticketing systems ala scrum, it's my observation that if a ticket is "assigned or owned by Jeff, I don't even think twice about it."

11:41

fair 'nuff. guess it depends on agreed-upon processes and overall mindsets of team, and so on...

11:42

So I wanted to avoid that. But going forward, I think that'll just some sort of optional field or whatnot. I'm excited to put some work into that as soon as I can here.

11:44

mod6: regarding visualizations, i'm just curious, did you have something particular in mind, as in, how do the svg visualizations at http://www.btcalpha.com/wot/trust/?from=mod6&to=mircea_popescu and http://www.btcalpha.com/wot/user/mod6/ look to the eye?

11:45

i suppose the idea could be to re-implement that, but using deedbot's view of WoT, and add additional things as desired.

11:52

asciilifeform: at least with public-static and phuctor boxes being separate, you'd have access to the latter if it were private. (but i guess you could object with "private/undisclosed box on the internet, what is this oxymoron!")

↖ ↖

11:57

re. visualization, i like stuff like this (mouse over on labels around the circle), but it's a hella lot of JS, and i share the hate towards the latter: http://bl.ocks.org/mbostock/7607999 - what's nice about btcalpha visualization is that it uses by-now standard html5 canvas directives (<path>) with no need for JS.

11:58

Framedragger: if you go down the 'phuctor visualizer/dash' route, you might consider leaning on pg streaming replication

11:59

phf: may be able to chime in on how much load that'd add to the db process but i don't think much

12:00

ben_vulpes: oh hmm, i've never used it, very interesting and thanks for the pointer

12:00

wot browser somewhat higher priority though

12:00

note to self/logs, check https://wiki.postgresql.org/wiki/Streaming_Replication

12:00

many affordable non-luxury cars today ship with front and rear-view wide multi-angle cameras

12:01

i wonder what a limited remote steering package on a toyota camry is worth

12:02

http://btcbase.org/log/2016-11-19#1570743 << i have, as you can surmise, own comp, that still works

12:02

Logged on 2016-11-19 16:52 Framedragger: asciilifeform: at least with public-static and phuctor boxes being separate, you'd have access to the latter if it were private. (but i guess you could object with "private/undisclosed box on the internet, what is this oxymoron!")

12:02

Framedragger: ben_vulpes: it probably makes the most sense for me to do the WoT browser

12:03

anyone else is going to have stale data

12:03

regarding visualization, a more condensed question: if a javascript-using thing were delivered, would this be hated upon (and berated by asciilifeform) and accepted if otherwise good and properly maintained, or hated upon and dismissed (and berated by asciilifeform)? :)

12:03

i imagine this'd have a lot of applications in, ie, mosul, raqqa, and more modern battlefields of the future

12:03

but it is neither here nor there, the server - falls down whenever enemy wants it to fall down.

12:03

asciilifeform: right. :)

12:03

trinque: sounds good to me, fwiw!

12:04

alright I'll take that and you can focus on phuctor viz

12:04

Framedragger: just deliver svgs, eh?

12:04

trinque: fine with me

12:04

even mike_c-'s thing only used js to populate the typeahead iirc

12:04

i utterly do not see , i confess, why phuctor needs 'visualizations' etc

12:04

and yeah, i guess one should go with svgs, mike_c used <path> and it was fine

12:05

it needs a key eater that doesn't saturate the db capacity 24/7 -- yes. a server that doesn't fall down when washington farts on it -- also yes. 'js visualizations' ??

12:05

asciilifeform: well, less so visualizations than easy ways to query and analyse data. but, i agree with mircea_popescu and yourself that concrete merits are to be discussed. maybe it's not needed.

12:06

i mentioned js in relation to WoT as it's more applicable there (lots of ready-made libraries for discrete graph visualizations and so on)

12:09

i too would like more detail in re 'shall provide a platform for those interested in analysing RSA keys'

12:10

the most i can offer to anyone is a static copy of the db. and that is supposing that the box comes back up, and stays up.

12:10

^ right way to analyze data

12:11

(backups are a monthly affair, any more often would -- again -- grind whole thing to a screeching halt)

12:11

'read only sql users to a replica of the phuctor db' is, while an interesting project, not wanted by the people doing most of the analysis afaict

12:11

i cannot offer anything REMOTELY resembling real-time replication, either.

12:12

db is under ~constant ~100% load as-is.

12:14

ben_vulpes: the main thing wanted by 'people doing most of the analysis' is a BOX THAT STAYS UP

12:14

replication works on the wal, not on the committed db, and so i don't think it would have the load impact you do.

12:14

nevertheless, academic.

12:14

which seems to be beyond the current technological state of the art.

12:14

at any price point

12:15

and i want a mig

12:15

mig -- actually exists...

12:15

box that stays up - different matter.

12:16

them who want mig -- are the lucky'uns. all ya need is a bag of money this-wide and this-tall, and here's yer mig.

12:16

box that stays up - likely, needs new internet.

12:16

just don't connect it to the internet, what

12:16

then it ain't 'box'

12:17

oho it isn't now

12:17

ben_vulpes: heathen-facing publicationtron is inherently contradictory thing .

12:18

ben_vulpes: i'll think about this next week. it may be that it's not really needed, yeah - i agree. and integration of things such as ssh server banners with phuctor keys is something that asciilifeform said he'd be up to do on phuctor's db itself..

12:18

Framedragger: all of this is quite academic if thing cannot be made to reliably stay up.

12:18

Framedragger: i am also curious to know what kind of requirements you have for a vps that your current loggotron doesn't serve.

12:18

(and , for folks with poor reading comprehension, i will repeat - entire box is ~unreachable, trb node, ssh, 80, etc )

12:19

*

ben_vulpes off for a while

12:24

Framedragger: so with regard of the wotperson to wotperson (http://www.btcalpha.com/wot/trust/?from=mod6&to=mircea_popescu) and all ratings to wotperson (http://www.btcalpha.com/wot/user/mod6/); I think we liked those very much as they were. Of course any sort of improvements could be added if they make sense, etc.

12:25

Framedragger: I do like this graph for sure, http://bl.ocks.org/mbostock/7607999

12:26

I think the idea would be to get back to some sort of analog of what mike_c had in place. And then add improvments as necessary.

12:26

This would be a really great project to work on, we're long overdue to get this working with deedbot, etc.

12:30

<+trinque> Framedragger: ben_vulpes: it probably makes the most sense for me to do the WoT browser << perhaps. i'd be up for anything really at this point.

12:30

<+Framedragger> trinque: fine with me << settled then?

~ 20 minutes ~

12:51

that was me saying I'll do it.

12:51

hacking on it now actually, maybe have something up in a week

12:54

werd

12:54

thanks trinque

~ 25 minutes ~

13:20

http://btcbase.org/log/2016-11-19#1570685 << it pings with ~2-300ms, however i see no web interface.

13:20

Logged on 2016-11-19 15:36 asciilifeform: is it just me or does nosuchlabs.com not ping ??

13:20

mircea_popescu: it's dead in the water.

13:20

trb, ssh, www, all.

13:21

31 packets transmitted, 31 received, 0% packet loss, time 30040ms < you actually can't ping it ?

13:21

can ping, 150ms or so, but that's all.

13:22

well... so you want me to what, datacenter reboot ?

13:22

1st ask'em 'wtf'

13:22

i can tell you what wtf right now : learn not to overload box.

13:22

it wasn't overloaded.

13:23

dollars to doughnuts thing is under syn flood.

13:23

weren't you just saying it's barely standing ?

13:24

the db!

13:24

*

mircea_popescu engages in this exercise.

13:24

the box per se is ultra-responsive, even when 'werker' is firing (i leave a core open)

13:24

http://btcbase.org/log/2016-11-19#1570801 << just to confirm that this was trinque saying he'd be up for doing it, and yeah sounds good :)

13:24

Logged on 2016-11-19 17:30 mod6: <+Framedragger> trinque: fine with me << settled then?

13:25

mircea_popescu: (syn flood suspicion because sockets don't respond with anything, even when possible to establish tcp connection. and yes ping does seem to work.)

13:28

http://btcbase.org/log/2016-11-19#1570793 << current loggotron also runs on vps, and in itself it requires very few resources. no db use, even. at this point there's a bunch of stuff and other people's sites running on that vps, i don't feel comfortable adding additional load.

13:28

Logged on 2016-11-19 17:18 ben_vulpes: Framedragger: i am also curious to know what kind of requirements you have for a vps that your current loggotron doesn't serve.

13:29

asciilifeform the wisdom of having the data in a cheap vps is becomingf ever more apparent.

13:29

because 1) other sites' experience may be impacted, and 2) phuctor db would place some load on things. why = because i'd create a few indices, those would hog some memory, and assuming users want to do quite a bit of sorting etc, would take some cpu time as well. just sayin'. nothing scientific.

13:30

(by other people's sites i mean sites that i'm responsible for.)

13:30

mircea_popescu: cheap vps will fall down just as readily when flooded (as will the trb node, mine or anybody's)

13:30

i will point out that, unless the box has been stolen, it is still processing keys at same rate

13:30

(needs 0 net pipe for this)

13:30

yes, but much easier to a) have a lot b) move around

13:30

this, yes.

13:30

i concur ^

13:30

yes, but there are two concerns that are separable : a) flood stops processing and b) flood stops display.

13:31

phuctor wasn't ever so snappy on display even when it was accessible.

13:31

i dun think anything short of unplugging the box will stop processing.

13:31

to the point of having a ready-made system image (no, does not imply need to use docker), deployable at vps center in a matter of minutes.

13:31

and the ONLY thing that interests the shitgnomes is display.

13:31

they really couldn't, and didn't, give a shit about phuctor working throughout the years it worked.

13:31

what hurts is you know, omfg, PEOPLE CAN READ!!1

13:31

hillary clinton is not worried about raping brown babies.

13:31

she just doesn't want anyone to see it is all.

13:31

but yes, mircea_popescu is quite right re 'ought to cut'em apart'

13:32

mircea_popescu: display/processing separation is why i mentioned http://btcbase.org/log/2016-11-19#1570743

13:32

Logged on 2016-11-19 16:52 Framedragger: asciilifeform: at least with public-static and phuctor boxes being separate, you'd have access to the latter if it were private. (but i guess you could object with "private/undisclosed box on the internet, what is this oxymoron!")

13:32

the 1 minus is that the -- already quite costly apparatus -- would cost yet moar.

13:32

in a sense it is "social responsibility", ie, "the data was provided, what did republic do with it".

13:32

Framedragger i've not yet got through the log because parsing threw exception on that line i quoted :D

13:33

fuck...

13:33

nono i was being metaphorical. exception in my head.

13:33

ah

13:33

stack explosion, ya

13:34

asciilifeform: wouldn't be sure re. costs. vps can be $5/$10 a month, and stuff i used for ssh key crawling (scaleway) can bill hourly

13:34

, too.

13:34

Framedragger: db would have to exist locally on the vps, for the thing to work more or less reasonably in real time

13:35

that's a few dozen GB.

13:35

i'm not saying that going for cheapest ad-hoc option is accetabru. just, a display box showing static content needs much less.

13:35

likewise bandwidth, considerably.

13:35

asciilifeform he just wants the popped stuff.

13:35

i don't know why you need true real time, tbh.

13:35

which is k of records * 10kb or such, not the end of world.

13:35

even if more than that - all of that shit can be cached, static html pages. maybe i'm oversimplifying.

13:36

'just popped' means that you can no longer use it, as previously suggested, in place of sks, for instance.

13:36

but yes, 'phuctored.html' is a few MB currently.

13:36

aha.

13:36

have a way (rss or better version of rss, or whatever) to sync it every $n hours.

13:37

http://btcbase.org/log/2016-11-19#1570692 << lol if so, more work needed.

13:37

Logged on 2016-11-19 15:52 asciilifeform: the folx whose chumpnet we blew open, with the debian boxes, i suspect -- trying to make their displeasure known.

13:37

though then you want to see the unpopped subkey siblings of the popped moduli, and start clicking, and you'll get zip.

13:37

asciilifeform no, those link back to phuctor.

13:37

it is html after all.

13:37

well yes, but each one is own page

13:37

so ?

13:37

currently they are generated programmatically. will have to rebake whole thing such that they exist as static texts on disk.

13:37

those sibling pages, why can't they hit once, and html be generated, too.

13:38

any collection of data will have to consist of references to other sources at the edges

13:38

or, caching server makes hits itself, and generates html.

13:38

i really ought to have made entire thing a c proggy that shits out static html once in a while.

13:38

like hourly or such, could work yea.

13:38

probably how www will work, if it at all even works, in near future.

13:38

and then imagine, deploying 'display' vps would become simpler still.

13:40

with regards to vps i could help, if help/hands are needed. i know you have other priority stuff asciilifeform. also, don't know what the meta-priority level here is. (i.e., compared to other projects etc)

13:40

like, scripting vps deployment etc

13:40

a couple of vps providers i used have nice APIs

13:40

for deploying and so on.

13:40

Framedragger: dunno if you read the broadcasts, but we have an actual product rolling off conveyor as we speak.

13:40

as in , physical device.

13:41

i've read, i am doubly interested due to vagueness of said broadcasts :p

13:41

very interesting indeed!

13:41

in other wtf : all https://archive.is/http://phuctor.nosuchlabs.com/phuctored are from the same one day in may ; while all https://archive.is/http://phuctor.nosuchlabs.com/ are from the same one day i nseptember.

13:42

my hands presently are quite full with polishing off the prospectus for said item, and www page / invoicetron.

13:42

this is pretty retarded.

13:42

mircea_popescu: there is an independent snapshot on my blog, also. but also a month old.

13:42

no doubt that's priority asciilifeform, and will patiently wait

13:42

http://www.loper-os.org/pub/snap/ph_snap.html

13:42

fri. sept. 23.

13:43

heh

13:43

and all-phyuctor snapshots are like this : 1 from nov 14. three from nov 3 . two from nov 1.

13:43

fucking worthless, really, there's a total of SIX links to phuctor that went in chan this month ?

13:43

i.... think not.

13:44

i dunno that anyone other than google and yandex ever made a ~full phuctor~ snapshot.

13:44

(pretty sure these 2 did. slowly, painfully.)

13:44

asciilifeform we have a bot run by peterl which supposedly snapshots EVERY LINK IN CHAN

13:44

asciilifeform: (instead of doing a more obscure "query-able phuctor-and-stuff db" thing i could help with some kind of phuctor-public-display-vps infrastructure setup. just sayin'. thing's not clear in my head.)

13:44

mircea_popescu: for quite some time, the rss thing did not display clickable links in chan for phuctor due to overflow

13:44

(and then , in debian run, the rss feeder itself overflowed )

13:44

a damn that's also there.

13:44

mircea_popescu: just fyi archive.is fails and/or timeouts on some pages

13:45

Framedragger: the fundamental problem is that i will have to rewrite WHOLE THING for any of this to possibly happen.

13:45

and reprocess ENTIRE db.

13:45

that's the question - will you have to actually do that.

13:45

the answer is 100% yes.

13:45

trinque http://btcbase.org/log/2016-11-17#1568940 ?

13:45

Logged on 2016-11-17 15:06 mircea_popescu: trinque can deedbot rss parsing be unprincipledly altered so that any succession of alphanum characters in excess of 16 spaces is replaced with first4[...]last4 ?

13:45

that would indeed be nontrivial and would take quite a bit of your time

13:46

mircea_popescu: that would make problem even worse, could no longer search logs for a gpg fp

13:46

yes you can, by that format.

13:46

but not raw.

13:46

you can't search worth a shit right now because overflow.

13:47

true, as of this week.

13:47

ie, you don't get to see ip or link. which /.... so search, what's it do.

13:47

asciilifeform: why can't a separate box be set up to just crawl through all of phuctor pages, and then determine which of them are 'static' / won't ever change, for starters. and re-query the dynamic ones (at least the /phuctored) every $x amount of time

13:47

Framedragger: because at its current speed this will take you 50 years.

13:47

for seriously you guys.

13:47

no, not rss overflow. line overflow.

13:47

if you have a fp or w.e and wanna see go to phuctor.

13:48

Framedragger your scriba could crawl all links in chan and archive them. phf said a while back he almost has this but i've yet to see it

13:48

asciilifeform: and the permalink pages identifiable via fingerprint, are they generated by the flask backend, too?

13:48

Framedragger: understand, EVERY page in phuctor is subject to change

13:48

and if there's to there's no loss.

13:48

Framedragger: it is called phuctoring

13:48

we fucking need this archival of links in chan, like years ago.

13:48

^

13:49

(i thought someone was supposed to retroactively archive all links in all logs of all times?)

13:49

that someone utterly failed at that task, as push coming to shove proves.

13:49

mircea_popescu: right, will get it done. i had started on it, got sidetracked by the python encoding problem, and the got sidetracked by other stuff. need to re-trace, and will first do the archival bit.

13:50

by 'it', i mean scriba submitting to archive.is

13:50

all links it newly sees.

13:50

or saving copies locally, or whatever.

13:50

yeah.

13:50

~what you were planning to do above.

13:50

(again, many phuctor pages will simply timeout, iirc. but maybe can adjust; and still worth doing.)

13:50

*

mircea_popescu is pretty fucking annoyed that the MOMENT the slightest disturbance in the force occurs, we suddenly discover there was really 0 defense in depth.

13:50

"oh maybe i have a page from 2005 somewhere"

13:51

yay for the home team!

13:51

this was my reaction also.

13:51

"oh it's on phuctor and why do i need to do anything", which is how we got "those debian experts are flyeyeing the code so why should i have a clue".

13:52

asciilifeform: have you profiled an http request to a permalinkable phuctor page? where's the bottleneck? curious if you could insert a thing into flask which crawls through everything and stores locally.

13:52

and, to continue to rain on the parade, if every www site has to be run like mpex, it will cost. and the range of things that can be provided 'for the public', 'for phreeee', will correspondingly shrink.

13:52

Framedragger: db being hammered 24/7 with 'do we have this hash' 'do we have this fp' 'add this and this' 1000/sec is the bottle.

↖ ↖ ↖

13:52

and there is ~0 way around it , other than by doing the static thing.

13:53

asciilifeform: http://btcbase.org/log/2016-11-19#1570745 may be worth a try, actually.

13:53

Logged on 2016-11-19 16:58 ben_vulpes: Framedragger: if you go down the 'phuctor visualizer/dash' route, you might consider leaning on pg streaming replication

13:54

quite sure the diff'ing / updates were thought out thoroughly, i.e. time complexity is constant.

13:54

there's a slave/clone db, it gets updates efficiently from master.

13:54

Framedragger: you do not seem to understand, so i will give illustration: http://www.swapmeetdave.com/Humor/Workshop/Lumber-Car-A.jpg

13:55

^ is the db

13:55

:D

13:55

we are at the farthest possibly limit of what can be done on one box, at anything like reasonable budget (whether paying the cost of a small european flat for a server for public service is 'reasonable' is separate question)

13:55

*possible

13:55

extracting rows identifiable by 'id > $num' also becomes super slow?

13:56

understand, if phuctor were an american 'corporate' production, it would make use of at least entire RACK of boxes

13:56

Framedragger: TOUCHING THE DB AT ALL IS SLOW

13:56

understand ?

13:56

if you want anything from the db, you wait. for , possibly, HALF A MINUTE

13:57

because, for instance, the key-eater is a separate process. and has nfi that someone else wants the db.

13:57

i have no doubt that a sillicon valley version would be a datacentre of mongodb nodes which constantly fail and corrupt data, and cost millions. :)

13:58

try to apprehend the scale of the thing.

13:58

it is not a wordpress blog with ~static 5 mil words.

13:58

it is a db of ~5 mil KEYS, each of which is from 1 to 20kB, and links, etc.

13:59

and moduli, stored separately, linked bidirectionally to/from keys, ditto factors for same.

13:59

i'm sorry but you didn't convince me in regards to the 'amount' of data. > 100mil row postgres with > 100 gb of data in a 8GB ram server ran fine. and while phuctor may be a more demanding beast, shouting '5 mil keys, MILLION!' doesn't convince

13:59

whenever ANYTHING is added , it first has to be searched for in each of these classes of item, in the case that it may already exist there.

14:00

Framedragger: it is doubtless possible to optimize .

14:00

yes, the latter is super demanding

14:00

but not effortlessly.

14:00

and not without rewriting entire thing.

14:01

no argument in relation to effort needed!

14:02

i'm thinking whether it'd be worth it to just have a static replica of db-as-it-currently-is, for now. as in for "i want to touch data, there's an outdated html file on loper-os i guess?" cases.

14:02

stop web app and other stuff, copy /var/lib/pgsql/data, start web app again, use data to set up separate db.

14:03

Framedragger: you are welcome to a copy of db as soon as i get my hands on it again

14:03

but please do not expect anything like regular update

14:04

asciilifeform: of course. and also don't treat this as high priority, i prolly wont look at it before wednesday anyway. but would be interesting to take a look - and i'd take a look

14:04

supposing i even see it again.

14:04

:/

14:09

asciilifeform "process fastwerker is killing the box, would you like reboot"

14:10

lying liars.

14:11

and unless maz actuallty reads the logs instead of doing his work, I TOLD YOU SO

14:11

get out of here. how do you propose came up with name ?

14:12

i have nfi. but process runs reniced.

14:12

(how can they even see the name? kvm? i thought it was baremetal?)

14:12

i'd like to know!

14:12

and what else they've been lying about.

14:13

Framedragger kvm.

14:13

( i certainly do not recall giving the monkey an account on the box )

14:14

dude... renice works half the time.

14:14

it's not like it offers some sort of hard guarantees.

14:14

understand, thing has gone like clockwork for ~year now.

14:14

so ?

14:15

it's a fucking computer.

14:15

anyway, make a call, do i have it rebooted or let it be for a while and see if it digs itself back out ?

14:16

such pdp-10 problems we're having in 2016.

14:16

also i will add, 'werker' saturates 15/16 cores for 20 MINUTES of any given runtime.

14:16

apparently - not.

14:16

so this story (and how the fuck does the monkey know process name ?) holds 0 water.

14:16

also at issue is something called "fastwerker". that the same thing ?

14:16

yes, in fact it is.

14:17

alright.

14:17

cputime per process logging may help to settle this for future cases :p

14:17

ip-kvm will show you a sort of top.

14:17

but i dunno that anyone other than mircea_popescu had any business knowing that the process were called.

14:17

ip-kvm will show only to folks who have login on the box.

14:17

well, until you insisted i ask, nobody did. once i ask, they gotta do specific things.

14:17

...

14:18

sometimes i wonder how you think computers work.

14:18

it's a fucking admin interface bridged into the fucking bus, what the everloving shit would it care about your derpy os's notions of "users". as if those fucking EVEN WORK irl.

14:18

bejaysus.

14:19

how would mircea_popescu react if one of his gurlz misplaced her undies, and his landlord came and said where they were ?

14:19

asciilifeform in point of fact i have asked the public force to locate car drunk ho abandoned "dunno where". they did. i didn't throw a shitfit about "o noes surveillance state". this because a) i asked them to do it and b) obviously they just put the number in the stolen cars interface and then a patrol saw it.

14:20

now stop thrashing about, write better code in general and make the above call so this can go on.

14:20

the code has guaranteed run bounds. so i cannot make any comment re 'in general', there is nothing to fix.

14:20

there is no such thing as code with guaranteed jack shit on linux. next question.

14:21

let'em reboot. and i will start on the sawing-apart of the werker and wwwtron as soon as my hands are again free.

14:21

least concern, for srsly.

14:22

though i must confess syn flood had a better ring to it altogether.

14:22

i'm still convinced re syn flood.

14:22

and idiot monkey ~will~ see 'cpu utilization 95%' and think 'killing the box'

14:22

it is WHAT WE BOUGHT IT FOR

14:22

for fucks sake

14:23

you probably ran into a disk or io locks issue.

14:23

newsflash: sshd doesn't care !!

14:23

ie, ssh moduli breakage didn't sink phf's lisp-frail stack. but it did sink YOURS

14:23

a RUNNING SESSION no less

14:23

mwahaha > 9k

14:23

it needs 0 disk access !

14:24

i had a RUNNING SESSION 24/7 on own display, and it ground to a half

14:24

*halt

14:24

aha.

14:24

0328 hours lithuania time.

14:25

the dc can't or won't supply packet chart ?

14:26

i dunno that they actualy bother keeping it for client managed boxes.

14:29

fwiw box is back.

14:29

0 evidence of what killed it.

14:30

http://btcbase.org/log/2016-11-19#1571040 << uh it's 9:29 pm there right nao

14:30

Logged on 2016-11-19 19:24 asciilifeform: 0328 hours lithuania time.

14:30

btw werker uses db for approx. 40 seconds of a run, to dump keys, which subsequently all load into memory.

14:31

but yeah, i've noticed that sshd works just fine (incl accepting new connections) even if cpu at ~100% and/or no free disk space. there's that.

14:31

wonder what was the memory status. maybe in syslog

14:32

(re. failures to allocate)

14:32

0 clue in logs.

14:32

0 faulures-to-anything.

14:32

*failures

14:38

(i remember having use from a super stupid `free > memory.log` cron job every minute or so)

14:38

maybe for future

14:38

ever since the theft of the original phuctor machine, i keep scrolling logs on SEPARATE LCD in real time, 24/7/365

14:38

these showed 0 thing of interest.

14:39

(mega-unsurprise, incidentally.)

14:40

by the way, i haven't ever used it, but from reading around it appears that streaming replication may indeed be quite efficient. every time row is inserted, row is sent off to remote replica. but this does not really require cpu. so maybe it wouldn't slow things down further / wouldn't be particularly slow even if db being clobbered 24/7

14:40

i'm not sure of course.

14:40

http://btcbase.org/log/2016-11-19#1570707 << well sure, and so will your eyes.

14:40

Logged on 2016-11-19 16:07 pete_dushenski: of lines of code and that many chinese sensors ~will~ fail at some point. and it will be unexpectedly. and catastrophically.

14:42

incidentally i have machines right here in my house subjected to far harsher loads , 24/7/365, and somehow -- mysteriously -- they never ONCE suffer 'martian' failures.

14:42

http://btcbase.org/log/2016-11-19#1570724 << well http://qntra.net/2016/11/phuctor-reveals-1-in-2700-ssh-capable-machines-on-the-internet-still-debianized/ was yesterday. this is the internet, gotta move with the minute.

14:42

Logged on 2016-11-19 16:16 asciilifeform: i wrote, in the qntra piece, 'examine debianized boxes for nsaware'. now 'owner' will have a chance to clean up before any mass 'examination' takes place.

14:43

a 'welded shut box' + dc supplies GB pipe + mains current dc would be mega-product.

14:43

i had nfi that we were even buying a 'fritz chipped' box.

14:44

of course you did ? iirc we discussed processor make etc way back when selecting ?

14:44

i dun recall specifying 'remote admin card'

14:44

processor - yes. ram - yes. disk - yes.

14:44

fritz ? nope.

14:45

uh. i still wonder how you think computers work.

14:45

i only know how mine work.

14:45

apparently the divergence is winder than previously realised. anyway - ipkvm capable chipsets come with ipkvm.

14:46

(server in DC you don't own, what illusions of hardware-safety...)

14:46

what exactly did you think you were asking when you asked for the dc to kvm in your box and fix it for you ?

14:46

well the one i subscribed to in the past, had a physical cart, with wheels

14:46

that plugged into vga jack and kbd.

14:46

ethernet snake on other end.

14:46

this costs ~200 an hour you realise.

14:47

by itself it did 0, customer had to log in with own pw.

14:47

it cost 0. (but there was a queue, sometimes a whole hour long!)

14:47

this was a fiddybuck/mo dc.

14:47

sounds very much like 1998.

14:47

anyway. in the intervening years there's this new approach that dcs favour because cheaper.

14:48

such cheap. logs in as root, deletes logs on the box (yes)

14:48

leaves 0 trace.

14:48

it doesn't log in ; nor does it interact with your os.

14:49

(this is not hard, but gotta wonder , what , this is sop in commercial kvm chip nao ?)

14:49

built-in ram forensictron ?

14:49

afaik it can't actually dump ram or do useful debugging. but it can reboot the system, which is you know, a crossed wire.

14:50

i had occasion to install such a machine a few yrs ago, somewhere, it came with 'kvm chipset', but this feature was not advertised. the advertised feature is just that, kvm, where you get a tty prompt.)

14:50

sort-of. by now they have derpy guis for it.

14:50

deedbot

http://qntra.net/2016/11/slock-prepares-charity-dao-for-your-loss/ << Qntra - Slock Prepares "Charity DAO" For Your Loss

14:50

physically it ~can~ dump ram, research proj i led at the time investigated how to coax it to do this, among other things

14:51

not saying it's not possible, but in practice above pay grade of most dc techs.

14:51

shinohai: fxd

14:51

(this is a case where i am more cynical than asciilifeform: a DC you don't know, what illusions does one have; okay, one could install a sensor which signals if box was 'tampered with'; they could still take box offline, clone memory, set up a replica. etc etc etc.)

14:51

http://btcbase.org/log/2016-11-19#1570729 << logically public projects can stay ownerless, but private ones should have assignable owner.

14:51

Logged on 2016-11-19 16:36 mod6: Anyway, now that it seems that private a more rich ticketing system is wanted, these things can be considered for sure. Salud!

14:51

s/know/own

14:51

Framedragger: which is why 'comes offline' is enough misbehaviour from my pov.

14:52

once it stops responding to shell, it may as well have been stolen.

14:52

asciilifeform you gotta learn to judiciously allocate your time ; Framedragger the problem isn't so much locking up the box - it's that the box will fuck up, this being "foss" bullshit, and then the owner will be like "i dun wanna pay for a box i can't use, notwithstanding this is what i claimed i want".

14:52

asciilifeform: then, i think, box needs to physically reside with someone in your WoT. i'm just saying. maybe cheap slippery slope sophism.

14:52

and i will add that , aside from the login keyz, there are 0 secrets on the box.

14:53

Framedragger: i would keep it personally if i had where. (see old thread.)

14:53

right, fair enough i guess.

14:53

and mircea_popescu has point. i'ma back to wurk, bbl.

14:53

mircea_popescu: sure, there's that...

14:54

point in case here. it is beyond even the shadow of the most absurd doubt that alf's code locked the box. but he's willing to spend time fighting the obvious, rather than anything else, because hey, this is a bitter pill to swallow.

14:54

i'm not for a second saying HE could have done much better / differently.

14:54

but the point remains, mtbf in linux world is NOT years.

14:55

and yes, we're woprking on fixing this. the work is wide, and we can't get over, some handsome rovers from town to town etc.

14:55

mircea_popescu: mtbf of the linux boxes at my ~house~ is measured in years... why do you suppose is this.

14:56

(and now i must surely bbl, this thread could go on and on.)

14:57

laters.

15:09

http://btcbase.org/log/2016-11-19#1570742 << that's pretty much exactly the idea.

15:09

Logged on 2016-11-19 16:45 Framedragger: i suppose the idea could be to re-implement that, but using deedbot's view of WoT, and add additional things as desired.

15:11

http://btcbase.org/log/2016-11-19#1570744 << works on my system. pretty. can certainly be an alt-view or such, killed nobody.

15:11

Logged on 2016-11-19 16:57 Framedragger: re. visualization, i like stuff like this (mouse over on labels around the circle), but it's a hella lot of JS, and i share the hate towards the latter: http://bl.ocks.org/mbostock/7607999 - what's nice about btcalpha visualization is that it uses by-now standard html5 canvas directives (<path>) with no need for JS.

15:12

might discover it fails via resource exhaustion in the very large dataset that is tmsr wot, but who knows. worth a shot.

15:17

http://btcbase.org/log/2016-11-19#1570755 << yes, but you're doing the payments lol. stale data is no good, but wait-forever project even worse. the trade-offs of resource investment!

15:17

Logged on 2016-11-19 17:03 trinque: anyone else is going to have stale data

15:19

http://btcbase.org/log/2016-11-19#1570756 << js-based thing can not be the thing. it can be an optional expansion on the thing.

15:19

Logged on 2016-11-19 17:03 Framedragger: regarding visualization, a more condensed question: if a javascript-using thing were delivered, would this be hated upon (and berated by asciilifeform) and accepted if otherwise good and properly maintained, or hated upon and dismissed (and berated by asciilifeform)? :)

15:21

re. JS, yeah makes sense. and http://btcbase.org/log/2016-11-19#1570796 << thanks for the feedback! (and i agree regarding comments, preferences make sense)

15:21

Logged on 2016-11-19 17:24 mod6: Framedragger: so with regard of the wotperson to wotperson (http://www.btcalpha.com/wot/trust/?from=mod6&to=mircea_popescu) and all ratings to wotperson (http://www.btcalpha.com/wot/user/mod6/); I think we liked those very much as they were. Of course any sort of improvements could be added if they make sense, etc.

15:22

tbh i think javascript is not ever hated in the context of its original domain, "make drawn man move his arms" ; it's whenever it tries to be other things that it draws ire.

15:24

sure. "javascript in the backend", etc etc etc.; i mean, it's still a horrible language. but yeah.

15:26

agree

15:26

http://btcbase.org/log/2016-11-19#1570790 << amusingly enough, this is actually true. "i would like to lecture these monkeys in modern psychology" "ok ?" "can you design a shitproof semipermeable membrane that still allows my precious words to reach them ?" "uh. not really. whatever the fuck it is, sooner or later the shit will clog it"

15:26

Logged on 2016-11-19 17:17 asciilifeform: ben_vulpes: heathen-facing publicationtron is inherently contradictory thing .

15:27

Framedragger: Wot browser must also maintain existing url structure to retrieve a nick's profile

15:28

Or there will be pain

15:31

BingoBoingo: k; but direct this at trinque, unless he gets convinced that he wants to do things sequentially and in payment-system-first order :p

15:31

or one could even dare to develop something collaboratively, but the republic would surely segfault then.

15:31

WoT browser is like log viewer, who the fuck cares how many there are so long as there is a non-zero number!

15:32

yah but there's also enough of stuff to be done :) if trinque does WoT browser i won't do in parallel just so that there are two

15:33

Framedragger technically yes you can - make a V root for it, like properly sane people, then he'll just import that / patch it himself when he's ready.

15:33

we have V specifically so it saves us from this box-owner / code-writer confusion.

15:33

i do need to try to V...

15:34

not that there's anything wrong with running your own service on your own box. but the pill for collaboration exists and is used.

15:34

There will always be enough stuff to be done in the future. This is not a reason to not make things and move into the done category.

15:34

BingoBoingo: no disagreement.

15:35

learning to write code so that other people ~actually find it better than writing themselves~ is squarely in that http://btcbase.org/log/2016-11-17#1568870 bracket.

15:35

Logged on 2016-11-17 14:43 Framedragger: (and on a side note, "hang out on #trilema after splitting with gf" has been one of the more constructive choices i've made in my life)

15:39

and from my pov, V is truly great in the sense that it allows a very simple test for when april next rolls around. phf's viewer trivially allows to see what signatures are actually active in the deployed branches of republican code. join that set with the set of box owners and you have a very good first approximation of the l1.

15:50

http://btcbase.org/log/2016-11-19#1570931 << and if there's two* i meant lol.

15:50

Logged on 2016-11-19 18:48 mircea_popescu: and if there's to there's no loss.

15:58

http://btcbase.org/log/2016-11-19#1570951 << for the record, separate dbs for selects and inserts is the way to go. from experience it can rescue a large project / save 9x% off the hardware costs.the way you do it is that you have a master db copy which is the only one that takes the inserts, and slave dbs which are the only ones that take the selects. replication can be at dedicated sql cluster level or above, slave dbs can

15:58

Logged on 2016-11-19 18:52 asciilifeform: Framedragger: db being hammered 24/7 with 'do we have this hash' 'do we have this fp' 'add this and this' 1000/sec is the bottle.

15:58

be marked as dirty after each insert if need be etc. possibly phuctor has grown industrial enough this is actulaly needed.

16:06

http://btcbase.org/log/2016-11-19#1570960 << for the record, phuctor costs more than what renting house costs in romania. flat - maybe in frankfurt or something.

16:06

Logged on 2016-11-19 18:55 asciilifeform: we are at the farthest possibly limit of what can be done on one box, at anything like reasonable budget (whether paying the cost of a small european flat for a server for public service is 'reasonable' is separate question)

16:09

(master + 1 single slave sounds reasonable to me fwiw)

16:09

that line can be quoted so out of context

16:10

lol. yeah, might work here, it's not clear. the problem, you have to understand, is how integrated the data is. sure a db with however many lines did fine in whatever box. the problem is that everything phuctor has, phuctor uses, and so it's very close to the unmitigated nightmare which is "random access"t

16:11

in general db optimization / low consumption success stories rest on a very opposite situation - the lines are only ever interesting in one perspective and light only lights one facet at a time sorta things.

16:12

a similar situation to how compression works great on literature and poorly on (proper) random strings.

16:12

this reminds me. you know, sometimes postgres prefers to do sequential read instead of using a reasonable index, because, as it estimates, using index would involve too much seeking etc. *but* with SSD, random access is much much faster. (and btw postgres does not automatically know about seek times in SSDs..)

16:13

so needs to be informed

16:13

i dunno that alf is a db engineer by trade, so it's entirely possible specific measures could help, especially if they're of the magic number ilk of "set X to Y in config file, we didn't docum,ent this anywhere but its tru!"

16:15

anyway, i imagine a bunch of d(a)emons fighting for i/o... and yeah no one is saying that there's a straightforward solution...

16:16

the entire stack is built pretty much to create this - first, we have a phuctor, and 2mn keys looks like the whole world, and any finds look improbable as shit. then some finds are found, and more keys are fed, so now 50mn looks like the whole world and a few finds a day are expected. BUT THEN a way is found to crack thousands of keys in a week, and well, the echafaudage which held up the original is struggling.

16:17

whoopdedoo, you push it until it dies by design then you wonder it dies now and again.

16:21

deedbot

http://www.contravex.com/2016/11/19/why-you-cant-afford-globalisation-anymore/ << » Contravex: A blog by Pete Dushenski - Why you can’t afford globalisation anymore.

16:27

http://btcbase.org/log/2016-11-19#1571048 << it's not a matter of "cpu at 100%" nor is it a matter of free disk space. if on that os ssh hangs off eg dbus, and if dbus gets locked out by kernel because "dirty page" or "waiting on journal update" or whatever similar idiocy, your process is stalled. and these are just random examples, so much can go wrong in a modern box it's not even worth my time drawing the broad strokes.

16:27

Logged on 2016-11-19 19:31 Framedragger: but yeah, i've noticed that sshd works just fine (incl accepting new connections) even if cpu at ~100% and/or no free disk space. there's that.

~ 16 minutes ~

16:43

and in other red cups, http://68.media.tumblr.com/f08279648ba09a168f9102da18af85d8/tumblr_mslu6c6xEq1ro2gueo1_1280.jpg

16:51

ive been enjoying this red cups series

16:56

*thumbsup*

16:59

OH FOR FUCKS SAKE

16:59

!?

16:59

i think archive.is archivers used to work

16:59

you know what archive.is tells me nao if i urlopen it from python?

16:59

"Access denied | archive.is used CloudFlare to restrict access"

16:59

bwahahaha

16:59

motherfucking cloudflare

17:00

clearly that archive.is thing needs replacement.

17:00

and so it goes.

17:00

(this explains why PeterL's or whoever's thing used to work, but stopped. i think this is recent.)

17:01

owait, curl still works. apparently "<h2 data-translate="what_happened">What happened?</h2>\n <p>The owner of this website (archive.is) has banned your access based on your browser\'s signature"

17:01

so i'll pretend to be someone else, but clearly this is not tenable long-term.

17:01

anyway, having scriba simply curl http://qntra.net/2016/11/st-louis-homeless-overdose-on-fake-weed-as-the-great-again-looms/ > 20161119.18 in response to http://log.mkj.lt/trilema/20161119/#18 is both useful and roughly sufficient.

17:01

Logged on 2016-11-19: [00:14:14] <BingoBoingo> Who makes counterfeit marijuana as mentioned http://qntra.net/2016/11/st-louis-homeless-overdose-on-fake-weed-as-the-great-again-looms/

17:02

then it could bundle all of these together in a base64'd blob each week and deedbot them.

17:02

i suppose so. do note that archive.is attempts to retrieve additional resources, including js needed for rendering some sites, etc etc.

17:02

(not that you'll call the latter "sites", i'm sure:)

17:03

this isn't a replacement for archive.is, but more of a defense-in-depth measure.

17:03

right.

17:03

i'm still giving phf his space to figure out his feelings re lisp archive.is

17:03

okay. i'll first do archive.is as i think i just need to change user-agent, and then some time next week can set up curl + bundling thing.

17:03

it's metrosexprual.

17:06

and yes, putting an -A "TMSR agent ; contact X for discussion" on all curls can't hurt anything.

17:06

as it's exactly the sort of you know, "advertising enigmae" you were talking about yest.

17:07

heh. sure!

17:09

in other horror stories, http://wotpaste.cascadianhacker.com/pastes/jt7n2/?raw=true

17:19

Dear Sreekumar, try LSD. Sincerely, shinohai

17:19

"will this fix my problem ?" "no, but it will give it color."

17:20

Why having a boring, colorless psychosis

17:27

speaking of which, /me walking noticed an interesting flower, long, trumpet like. plucked it, smelled it, very nice smell. checked it out on the internets - it's datura. grows wild here.

17:27

*

mircea_popescu went and washed hands.

17:29

Ah Devil's trumpet

17:29

yeah. "mataperros" here.

17:30

iirc wasn't that abused in Roanoke ?

17:31

what roanoke ?

17:32

Virginia, I think I remember the colonists got all high on it.

17:32

really ? i didn't think it grew where it frosts.

17:32

also i dunno anyone'd wanna try get high on this, iirc mostly scopolamine.

17:33

https://en.wikipedia.org/wiki/Datura_stramonium <<< this?

17:33

nah

17:34

d. inoxia i think it was.

17:35

in any case white flower, plain leaf.

17:35

i think some folks in .lt used to (attempt to) get high on it, to obtain some or other state of delirium

17:36

tropane alkaloids aren't usually abused recreationally, because well... not what people usually think is fun.

17:36

huh apparently may still be used in some neopagan 'rituals'. citation needed tho

17:37

but then, people take that salvia thing, whatever it's called, and apparently better part of all experiences end up with dissociative-of-not-the-nice-sorts delirium state..

17:38

Used to could buy salvia in pouches at local gas station.

17:38

i actually planted a bunch in romania, for the flowers.

17:51

Restarting for archive.is

17:51

it doesn't log in correctly you know.

17:52

i know, i know.

17:52

!$ archivestats

17:52

Number of URLs seen since bot startup: 1; number of URLs newly archived: 1.

17:52

o nice.

17:52

aaa https://asdfasfd-asdfafd80232sdf.blasdf-asdfas22df.blasadf-asdf/asdfsdf/sdf-asdf209sdf/sth.asdf?asfd=asdfw2&psdfo2=oij2oijoij aaa

17:52

(this should be invalid)

17:52

!$ archivestats

17:52

Number of URLs seen since bot startup: 2; number of URLs newly archived: 1.

17:52

aaa http://fd.mkj.lt/ aaa

17:52

!$ archivestats

17:52

Number of URLs seen since bot startup: 3; number of URLs newly archived: 2.

17:52

pete_dushenski : update for http://bots.contravex.com in order :)

17:53

!$ archivestats

17:53

Number of URLs seen since bot startup: 4; number of URLs newly archived: 3.

17:53

seems to be fine.

17:54

mircea_popescu: um. while it will not increase "number of URLs newly archived" if URL is actually not valid as reported by archive.is, but it *will* get increased for the same URL, even if archive.is had already archived it.

17:54

this will get fixed soon.

17:54

*

Framedragger sleeps

17:54

s/but it/it/

17:55

(should be possible to find the signal from html returned.)

17:55

i dun see the problem either way.

17:55

true. just wanted to have it be real "proper"

17:55

!$ hello

17:55

Hello, world! My uptime is 0:04:32.

17:56

k. bbl!

17:58

http://fd.mkj.lt/actuallynewurl-check.txt

17:59

lol ?

17:59

ehehe

17:59

wanted to check by going to actual archive.is and seeing if the new url got archived, and when. (all good).

18:01

https://www.youtube.com/user/HarrisHarrington << this just takes the cake of poptardation. so here's a guy too young to have an undergraduate degree doing self-help posturing over youtube.

18:02

who the everloving fuck can live in this world where the yearlings think themselves experts in things jesus f christ!

18:02

that they give each other advice on "how to talk to girls" on the basis of you know, one of them actually once did it / someone once overheard someone doing it is one thing.

18:03

but really, psychic surgery over the tubes ?

~ 19 minutes ~

18:22

speaking of american teenagers, vintage lel : https://archive.is/ICxt

~ 24 minutes ~

18:46

ahahaha super bowl nuclear bomb

18:47

this is pretty good ftl.

~ 15 minutes ~

19:02

When your shitcoin needs a hardfork to fix a hardfork: https://github.com/ethereum/go-ethereum/releases/tag/v1.5.2

19:13

Restarting for daddy (update help to list owner and to include new command)

~ 40 minutes ~

19:53

mircea_popescu lulz; https://www.reddit.com/r/Bitcoin/comments/5dth6w/who_is_mircea/da7k3ea/?context=3

~ 52 minutes ~

20:46

i suppose these will pop up periodically until the end of time.

20:48

Full of luke-jr butthurt

20:50

and in other "the summer is here" news, http://68.media.tumblr.com/6cc71fb0286ffd76f249db75faed296c/tumblr_mslu4iBVoh1ro2gueo1_1280.jpg

20:52

!$ archivestats

20:52

Number of URLs seen since bot startup: 3; number of URLs newly archived: 3.

20:52

i only count 2.

20:53

http://btcbase.org/log/2016-11-17#1568940 << next one that pops should obey that rule.

20:53

Logged on 2016-11-17 15:06 mircea_popescu: trinque can deedbot rss parsing be unprincipledly altered so that any succession of alphanum characters in excess of 16 spaces is replaced with first4[...]last4 ?

20:55

win.

20:55

http://btcbase.org/log/2016-11-19#1571137 << not allergic to collab at all; in a couple of days, a V genesis shall be forthcoming on a tidy wad of html generating lisp for the WoT pages.

20:55

Logged on 2016-11-19 20:31 Framedragger: or one could even dare to develop something collaboratively, but the republic would surely segfault then.

20:56

o wait, and it needed no restart ? pretty cool.

20:56

yep, feeds is a separate service that adds entries to a queue

20:56

smart.

20:57

ty

20:58

what were you running your lisp on, was it clim ?

20:59

I tend to use sbcl as compiler, then clim is the UI doodad when one's needed

21:01

so normally you just compile into bytecode via sblc on your box and send the binary to prod box ?

21:01

... which one of you nuts wanted a copy of phuctor db ??

21:02

i'll take one anyway.

21:02

bring a 12GB-sized bucket

21:04

mircea_popescu: yep, that's rather nice. save-lisp-and-die saves the memory state of your environment, then runs a specified function when it's awoken.

21:04

anyone else who wants, please pm asciilifeform

21:05

so you make one of those, it tacks the lisp executable to the front of the thing, huck the whole wad at server and restart service

21:06

if it's a development instance of something, I send along "swank" which lets me summon a repl out of the thing and fiddle internals.

21:17

and the cycle penalty is not significant huh.

21:17

asciilifeform done ty. 100 10.5G 100 10.5G 0 0 97.3M 0 0:01:51 0:01:51 --:--:-- 101M << gotta love real tubes.

21:18

neato

21:19

it helps that we're 'neighbours'.

21:19

kinda the point.

21:21

*

trinque bash scripts everything significant he does, runs that thing next time.

21:21

whole folder full of deploy-*

21:22

trinque ever benchmarked something for resource consumption ? like i dunno, lisp-sha in your implementation vs stock sha in c as provided ?

21:24

I will say that I am not fully sold on automatic memory management yet.

21:24

*

trinque ducks

21:25

I'm sold on the structure of the language, but I have had times where I wonder why the fuck the gc hasn't kicked on yet, find folks saying "oh you tricked the gc" researching the problem

21:25

simplest example I can come up with is that I was consing thing to the front of a list, nil-ing things off the back

21:26

and at great volume; this caused sbcl to not want to gc the shit falling off the tail

21:26

trinque: yer doing destructive ops and then wondering where gc was?!

21:27

please explain to me how a list operation as simple as that in "list processing" language is not right in the middle of the groove of the language

21:27

because it isn't magic

21:27

so sure, I called gc myself sometimes, worked

21:27

has heuristics for when to fire. if you don't like'em, gotta (gc)

21:27

or pollute less.

21:28

and it's for 'list processing' in same way as fortran-90 is for 'formulae translation'...

21:30

yes well not everything can be an immutable structure.

21:31

my list contained identifiers for things coming at me from a queue, where the queue may shit itself and send dupes, but with a maximum possible age of message that is duped

21:32

and dupes are most likely to come from recent messages (say a queue server died before noting that the message was sent)

21:32

seems like a reasonable data structure for it

21:34

it makes me less anti-lisp and more in favor of using some kind of namespaced wad of memory which I can trash all together explicitly.

21:35

trinque: peruse the docs for your particular lisp, most of'em provide knobs for 'dangerous' manual manipulation of the 'auto gearbox', if you will

21:36

most of the time gc works fine; I am certainly still a student of the language.

21:36

trinque: also this'll be good practice for writing 'cons-free' manually-driven lisp, for when we do os bootloader, kernel, similar.

21:36

neat

21:38

asciilifeform: any reading material on the "cons-free" subj?

21:39

trinque: afaik only the leaked smbx lm sources.

21:39

one of the difficulties of the student is that most 'adult' lisp work took place in the era before 'must publish everything', 'open sores', etc.

21:41

there are books even on flying mig. but there are afaik none on flying mig under a non-liftable sea bridge (there is photo of this!). 'cons-free lisp' resembles the latter more than 'sop' operation.

21:43

<mircea_popescu> i actually planted a bunch in romania, for the flowers. << Different salvia. Tripping balls salvia has sparse shitty flowers if it flowers.

21:45

I might be a sick person, but I thought that stuff was great fun.

21:45

completely disorienting

21:46

iirc -- pharmacologically unique

21:46

(in family by itself)

21:47

fwiw i once ate some of its liquid extract (it was at one point freely available in usa, was literally the only type of dope i could physically get !) and it did... zero.

21:48

( did have strange dream that night, but i often do anyway )

21:49

iirc later it was banned, so it must work on ~somebody~

21:49

gotta torch at high temperature.

21:49

I had the trinque experience of being completely out-of-head for like 10 minutes

21:50

asciilifeform: 1st pass metabolism is a bitch. Lungs ftw

~ 22 minutes ~

22:13

<asciilifeform> ... and it did... zero. << Or you were wired for addiction to that weirdo substance before trying? Lacked the apperception to realize was tripping balls.

22:14

mircea_popescu: done :)

22:15

!~calc 29^14

22:15

jhvh1

pete_dushenski: 29^14 = 2.9755823267579947E20

22:16

didn't realise that jhvh1 also adopted the calculator until saw it in-chan yesterday. that sneaky shinohai ! bots directory updated accordingly.

22:18

*

trinque gets space-saving rss code in just as phuctor shuts up

22:18

lol

22:19

not for long, I'm sure.

22:29

mircea_popescu: the difference between eyes and electronic sensors on cars, ofc, is that you know when eyes stop working before it's too late. either way, if trumpenreich is to restore full employment to amerika, abolishing the minimum wage will make a full-time chauffeur affordable to anyone who can also spend $100k on a tesla or equiv.

~ 35 minutes ~

23:05

pete_dushenski: watched any of clarkson's new show?

23:10

pete_dushenski: Not really. Most muricans suck at driving.

23:11

Not to mention most muricans are broken in their temperment, unsuitable for service work where they interact with customer or customer/employer and require herding employer.

23:11

if employable at all.

23:14

ben_vulpes: 'the groand tour' isn't available outside us, uk, and i think oz for another month so... nope. looking forward to it though. you ?

23:14

grand*

23:17

BingoBoingo: 'most' isn't really an argument now is it. nor was it my point. talking solely about folks buying $100k cars, they're better off sifting through resumes to find a non-retarded driver than shipping their cash to musk et al., what with the rapidly sinking obamanchor tied around their weedy necks.

23:19

pete_dushenski: I don't think you understand the proportion represented by "most" in that usuage. We are talking about a mass economically crippled by their various "employable if X" bullshits.

23:25

tis possible. tis also possible that the world won't be worse off if they car-crash themselves off the face of it, be it by their own hand or that of their robot overlords.

23:26

http://archive.is/SG5uV << in other overlords, bannon compares himself to cromwell.

23:26

Failed to archive. HTTP code: N/A. Exception raised: True

23:28

thomas, not oliver.

23:28