jhvh1: asciilifeform: The operation succeeded.
pete_dushenski: mircea_popescu: i'd like to defend him by noting that social network != social media but dude's command of english leaves enough to be desired that this distinction isn't much of a leg to stand on
pete_dushenski: though i'm entirely open to the criticism that it's like sprinkling holy water on a dented can in the hopes that clostridium botulinum feels the wrath of my gods
pete_dushenski: looks like those xmas teas from lt bought some more business for uab cherry servers
mod6: oh hey... sorry, just catching up now. was crazy day.
mod6: <mircea_popescu> list of items for mod6's convenience : 1) nice job! 2) successive presses overfil .gitignore and possibly other files 3) deps directory should prolly be same level as patches and .seals<< sorry you encountered some issues on 2nd press. Thanks for your comments. Duly noted.
mod6: Sounds like you got it compiled and working, for this, I'm glad.
mod6: Nice. I've got mine built too with 'wires'. Just need to get a minute to hook it up to alf's node.
mircea_popescu: pete_dushenski i could make that read 5k by pushing a switch. idiots have nfi, counting all the ips you feel like exposing.
mircea_popescu: mod6 consider patching .gitignore out altogether next release, not like it's used
mod6: it's only current use, is that it allows directories not to be removed.
mod6: long ago, we all talked about this. and perhaps it is time to prune these directories. but at the time, it was decided to leave them as-is
mod6: let me dig it up -- bumeanwhile, it's regarding src/obj, src/obj/nogui, src/obj/test
mod6: Well, there are a bunch of references...
a111: Logged on 2015-09-24 00:25 mod6: <+phf> shinohai: you probably have rm_gitignore.patch applied, which removes .gitignore files from src/obj/nogui and nukes the folders along the way? << i said to disregard this patch. reason is, it wipes out output dirs required by the bitcoin makefile.
mod6: (and by bitcoin makefile, this meant 'makefile.unix' in the src directory)
a111: Logged on 2015-07-19 21:31 ben_vulpes: mod6: re gitkeeps do we even know why those empty dirs are necessary for a succesful compile?
a111: Logged on 2015-07-19 22:04 mod6: <+phf> ben_vulpes: because makefile.unix expects them, (search for obj/nogui/%.o: %.cpp, etc.) << ben_vulpes yeah this stuff
mircea_popescu: makefile bs. alright, well, i guess we'll need a test for file exists or some such hair.
mircea_popescu: i suppose alternatively could just say DO NOT PRESS TWICE! DELETE DIR BETWEEN PRESSES.
mod6: but I think now that we have the outside makefiles integrated and working well, maybe we can prune out those dirs and create them as needed? certainly worth looking into. certainly would be nice to get rid of those things finally
mod6: <+mircea_popescu> i suppose alternatively could just say DO NOT PRESS TWICE! DELETE DIR BETWEEN PRESSES. << agree. there obviously needs to be a check in there - if the press directory already exists, warn the user and stop.
☟︎ mod6: <+mircea_popescu> if you move /deps/ that'll work fine. << this part is a bit more tricky. but might be possible.
ben_vulpes: i'd rather it were transparently idempotent.
ben_vulpes: sorry. i'd rather it rip out the press target if it existed.
mod6: just automagically blow it away and create a new one?
ben_vulpes: seems the cleanest, least error prone thing to do.
ben_vulpes: warning and stopping doesn't really help, as i see it.
mircea_popescu: imagine if pregnancy worked this way. get girl pregnant, fuck her again a month later get her pregnant again, except timer reset.
mod6: blowing away someones dir without notice seems like the wrong approach, no matter how clean it might seem.
☟︎ ben_vulpes: what would the point be? destroys press integrity.
ben_vulpes: "oh i took this priceless jewel and scribbled my name in it"
mod6: i just wanna be considerate before agreeing to this. i'll think on it.
mircea_popescu: mod6 well you delete his files, but he can regenerate them anyway, is the idea.
mod6: yeah, im maybe considering the bad-form-case where someone modifies the source of a pressed directory, forgets about his changes, comes back 12 years later, then re-presses over said changes, now lost.
☟︎ mod6: but maybe this is just that, a bad-form-case. do not do this.
mircea_popescu: i mean, wtf would it be, how'd he know what of his changes were kept in the first place
ben_vulpes: "if it's not backed up, it's clearly not important"
mod6: there would be nothing to denote any said changes in a pressed directory.
mod6: this change is pretty trivial. probably will update V99994 before it comes out (was about to release this weekend). will need to add at least an automated test too. not a big deal.
mod6: the harder part is trying to move 'deps' ..//..
mod6: anyway. ya, something to look at.
mod6: interestingly, i just tried moving 'bin', 'deps', and 'build' and all of their makefiles + a few small tweaks up ../../.. and was able to build.
mod6: problem will be with pressing and a vpatch. press wants to put everything underneath of the given press directory. not place contents adjacent to a press directory.
mod6: so that'll be the next thing to take a look at.
mod6: ('bin', 'build', 'deps', being adjacent, meaning, along side of 'patches', '.seals', 'v.pl', etc.)
mod6: this would leave, for instance, a 'trb' output press directory along side of the above stated dirs. of which, all that would be contained in there is a bitcoin directory, and the underlying 'src' directory.
trinque: nah makefile can make those dirs if absent
trinque: this is the kind of thing I provided defaults for, like what your copy of your checksum proggie is
trinque: could do the same for deps path, provide some default, and if you want something else:
trinque: make DEPS=/wherever/I/stick/it
trinque: whole idea is to get them outside the press, so you don't have to copy them in each time you press
trinque: I tend to have a couple trees pressed right next to each other, when testing a patch, when writing my own
trinque: could just drag deps/Makefile into the root Makefile, and put the dep-files someplace else
mod6: that's fine. could just reference something like: ../../../deps/*.asc or something
mod6: but having it set up as i just did in the above paste is unpressable in current vtronics.
mod6: so yeah, something other than what I did is preferred (at least to me).
trinque: makefile would make the dir, maybe ~/trb-deps as default, or w/e
trinque: then they can sit there for the rest of your life and that's that
mod6: so either user creates ~/trb-deps and then pulls all .asc's from deedbot manually in offline mode. or makefile creates ~/trb-deps and pulls .asc files in `make ONLINE=1` mode and dumps files in there.
mod6: ok. i think we've got something we can work with. :]
davout: asciilifeform: what's your rigorous criterion for spamolitude?
mircea_popescu: and in other news, anyone with a background in sales who's damned good at it want a job ?
Framedragger: mircea_popescu: re. blackholing and to be alf-pessimistic for a second, node is exposed to risk of being blackholed as long as it uses TCP; because not only can enemy make it read packets (unavoidable in the end it seems), but there may be ways to making it send packets back.
☟︎ Framedragger: (i guess "silently drop connections" (TCP DROP) from any non-whitelisted IP is one way around it, sorta.)
Framedragger: (this is just to juxtapose topics of trb and gossipd for a second and to maybe show why some folks really like the lighthouse idea) :)
Framedragger: asciilifeform: hm, sure. (and i forgot about arbitrary RSTs by anyone in the path which destroy the session, heh.) do you know how the resource allocation compares to udp in a typical (say, linux kernel) networking stack?
Framedragger: i'm sure it's less than that especially as you don't need to save state/session info
Framedragger: yes that i know, but you still gotta reserve things to parse things...
mod6: <+mircea_popescu> mod6 no, JUST /deps/ << ok.
Framedragger: aha. it's funny this little difference is 'forgotten' in most tutorials etc. heh.
Framedragger: asciilifeform: can you not use raw sockets (with some kind of linux cap to allow program to open them without root), defining udp-like struct within? (i'm sure performance tuning may not be easy, i mean to achieve same level of optimisation as in kernel stack)
Framedragger: also, ip packets with custom proto number would (1) stand out more easily to enemy, and could be more easily filtered out (vs. udp header with rng-data within) - see how chinese firewall blocked tor bridges etc etc; and (2) i'm sure quite a few appliances would filter them out by default (like how they filter out icmp, etc.)
Framedragger: (one day i'll dup up logs why tae fuck mircea_popescu allegedly filters out udp by default)
Framedragger: how do you amplify udp? i guess application-layer-specific stuff like bittorrent's uTP (which has some amplification vuln shit iirc), etc.; also, dns
Framedragger: asciilifeform: forged as in e.g. where source ip addr != actual source?
Framedragger: aha, good stuff, i guess the point is that this kind of ddos is *easy*.
Framedragger: asciilifeform: btw regarding bitcoin, sure re. 1000 machines but if/when there's only 15 or however many trb nodes....
Framedragger: i think it makes sense, thinking of the future, to eventually move to a model where there is no 'public trb node ip list'.
Framedragger: ..which means that the latter should be dropped eventually, i guess (but then cue my question 'why not just work on trb-i if no compatibility with heathen prb network')
a111: Logged on 2017-02-28 13:38 asciilifeform: to wind it up, the casks algo is asciilifeform's attempt at the 'high vacuum pump' from earlier -- to get the max possible removal of something-to-allcomers element , to the extent possible without running an entirely closed wotronic system (and consequently turning into visa or swift)
Framedragger: i won't insist - you're right, we've already just had that chat on #b-a
Framedragger: then again, it'd be a more powerful filter to deter stupid people.
Framedragger: some suggest to try with stock openssh server, to see if same happens. anyway, curious if sshd_config on your end has custom knobs (e.g. prefers one crypto format over another), but i guess the whole thing's a timesink
Framedragger: oh, did mp's client connect to server before, without autossh etc? need to confirm fingerprint
Framedragger: (i wonder if he tried just-ssh'ing, no autossh first. because he needs to answer "yes" to "trust this fingerprint?" prompt first, it seems. just that.)
Framedragger: then i retract my previous snappy remark. ssh lol.
ben_vulpes: forwarding is swiss cheese, i don't even think that's for sale
a111: Logged on 2017-03-14 14:07 Framedragger: mircea_popescu: re. blackholing and to be alf-pessimistic for a second, node is exposed to risk of being blackholed as long as it uses TCP; because not only can enemy make it read packets (unavoidable in the end it seems), but there may be ways to making it send packets back.
a111: Logged on 2017-03-14 14:36 asciilifeform: ( but to make alt-udp you gotta patch the kernel. )
Framedragger: luckily there's no conceivable way to forge "source ip" field in headers :p
mircea_popescu: sure there is. but, not for garden variety gnome, not for free, etc.
mircea_popescu: it is a tempting naivism to imagine the empire as a sort of frictionless monster, in which every limb gets full and impeded access to any tool it needs whenever it needs it. this however is not how any empire works irl.
mircea_popescu: irl, the relative gain from "lording it over" the "lower" limbs that need tools to do their job is often more significant than the gain to the decident of the actual correct usage.
mircea_popescu: this is periodically discussed in the logs, but i suppose pretty out there to grok.
mircea_popescu: average gauleiter actually gains more from secretaries NOT HAVING ink ribbons so they can't do their job than from their HAVING ink ribbons and doing their job.
mircea_popescu: because every empire ever is backed by the trade in flesh, and well... the former pays in flesh. the latter - in scrip.
mircea_popescu: yes, but then asciilifeform gets pissed off that the heads aren't cut :D
Framedragger: "you using ssh-keygen, are you out of your mind!1"
a111: Logged on 2017-03-14 14:54 asciilifeform: this went on for, iirc, a year or so, then got tired.
mircea_popescu: it's a not terrible "garage chemistry lab" project, costs maybe fiddy bux a month or such\
mircea_popescu: if you already have with what it is prohibitebly expensive, because of opportunity costs
mircea_popescu: wich is why rayon plant isn't engaged in the practice of "see kids, organic chemistry!" experiments
mircea_popescu: sort-of how the bare-under-raincoat sexual behaviour is not common in men who get sucked off whenever they feel like.
a111: Logged on 2017-03-14 15:08 asciilifeform: ( what ~can~ be done is to build a thing that could function in place of a dead bitcoin. )
mircea_popescu: and the idea is to have it more or less ready before the prototype actually dies.
mircea_popescu: which, considering the sad confluence of not dumb people who (self servingly, and shamelessly), "think" bitcoin code is perfect (hey, alpha dun wanna be bothered, yes ?) and imbeciles who "are being like suparheoes power rangers"... is a virtual certainty.
a111: Logged on 2017-03-14 15:11 asciilifeform: and 'dead bitcoin', esp. if it dies on enemy's terms, would imho be a technogenic catastrophe, quite comparable to, e.g., chernobyl. ( not for mircea_popescu 'i'm rich anyway, fuck everyone' , and not for other folx, who might not even have any; but for the concept of 'gold sans the guard labour')
mircea_popescu: -293.0-292.1 ie, mp "who is not truly a billionaire because reasons" is more than willing to sacrifice tens of thousands of the coins.
mircea_popescu: i have nfi what people think, but from eg mpoe-pr's experience, it'll be something shockingly stupid and in all the joins nonsequiturial.
Framedragger: ...and all this is exposed to ecdsa and the particular parameters (secp256k1) not breaking... ouch.
mircea_popescu: asciilifeform i wonder why they even bother spelling that shit out by now, truly should just have pictorgraphs.
mircea_popescu: Framedragger a little more layered a pancake, but anyway.
mircea_popescu: asciilifeform you know ? " We aim to empower our customers" = hun go wun bo = # ; "is a leading full service supplier of high-quality" = bing ding shing ling = @
mircea_popescu: and in other hieroglyph lulz, me at travel agent, buying tickets. calls girl, "send so and so email a picture". beet red travel agent. "you know... i needed a copy of the passport". "what did you get ?" > nudegirl.jpg
a111: Logged on 2017-03-14 15:16 asciilifeform: hence why we have 1 mircea_popescu , and not 2 or 3.
a111: 2017-02-17 <mthreat> people who don't shower in jail/prison are called 'vikings', or another name in Spanish which I forget. There is peer pressure to not stink in there.
mircea_popescu: amusingly, in an earlier, simpler time of "men inhabit bare stone dugouts together", the ability to not stink without washing was regarded as about half of what made monk a candiate for sainthood.
mircea_popescu: to this day "pleasantly scented remains" is on the orthodox list of santly merits.
a111: Logged on 2017-03-14 15:17 asciilifeform: it is in the empowered whinerism.
davout: "être en odeur de sainteté"
mircea_popescu: asciilifeform yep, afaik actually imported from india via persia and the lebanese/syrian/etc early christianhood.
davout: asciilifeform: .jp thing i reckon
davout: asciilifeform: Sokushinbutsu according to pedopedia
mircea_popescu: these discussions predate stilpniks, we're talking 4-3 bce
mircea_popescu: this one's way the fuck more appealing to my taste than "and everyone will be nice to everyone and there will be no poverty or suffering"
a111: Logged on 2017-03-14 15:25 asciilifeform: i suspect that 'greybeard' holdouts, taleb et al, who 'bitcoin is a fad, will go into the sands of time, snoar' are on a sound logical footing, 'every OTHER attempt to combat massed stupidity has ultimately fallen, nuked by OurDemocracy, and the more brazen the assault, the quicker'
a111: Logged on 2017-03-14 15:50 asciilifeform: Framedragger: nothing to do with keys
a111: Logged on 2017-03-14 15:53 asciilifeform: 'no, increase the mtu!'
Framedragger: mircea_popescu: i do. the inconsistency that my (naive) mind spots is the "eh empire is not powerful anyway" + "there is no current alternative for X than Y" in conjunction with "how dare you use Y!!!"
mircea_popescu: two men land on deserted island, let's call them mrame pragger and fircea dopescu. they observe that a) monkeys are powerless b) there are no proper latrines and c) how dare you shit out in the open you stinked up the whole island. problem ?
mircea_popescu: the ~only~ think these two fellows, mrame and fircea, ARE NOT going to engage in is any sort or kind of ancestor worship.
☟︎ mircea_popescu: leaving aside how that's uninsurable -- it's not even legal.
mircea_popescu: (and if you're going to run an illegal boat ... in what sense was it hijacked by pirates. already was pirate yes ?)
mircea_popescu: well if you're using a us navy ship you're definitely a pirate.
mircea_popescu: incidentally, i wonder how many people took a ho over to graf zepellin and fucked her there, during the 30s.
mircea_popescu: (taking schoolgirl to abandoned / out of service train and fucking her is, i expect, kindergarten-universal yes.)
davout would love to visit asciilifeform and mircea_popescu's kindergartens
mircea_popescu: o, right, there's no trains where you live. well... it's basically a tin can on wheels.
mircea_popescu: nah, just some perma-moored german ship. wanna-be carrier or such.
a111: Logged on 2017-03-14 18:09 mircea_popescu: the ~only~ think these two fellows, mrame and fircea, ARE NOT going to engage in is any sort or kind of ancestor worship.
mircea_popescu: you'd be surprised how often the defense is "oh, nobody TOLD ME" when cornered.
mircea_popescu: (this, fwiw, carries 0 water in mp's own dungeon. girl that had no way to know hangs just along well warned one. it IS a little chilling but what can you do.)
BingoBoingo still continuing full sync experiment, less than 35 kiloblocks to go. Prolly should have sprung for an SSD.
davout: BingoBoingo: welcome to the club
davout: I'm syncing since last december
davout: I should be current in a few hours \o/
BingoBoingo: Well node's in August 2016, so made it past the mp-constipator block
BingoBoingo: Still has required a few resets to overcome crapolade blackholes, but far less common after passing block 400,000 compared to 250,000 -350,000 where PRB unordered shit spewing of blocks complicates things
BingoBoingo: no real srs diagnostics done. simply restarted bitcoind on freezes
a111: Logged on 2014-10-22 04:34 asciilifeform: but all the semantics of original must be preserved, to begin.
a111: Logged on 2016-12-20 19:16 asciilifeform: which potentially changes the semantics of EVERYTHING
a111: Logged on 2017-02-22 04:43 asciilifeform: (it'd require, absolutely, removing the locks. which ~unquantifiably alters the semantics of the entire thing.)
a111: Logged on 2017-03-09 20:33 asciilifeform: and incidentally this will also change the semantics of the block-saver, unless somehow kludged around (e.g. via locking)
mircea_popescu: it would actually be VERY INTERESTING to have a semantics- unpreserving client. let's see dude, wtf DOES happen ?
mircea_popescu: consider : ssh-tunnel ze bomb ? what diiference does it make.
mircea_popescu: well obviously. that's the idea. do it know while they are.
mircea_popescu: that's the fucking rule. something scares you ? DO IT NOW! now, that master's here, now thart you're still young and it heals, now, now, NOW.
BingoBoingo: asciilifeform: Well with these things even a mere 20 km can make a huge difference in timing
trinque: chilly and sunny here in TX today, been enjoying it on the balcony all day.
davout: so apparently "bitcoin unlimited" nodes are sniped off the network by some remote-crash bug
davout: i'm in their IRC room for lulz, it's absolutely delicious
trinque: inb4 bug being simply, a very large block.
Framedragger: :D so, if not "thing block", then CRASH (`assert(0)`)? A THOUSAND EYES
Framedragger: is there a way to short all these scamcoins easily? lol.
Framedragger: (right, this is basically *remote* probably-easy way to make it crash. nice knob.)
davout: BingoBoingo: will make an attempt
mircea_popescu: this is embarqassing, there's nobody there. buncha dorks on eg [marilyn] (~androirc@c-68-37-99-241.hsd1.mi.comcast.net): Android IRC Client wtf.
mircea_popescu: the "consumer computing community" functions approximately as a spotty porn offline backup.
pete_dushenski: asciilifeform: i'll be working on it between a couple of my nodes this afternoon. got a slave set up. now.... master
davout: BingoBoingo: feel free to edit ad lib obviously
mircea_popescu: * marilyn has quit (Ping timeout: 258 seconds) << dude... why. srsly nao.
Framedragger: well now that it's fixxord i'm certain it's smooth sailing from here on.
mircea_popescu: Framedragger the eth recipe is apparently un-fucking-escapeable, no matter what. like the fucking "restaurants" in the us, which WILL be feeding people HFCS no matter the fuck what and how it blows up.
Framedragger: "the audience doesn't change, why should the food change"
pete_dushenski: hfcs trend well underway : "Nobody was predicting what happened in Clarendon in 2016 -- a wave of failures that amounted to a local restaurant die-off. Within the space of a few months, nearly a dozen major restaurants closed their doors, many of them well liked by reviewers and locally popular, some of them fixtures in the neighborhood going back a decade or more."
Framedragger: pete_dushenski: wait they had GOOD REVIEWS ON YELP, omg HOW could they have failed
pete_dushenski: lol somehow their cousins and cheap friends weren't enough to buoy the whole establishment
ben_vulpes: > what is the name of your youngest child
ben_vulpes: because that's a constant string in the universe
ben_vulpes: > in what city does your nearest sibling live
ben_vulpes: oh OH and then while the little javascripty checkboxes next to the fields of pointlessness show green the ACTUAL POSTBACK VALIDATION FAILS
ben_vulpes: what the FUCK is the point of javascript if such RETARDS are going to wield it
Framedragger: i personally have a special place in my heart for the static content websites which refuse to render a single legible char without heavy javascript. not that the rendered text is full of glorious content, of course
trinque: Framedragger: is it one of those special places full of sadomasochism and despair?
Framedragger: trinque: hah. only pure unfiltered rage - cleaner that way
Framedragger: mircea_popescu: yer outta teh loop, you're supposed to abstract it with the great framework of jquery, no i mean angular, no i mean backbone, no i mean ember, no i mean react
Framedragger: ben_vulpes: you may have a point i'faith. that said, _why_ is she so fat?..
ben_vulpes: proud, upstanding slut with no pretension
mircea_popescu: in other lelz, aparently trilema ranks for "the forums of the most serene republic". with the right article, too.
ben_vulpes: Framedragger: that's not a nice way to speak about a girl who's so charitable
mircea_popescu: venereal disease is supposed to thin one out. this is shenaniganich.
mircea_popescu: ben_vulpes incidentally, ever run into one of those ? they're most common on teh western seaboard.
mircea_popescu: anyway, i guess that whole BUTT thing was moderately entertaining.
Framedragger: we missed out on quality butthurt because they have a private safespace channel heh.
ben_vulpes: mircea_popescu: one of which? venereal-laden?
trinque: aha, they'll ignore your system root certs because fuck you
Framedragger: asciilifeform: unrelated - re. fuckoff.c, aren't you supposed to register that sig_handler() to kernel first (via invoking signal()) for it to even work?
trinque: ben_vulpes: openbsd's package for the thing is neutered to use the system
Framedragger: asciilifeform: just fyi confirmed that in fuckoff.c, sig_handler() code is redundant as the function is not registered. (adding `signal(SIGHUP, sig_handler);` in main() changes behaviour and fixes this)
Framedragger: i'm not too sure why you even needed a handler for HUP, it closes by default
Framedragger: mircea_popescu: ^ you can remove that sig_handler definition from fuckoff.c (as well as the signal.h #include) fwiw, as that that code won't ever be executed