Show Idle (>14 d.) Chans


← 2020-07-18 | 2020-07-20 →
00:31 thimbronion asciilifeform: I added in some cacheing that should prevent repeated requests for the robots.txt file, but I need to let this run finish so sorry in advance for the log spam.
~ 8 hours 10 minutes ~
08:41 gregorynyssa http://logs.nosuchlabs.com/log/asciilifeform/2020-07-18#1016686 << no truly exemplary platform exists, but pre-Unicode Tcl comes closest, I would say. have you tried that yet? the next closest competitors would be Forth and Smalltalk. all three of these languages are "not quite LISPs."
08:41 snsabot Logged on 2020-07-18 19:59:23 asciilifeform: trinque: for me, the big wake-up re 'there is NO reasonably clean script lang atm' was when wrote 'litmus' in what ( asciilifeform naively thought was ) pure 'sh'
08:52 gregorynyssa http://logs.nosuchlabs.com/log/asciilifeform/2020-07-18#1016681 << might I ask what you mean exactly by "dethompsonizing?" do you simply mean, the absence of self-hosting? (a reference to that lecture Reflections on Trusting Trust) searched the logs, but still not sure.
08:52 snsabot Logged on 2020-07-18 19:55:54 asciilifeform: trinque: since you mentioned script langs: considering, after ffa, to attempt a 'dethompsonizing' simple gc-less scheme in asm, in style of 'M' as a scripting lang. can't speak for erryone, but i've wanted a <32kB scripting lang that 'compiles with bare hands' for many yrs.
~ 25 minutes ~
09:17 trinque no, he means one that doesn't require someone else's compiler to bootstrap.
09:21 trinque in re: tcl, it's just another unix string-munging language. unix-string-munging is the whole of the problem with unix, so tcl goes right on the pyre with the rest of them.
09:22 trinque why do I want a "not quite lisp" instead of a lisp?
09:24 gregorynyssa http://logs.nosuchlabs.com/log/asciilifeform/2020-07-19#1016709 << because Common Lisp is "bloated and unorthogonal" while Scheme is "minimal to the point of uselessness."
09:24 snsabot Logged on 2020-07-19 09:22:37 trinque: why do I want a "not quite lisp" instead of a lisp?
09:25 trinque don't quote other people's opinions to me. python is also bloated and unorthogonal, and tcl is an even worse shit-bolted-to-the-side monster by now.
09:27 gregorynyssa trinque: Tcl has cleaner semantics than Python, and is homoiconic. it has strong support for DSLs similar to the LISPs.
09:29 gregorynyssa (since its syntax/semantics are cleaner, it has earned a reputation of very low inconsistency between platforms/implementations.)
09:30 trinque wtf is this "it has earned a reputation" device you're using?
09:30 gregorynyssa http://logs.nosuchlabs.com/log/asciilifeform/2020-07-19#1016707 << thanks for the reply btw. I very much agree. I would like to build a language purely from the bedrock.
09:30 snsabot Logged on 2020-07-19 09:17:13 trinque: no, he means one that doesn't require someone else's compiler to bootstrap.
09:30 trinque 9 out of 10 doctors agree?
09:31 gregorynyssa my point is, I consider its semantics in a different league from Python, Perl, PHP, Ruby...
09:32 trinque when you offer up the thinking by which you so consider, I'll bother discussing it with you.
09:32 trinque meanwhile deedbot and freenode don't seem to be getting along today
09:36 gregorynyssa for instance, Tcl follows Smalltalk in not having control-structures. conditions and loops are just function-calls with delayed evaluation. that makes the semantics more concise.
09:38 trinque why does that not bias you towards smalltalk rather than tcl?
09:39 gregorynyssa I am a fan of both. however, Smalltalk has a heavier runtime.
09:45 gregorynyssa (also, I said pre-Unicode; the 8.0+ especially 8.5+ versions of Tcl have somewhat betrayed the language's original promise of transparency, no argument there)
~ 21 minutes ~
10:06 trinque I'm skeptical of anything that bills itself as a "scripting language" and therefore it gets away with things like that uplevel directive, because it's "not real programming"
10:08 trinque do I also understand correctly that tcl represents the result of every expression as a string?
10:09 trinque magic variables, bleh!
~ 34 minutes ~
10:44 gregorynyssa http://logs.nosuchlabs.com/log/asciilifeform/2020-07-19#1016726 << the "uplevel" primitive is a general way of allowing functions to choose whether they are pass-by-value or pass-by-reference. it enables other tricks as well, including, I believe, Scheme-like continuations.
10:44 snsabot Logged on 2020-07-19 10:06:49 trinque: I'm skeptical of anything that bills itself as a "scripting language" and therefore it gets away with things like that uplevel directive, because it's "not real programming"
10:55 gregorynyssa magic variables were a mistake. tcl_precision struck me as particularly inelegant.
10:59 gregorynyssa http://logs.nosuchlabs.com/log/asciilifeform/2020-07-19#1016727 << before 8.5, everything was either: a string, a (nestable) list of strings, or a hashmap. lists of strings were encoded as strings (like JSON but simpler; the AST of every program itself was represented this way). a hashmap was a (one level) mapping from string to string. file-handles and sockets were represented as strings (such as "file1," "file2," ...) which were understood
10:59 snsabot Logged on 2020-07-19 10:08:10 trinque: do I also understand correctly that tcl represents the result of every expression as a string?
11:04 gregorynyssa the language has first-class environments. you can construct arbitrary ASTs and environments/scopes at runtime, and eval any of them together.
~ 1 hours 18 minutes ~
12:23 asciilifeform http://logs.nosuchlabs.com/log/asciilifeform/2020-07-19#1016703 << i confess, still entirely stumped re what is appealing in tcl... ( as for smalltalk, iirc already commented; and as for forth, in fact wrote a forth... )
12:23 snsabot Logged on 2020-07-19 08:41:53 gregorynyssa: http://logs.nosuchlabs.com/log/asciilifeform/2020-07-18#1016686 << no truly exemplary platform exists, but pre-Unicode Tcl comes closest, I would say. have you tried that yet? the next closest competitors would be Forth and Smalltalk. all three of these languages are "not quite LISPs."
12:24 asciilifeform http://logs.nosuchlabs.com/log/asciilifeform/2020-07-19#1016707 << correct. recall the '80s comp. enthusiast magazines? where you ~typed in~ proggy in hex ? a proper script lang oughta be bootstrappable this way.
12:24 snsabot Logged on 2020-07-19 09:17:13 trinque: no, he means one that doesn't require someone else's compiler to bootstrap.
12:25 asciilifeform http://logs.nosuchlabs.com/log/asciilifeform/2020-07-19#1016710 << the funny thing is, these are ~both~ fixable...
12:25 snsabot Logged on 2020-07-19 09:24:20 gregorynyssa: http://logs.nosuchlabs.com/log/asciilifeform/2020-07-19#1016709 << because Common Lisp is "bloated and unorthogonal" while Scheme is "minimal to the point of uselessness."
12:26 asciilifeform http://logs.nosuchlabs.com/log/asciilifeform/2020-07-19#1016722 << imho this kinda thing (in other threads i referred to it as 'haskellism') is a sin.
12:26 snsabot Logged on 2020-07-19 09:36:47 gregorynyssa: for instance, Tcl follows Smalltalk in not having control-structures. conditions and loops are just function-calls with delayed evaluation. that makes the semantics more concise.
12:27 asciilifeform recall t. edison's last biz venture ? chances are you don't: he sold cement furniture. 'clean, concise'...
12:28 asciilifeform http://logs.nosuchlabs.com/log/asciilifeform/2020-07-19#1016729 << oblig.
12:28 snsabot Logged on 2020-07-19 10:44:13 gregorynyssa: http://logs.nosuchlabs.com/log/asciilifeform/2020-07-19#1016726 << the "uplevel" primitive is a general way of allowing functions to choose whether they are pass-by-value or pass-by-reference. it enables other tricks as well, including, I believe, Scheme-like continuations.
12:29 asciilifeform http://logs.nosuchlabs.com/log/asciilifeform/2020-07-19#1016732 << this is supposed to be appealing ?! where ~any nontrivial proggy will run in geological time ?
12:29 snsabot Logged on 2020-07-19 10:59:36 gregorynyssa: http://logs.nosuchlabs.com/log/asciilifeform/2020-07-19#1016727 << before 8.5, everything was either: a string, a (nestable) list of strings, or a hashmap. lists of strings were encoded as strings (like JSON but simpler; the AST of every program itself was represented this way). a hashmap was a (one level) mapping from string to string. file-handles and sockets were represented as strings (such as "file1," "file2," ...) which were understood
12:31 asciilifeform gregorynyssa: outta curiosity -- didja ever actually program in that horror, or simply find appealing in theory ?
12:31 asciilifeform !w poll
12:31 watchglass Polling 12 nodes...
12:31 watchglass 205.134.172.4:8333 : (172-4.core.ai.net) Alive: (0.083s) V=70001 (/therealbitcoin.org:0.7.0.1/) Jumpers=0x1 (TRB-Compat.) Blocks=639901
12:31 watchglass 205.134.172.27:8333 : Alive: (0.084s) V=99999 (/therealbitcoin.org:0.9.99.99/) Jumpers=0x1 (TRB-Compat.) Blocks=639901 (Operator: asciilifeform)
12:31 watchglass 205.134.172.26:8333 : Alive: (0.082s) V=99999 (/therealbitcoin.org:0.9.99.99/) Jumpers=0x1 (TRB-Compat.) Blocks=639901
12:31 watchglass 205.134.172.6:8333 : (172-6.core.ai.net) Alive: (0.084s) V=99999 (/therealbitcoin.org:0.9.99.99/) Jumpers=0x1 (TRB-Compat.) Blocks=639901
12:31 watchglass 108.31.170.3:8333 : (pool-108-31-170-3.washdc.fios.verizon.net) Alive: (0.158s) V=99999 (/therealbitcoin.org:0.9.99.99/) Jumpers=0x1 (TRB-Compat.) Blocks=639901 (Operator: asciilifeform)
12:31 watchglass 192.151.158.26:8333 : Alive: (0.141s) V=70001 (/therealbitcoin.org:0.7.0.1/) Jumpers=0x1 (TRB-Compat.) Blocks=639901
12:31 watchglass 143.202.160.10:8333 : Alive: (0.161s) V=70001 (/therealbitcoin.org:0.7.0.1/) Jumpers=0x1 (TRB-Compat.) Blocks=639901
12:31 watchglass 208.94.240.42:8333 : Alive: (0.234s) V=99999 (/therealbitcoin.org:0.9.99.99/) Jumpers=0x1 (TRB-Compat.) Blocks=639901
12:31 watchglass 176.9.59.199:8333 : (static.199.59.9.176.clients.your-server.de) Alive: (0.239s) V=99999 (/therealbitcoin.org:0.9.99.99/) Jumpers=0x1 (TRB-Compat.) Blocks=235865 (Operator: jurov)
12:31 watchglass 213.109.238.156:8333 : Alive: (0.421s) V=99999 (/therealbitcoin.org:0.9.99.99/) Jumpers=0x1 (TRB-Compat.) Blocks=639901
12:31 watchglass 188.121.168.69:8333 : (rev-188-121-168-69.radiolan.sk) Alive: (0.365s) V=99999 (/therealbitcoin.org:0.9.99.99/) Jumpers=0x1 (TRB-Compat.) Blocks=639902
12:31 watchglass 103.36.92.112:8333 : (terebe.ns01.net) Alive: (0.816s) V=99999 (/therealbitcoin.org:0.9.99.99/) Jumpers=0x1 (TRB-Compat.) Blocks=639901
~ 39 minutes ~
13:10 adlai gregorynyssa: re:unorthogonality, C++ is also unorthogonal to the point of eating its own spine, yet somehow one or two blubhubs still seem to hire C++ programmers
13:12 adlai the old "fqa" where you can just gram-schmidt the antifeatures until all that is left is orthogonal can be seen as a warped skew-parallel to greenspun's 10th
13:19 trinque I'm by now highly suspect of any "no, this, *THIS* is the one string munger to rule them all" in unixland
13:19 trinque sed, *awk, m4, tcl, *sh, fuck me, I'm tired of this list already
13:20 trinque the worst part is that they use ALL OF THEM TOGETHER
13:20 trinque I'm over here wondering why I can't get flex 2.5.39 to compile statically, as it does not respect LDFLAGS=-static
13:21 trinque if the GNU fuckwits intended the whole industry to so completely choke on their wad that it would never move on, congrats.
13:23 trinque lemme tell you just how long a project my assertion that "fuck you, there will only be one autoconf in this pile" has made.
13:24 trinque because the various dipshits all used different string-mungers and different versions of same string mungers in the dependencies leading up to gcc.
13:24 trinque and sure, could instead use whomever's pre-generated Makefiles and pretend autoconf isn't there, and then forever muck about in generated code. no.
13:25 trinque so no, this "I found in this found-object a little piece of myself" dies in the fire.
13:26 asciilifeform trinque: and on top of this 'thought you hit bottom, then somebody knocks from below', wait till you consider the q of 'why does 'flex' even exist' etc
13:26 trinque oh, I know.
13:27 * asciilifeform reminds readers that orig. DEC unixen did useful work in 256-512kB total storage..
13:31 trinque my wife commented to me once that "revolutionaries always fixate on the putsch, and never on what comes after".
13:31 asciilifeform http://logs.nosuchlabs.com/log/asciilifeform/2020-07-19#1016763 << almost certainly this aint news, but asciilifeform cpp for a living, for years. (and on microshit, no less.)
13:31 snsabot Logged on 2020-07-19 13:10:45 adlai: gregorynyssa: re:unorthogonality, C++ is also unorthogonal to the point of eating its own spine, yet somehow one or two blubhubs still seem to hire C++ programmers
13:32 asciilifeform in microshitistan, thing is still king. and 'jobsecurity lang.'
13:32 adlai well now, some folks studied the liquishit for years, some only dabble, and others have it poured on them until they drown
13:32 asciilifeform trinque: wife aint wrong, there
13:32 trinque I think this is apt for the GNUtards, only were concerned with infecting commercial unix with their spread shit
13:34 * adlai doubts that eugenics and muskism has yet reached the point where babies are taught C++ before they learn latin, although perhaps this is not far away.
13:34 asciilifeform trinque: see oblig. naggum also.
13:35 trinque bless him, "go away", he was such an optimist.
13:35 * trinque has read this one
13:37 asciilifeform adlai: if were not for trb legacy pile o'shit, i would not recommend to any n00b to 'dabble' in cpp. it's arguably 'a cobol', will slowly destroy yer brain if you start to take it seriously, unless you have ~very~ strong immune system
13:37 trinque and psychological autoimmune disorders are also a thing.
13:39 asciilifeform the fella who remarked to a young asciilifeform 'after N lines of coad, you will smash yer comp and build unabomber cabin' was a cpp programmer.
13:49 asciilifeform elsewhere: without biting the bait of the butthurt of folx who couldn't be arsed to bake a wot isp, large or small -- there are actually quite a few trb-compat. nodes not advertised anywhere but the p2p net itself, and it is possible to map'em out
13:49 snsabot Logged on 2020-05-06 14:28:35 asciilifeform: !w peers 205.134.172.27
13:50 asciilifeform i even posted 100% of the necessary coad, to do this. if someone gives a damn, can map.
13:52 asciilifeform subj.
14:02 whaack asciilifeform: ostensibly "important" folks have a deal with some miner/other node - "plz keep me up to date with block data, and do not advertise my node as I wish to remain in my closed network."
14:03 asciilifeform whaack: the beauty is, tho, that you can't have an entirely 'closed net' and still bitcoinate.
14:03 asciilifeform (well, rather, you can, it's called a fork..)
14:03 asciilifeform as for 'plz keep me up to date but not advertise mine', it's a built-in knob, called '-connect'...
14:04 asciilifeform http://logs.nosuchlabs.com/log/asciilifeform/2020-07-19#1016766 << before this slips away -- i'd include php in the list.
14:04 snsabot Logged on 2020-07-19 13:19:48 trinque: sed, *awk, m4, tcl, *sh, fuck me, I'm tired of this list already
14:04 trinque sure, php perl python ruby aaaa there goes my gall again
14:05 asciilifeform 9000 sadlangs to half-arsedly do exactly same job.
~ 32 minutes ~
14:37 whaack asciilifeform: My point is that although the graph is fully connected, it's not true that every node can see the whole graph.
14:39 asciilifeform whaack: this is elementarily true, nodes that operate via '-connect' -- or normally, but sitting behind nat -- aint connectable to externally.
14:39 asciilifeform (and don't get advertised as peers by peers)
14:39 asciilifeform this is actually most of the net (when taking whole net , prb etc included) by mass.
14:42 asciilifeform whaack: at 1 time, asciilifeform wrote a mechanism, 'wires', to make it simple to create 'private' cliques of noads that can connect to one another. but abandoned from lack of in-wot interest. (didn't help that the 1st version was buggy, either)
14:43 asciilifeform i did this when still entertained the hypothesis that trb suffered slow sync on acct of interference from isps/nsa/etc
14:43 whaack alright, i guess i was confused by your point that you can map out the 'nodes not advertised anywhere but the pwp net itself'
14:43 whaack p2p*
14:43 asciilifeform can only map out the publicly-talkable ones, aha
14:44 whaack you still can't map out the whole graph, which makes it silly to say 'the nodes i can see are the only nodes!'
14:45 asciilifeform defo can't map out nodes that dun answer on public ports.
14:45 asciilifeform however, if you think you found a public noad that is trb-compat, can, e.g. :
14:45 asciilifeform !w probe 103.36.92.112
14:45 watchglass 103.36.92.112:8333 : (terebe.ns01.net) Alive: (0.775s) V=99999 (/therealbitcoin.org:0.9.99.99/) Jumpers=0x1 (TRB-Compat.) Blocks=639920
14:46 asciilifeform ( see src re how decides 'trb-compat' , is quite simple )
14:47 asciilifeform prb's will look like e.g. :
14:47 asciilifeform !w probe 176.122.101.230
14:47 watchglass 176.122.101.230:8333 : Alive: (0.276s) V=70015 (/Satoshi:0.19.1/) Jumpers=0x409 (PRB: WITNESS;NETWORK_LIMITED) Return Addr=0.0.0.0:0 Blocks=639921
14:47 asciilifeform ^ bot lists evidence re 'why this aint trb'
14:49 whaack asciilifeform: nifty, i'll try to give the source a read when I have a moment
14:49 asciilifeform note that currently milking prb noades for peers doesn't work, they use an incompat. protocol .
14:49 snsabot Logged on 2020-03-01 14:38:20 asciilifeform: shinohai: this btw confirmed. i did experiment, if one sends that extra byte, prb noades do in fact send addrs. (and yes some of'em ipv6, have to be thrown out)
14:50 asciilifeform whaack: you may or may not need this for yer application.
14:50 asciilifeform (for 'block viewer' is of 0 use . this is for if yer mapping out the net etc )
14:51 asciilifeform all of this is strictly in re the q of 'what can be determined re size/shape of bitcoin p2p net' .
~ 3 hours 11 minutes ~
18:03 shinohai https://www.zerohedge.com/technology/zero-logs-vpn-company-exposes-millions-user-logs <<< evidently, another COVID-19 casualty. xD
~ 42 minutes ~
18:46 asciilifeform shinohai: lulzy.
18:46 asciilifeform shinohai: tastes great w/ earlier sauce.
18:47 shinohai aha!
~ 1 hours 40 minutes ~
20:28 asciilifeform shinohai: it entertains me that incidents like the linked item have 0 measurable effect on vpnism enthusiasts. ( rather like e.g. n-th, n+1st, ... , 'silkroad' trials had similarly 0 effect on the tor smokers )
20:29 asciilifeform will paste for the l0gz :
20:29 asciilifeform 'A Hong Kong-based UFO VPN - which claims a 'zero logs' policy, maintained a database without any password, exposing over 20 million user logs per day which consisted of 894 GB of data. The logs reportedly included passwords, IP addresses, geographical location, connection timestamps, session tokens, device information and the OS used. This is in stark contrast to UFO VPN's stated privacy policy that "We do not track user activ
20:29 asciilifeform ities outside of our Site, nor do we track the website browsing or connection activities of users who are using our Services."'
20:30 shinohai "We don't track, but everyone else can!" (tm)
20:31 asciilifeform oblig. see also.
20:32 shinohai But yeah, you're right, burned customer will simply move on to *yet another vpn provider* "surely this one is better"
20:33 asciilifeform the chukchas and neighbouring tribes had similar religious practice. if a god 'wasn't pulling his share', i.e. praying to $idol wasn't delivering 'luck', they'd smear the idol with shit, or -- in 'extreme' failures -- throw in fire, and pick another..
20:34 asciilifeform at no pt would question the 'algo' tho.
~ 1 hours 43 minutes ~
22:17 asciilifeform dpb: problem with box ?
22:17 dpb asciilifeform, terribly sorry, I am an idiot, i shutdown my rockchip by accident
22:18 dpb no rush, just wanted to give you a heads up as i don't think i can turn it back on from where i'm at
22:18 dpb gtg now
22:18 asciilifeform ugh ok. this'll req. a visit. will count against your 2h/y repair .
22:18 dpb that's fine. thank you
22:18 asciilifeform will do it when i wake up ( about to bed ) if that's ok
22:18 asciilifeform aite.
22:18 dpb yep
22:18 dpb gn
22:25 feedbot http://mvdstandard.net/2020/07/california-juvenile-forcibly-dosed-with-estrogen-in-detention-facility-for-behavior-modification-purposes/ << The Montevideo Standard -- California Juvenile Forcibly Dosed With Estrogen In Detention Facility For Behavior Modification Purposes
~ 1 hours 23 minutes ~
23:49 feedbot http://mvdstandard.net/2020/07/yves-rausch-black-forest-rambo-allegedly-captured-by-police/ << The Montevideo Standard -- Yves Rausch, Black Forest Rambo Allegedly Captured By Police
← 2020-07-18 | 2020-07-20 →