Show Idle (>14 d.) Chans


← 2022-01-15 | 2022-01-17 →
gregory5: http://logs.nosuchlabs.com/log/asciilifeform/2022-01-15#1073310 << I am going to wait for the new spec. before finalizing my station, hope you don't mind.
dulapbot: Logged on 2022-01-15 22:47:09 asciilifeform: will give folx time to comment, then cut a 0xFB patch some time tomorrow evening.
gregory5: http://logs.nosuchlabs.com/log/asciilifeform/2022-01-16#1073313 << I read that book a few months ago. the writing style was very strange.
dulapbot: Logged on 2022-01-16 00:46:36 scoopbot: New post on A Syndication of Verisimilitudes: A Review of ``Softword: Provenance for the Word 'Software''' and Other Work by Paul Niquette
gregory5: {by calling ``bugs'' in software ``flaws''} << I do this as well, "verisimilitude," but I also use the word "error."
gregory5: some programmers will object, saying that "error" implies the occurrence of a flaw (at run-time), not the flaw itself.
gregory5: however, "error" does not have that connotation in philosophy and theology.
gregory5: http://niquette.com/books/101words/be.htm << Niquette is a proponent of "E Prime."
gregory5: the man's style, and way of thinking, is highly reminiscent of other Catholic intellectuals which I have met.
gregory5: white Americans who are well-read in English, but do not have the English heritage.
mats: asciilifeform: i dunno, good questions
asciilifeform: mats: gnarly. asciilifeform doesn't have a sufficiently detailed grasp of that particular shitcoin to 100% grasp on 1st pass
asciilifeform: http://logs.nosuchlabs.com/log/asciilifeform/2022-01-16#1073314 << which version of spec do you intend to wait for the implementation for ?
dulapbot: Logged on 2022-01-16 01:23:25 gregory5: http://logs.nosuchlabs.com/log/asciilifeform/2022-01-15#1073310 << I am going to wait for the new spec. before finalizing my station, hope you don't mind.
asciilifeform: gregory5: for instance asciilifeform is cleaning up 0xfb. but even after vpatch, will need to wait for folx to fix the coad
asciilifeform sadly presently hasn't the time to do w/ own hands the latter
dulapbot: Logged on 2022-01-16 01:26:13 gregory5: http://logs.nosuchlabs.com/log/asciilifeform/2022-01-16#1073313 << I read that book a few months ago. the writing style was very strange.
asciilifeform: http://logs.nosuchlabs.com/log/asciilifeform/2022-01-16#1073322 << dunno, imho there was nuffin obv. 'aquinasistic' about it
dulapbot: Logged on 2022-01-16 01:39:16 gregory5: the man's style, and way of thinking, is highly reminiscent of other Catholic intellectuals which I have met.
asciilifeform: $ticker btc usd
busybot: Current BTC price in USD: $43435.76
asciilifeform: !w poll
watchglass: Polling 14 nodes...
watchglass: 205.134.172.26:8333 : Alive: (0.081s) V=99999 (/therealbitcoin.org:0.9.99.99/) Jumpers=0x1 (TRB-Compat.) Return Addr=0.0.0.0:8333 Blocks=718998
watchglass: 205.134.172.6:8333 : (172-6.core.ai.net) Alive: (0.021s) V=99999 (/therealbitcoin.org:0.9.99.99/) Jumpers=0x1 (TRB-Compat.) Return Addr=0.0.0.0:8333 Blocks=718998
watchglass: 54.39.156.171:8333 : (ns562940.ip-54-39-156.net) Alive: (0.059s) V=99999 (/therealbitcoin.org:0.9.99.99/) Jumpers=0x1 (TRB-Compat.) Blocks=718998
watchglass: 205.134.172.4:8333 : (172-4.core.ai.net) Alive: (0.082s) V=70001 (/therealbitcoin.org:0.7.0.1/) Jumpers=0x1 (TRB-Compat.) Blocks=718998
watchglass: 71.191.220.241:8333 : (pool-71-191-220-241.washdc.fios.verizon.net) Alive: (0.131s) V=99999 (/therealbitcoin.org:0.9.99.99/) Jumpers=0x1 (TRB-Compat.) Blocks=718998 (Operator: asciilifeform)
watchglass: 205.134.172.27:8333 : Alive: (0.144s) V=99999 (/therealbitcoin.org:0.9.99.99/) Jumpers=0x1 (TRB-Compat.) Blocks=718998 (Operator: asciilifeform)
watchglass: 205.134.172.28:8333 : Alive: (0.144s) V=99999 (/therealbitcoin.org:0.9.99.99/) Jumpers=0x1 (TRB-Compat.) Return Addr=0.0.0.0:8333 Blocks=718998 (Operator: whaack)
watchglass: 208.94.240.42:8333 : Alive: (0.143s) V=99999 (/therealbitcoin.org:0.9.99.99/) Jumpers=0x1 (TRB-Compat.) Blocks=718998
watchglass: 54.38.94.63:8333 : (ns3140226.ip-54-38-94.eu) Alive: (0.262s) V=88888 (/therealbitcoin.org:0.8.88.88/) Jumpers=0x1 (TRB-Compat.) Blocks=718998
watchglass: 94.176.238.102:8333 : (2ppf.s.time4vps.cloud) Alive: (0.309s) V=99999 (/therealbitcoin.org:0.9.99.99/) Jumpers=0x1 (TRB-Compat.) Blocks=718910
watchglass: 82.79.58.192:8333 : (static-82-79-58-192.rdsnet.ro) Alive: (0.338s) V=99999 (/therealbitcoin.org:0.9.99.99/) Jumpers=0x1 (TRB-Compat.) Blocks=718910
watchglass: 103.36.92.112:8333 : (terebe.ns01.net) Alive: (0.583s) V=99999 (/therealbitcoin.org:0.9.99.99/) Jumpers=0x1 (TRB-Compat.) Blocks=718998
watchglass: 75.106.222.93:8333 : Alive: (0.452s) V=99999 (/therealbitcoin.org:0.9.99.99/) Jumpers=0x1 (TRB-Compat.) Blocks=718998
watchglass: 143.202.160.10:8333 : Busy? (No answer in 100 sec.)
asciilifeform: mats: reread your meth fishwrap link, seems that the authors blame impurities. but unsurprisingly didn't propose the simple solution of 'legalize'(tm)(r) where 100% pure product simply sold in shops otc for pennies.
asciilifeform: 'warriors on drugs' simply not interested in even admitting that folx take'em for reasons which often make sense, i.e. where life in usaschwitz in fact suxx palpably less on the dope than off
dulapbot: Logged on 2022-01-15 16:05:27 asciilifeform: the 'permanently changed' thing may even be that the eater (an overworked 'blue collar') simply realizes that 'yes it is possible to be awake'
dulapbot: Logged on 2022-01-15 16:07:52 signpost: pleb gets knocked from two jobs to three to stay alive, finds crank
asciilifeform: in ru, half dozen rulers historically (most recently gorby) tried to clamp down on vodka, w/ similar results
asciilifeform: life in 'nigeria with snow' unendurable for over9000 folx w/out subj...
asciilifeform: mats: observe also that ~0 whine pieces in fishwraps re wallstreet folx and their (generally pure, being obtained via Official channels) meth etc
dulapbot: Logged on 2022-01-15 13:20:12 mats: mentally,” Jobe said. “Before, we didn’t keep anybody more than nine months. Now we’re running up to 14 months, because it’s not until six or nine months that we finally find out who we got.” Some can’t remember their life before jail. “It’s not unusual for them to ask what they were found guilty of and sentenced to,” she said.
asciilifeform: nasdaq floor monkey wanting to be awake to do his work or party after is a++ Officially. but welder or trucker, ohnoez, 'addict', 'how do we cure'
asciilifeform: dopaminergics dun produce physical 'withdrawal' in the style of opium. the 'withdrawal' is simply 'normal life', which for over9000 folx simply sux.
dulapbot: Logged on 2022-01-16 11:26:00 asciilifeform: gregory5: for instance asciilifeform is cleaning up 0xfb. but even after vpatch, will need to wait for folx to fix the coad
gregory5: http://logs.nosuchlabs.com/log/asciilifeform/2022-01-16#1073331 << that was how I heard of the book, obviously.
gregory5: http://logs.nosuchlabs.com/log/asciilifeform/2022-01-16#1073333 << he has that native-but-not-native way of writing. hard to describe.
dulapbot: Logged on 2022-01-16 11:28:09 asciilifeform: http://logs.nosuchlabs.com/log/asciilifeform/2022-01-16#1073322 << dunno, imho there was nuffin obv. 'aquinasistic' about it
gregory5: http://niquette.com/books/101words/toc.htm << this is also sort of Thomistic.
gregory5: willing to release the tarball of my code, if people don't make fun of me for using Java.
gregory5: http://logs.nosuchlabs.com/log/asciilifeform/2022-01-16#1073360 << if the allegations about Wall Street and illicit drugs are true, the implications are quite horrifying, more than at first glance.
dulapbot: Logged on 2022-01-16 11:39:07 asciilifeform: mats: observe also that ~0 whine pieces in fishwraps re wallstreet folx and their (generally pure, being obtained via Official channels) meth etc
gregory5: likewise with similar allegations which I have heard about the Ivy League universities.
jonsykkel: readed section 4 good stuf. intake algo is now quite diffrent to what i ended up with to make everything work, gonna have to put changes in program before can give very meaningful comments. maybe most significant diff is that i added a third buffer for messages that dont fit anywhere in chain, where they sit and wait a sec or so for getdata response, to avoid chain breaking if packets are dropped
jonsykkel: or reordered, or if in process of catching up cuz station was turned off etc. maybe this is outside scope of spec anyway, dunno
asciilifeform: http://logs.nosuchlabs.com/log/asciilifeform/2022-01-16#1073372 << lol! but wainot ( asciilifeform promises not to make fun of even cobol implementation, if such appears )
dulapbot: Logged on 2022-01-16 11:55:07 gregory5: willing to release the tarball of my code, if people don't make fun of me for using Java.
asciilifeform: gregory5: asciilifeform doesn't keep a javatron around and likely will not be able to test tho
asciilifeform: http://logs.nosuchlabs.com/log/asciilifeform/2022-01-16#1073376 << actually asciilifeform is writing even nao a revised-again sect4 where third, 'output buffer', entirely as you described
dulapbot: Logged on 2022-01-16 12:06:07 jonsykkel: readed section 4 good stuf. intake algo is now quite diffrent to what i ended up with to make everything work, gonna have to put changes in program before can give very meaningful comments. maybe most significant diff is that i added a third buffer for messages that dont fit anywhere in chain, where they sit and wait a sec or so for getdata response, to avoid chain breaking if packets are dropped
asciilifeform: (normally output buffer empty, console outputs go straight to console. but 'clutch engages' if there's a getdata hole)
asciilifeform: i.e. if a msg makes it to output stage where the chain doesn't link up with the immed. chronological predecessor
asciilifeform attempting to remove all 'handwaves' from spec and turn it into sumthing reasonably like a straight pseudocoad, implementable w/out having to bash head against problems which 'hey you must solve' but spec aint saying just how
asciilifeform: jonsykkel, thimbronion , et al : lemme know if the new 'unified' sect4 algo makes 100% sense to you. imho the original one asciilifeform came up with was braindamaged and just about impossible to implement deterministically
punkman: "Radakin said there had been 'a phenomenal increase' in Russian submarine activity over the past 20 years, adding: 'Russia has grown the capability to put at threat those undersea cables and potentially exploit them.' "
gregory5: http://logs.nosuchlabs.com/log/asciilifeform/2022-01-16#1073380 << I would like to see the JVM in the form of a single executable (statically linked of course). it is tragic that this doesn't already exist.
dulapbot: Logged on 2022-01-16 12:37:26 asciilifeform: gregory5: asciilifeform doesn't keep a javatron around and likely will not be able to test tho
gregory5: http://logs.nosuchlabs.com/log/asciilifeform/2022-01-16#1073385 << that is such a noble goal. if everyone did this, the exegesis of RFCs might have been one of the pillars of the undergraduate computer-science curriculum.
dulapbot: Logged on 2022-01-16 12:40:56 asciilifeform: attempting to remove all 'handwaves' from spec and turn it into sumthing reasonably like a straight pseudocoad, implementable w/out having to bash head against problems which 'hey you must solve' but spec aint saying just how
asciilifeform: gregory5: asciilifeform hesitated to do this, it will inevitably make for a more obese spec. but in the end decided it aint escapable. atm wrapping up a 'data types' section, will upload half-finished item shortly so folx can see what meant.
asciilifeform still in the midst of actually massaging whole thing to refer to that sect., rather than the adhoc peppering of 'here's a 256-bit...' prev.
asciilifeform will bbl
gregory5: asciilifeform: section 3.1.4, the phrase "at most N characters in length... at all times occupies precisely N bytes" is contradictory (subject to one's understanding of hypotaxis/parataxis).
gregory5: might be better to say: A UPestString[N] is a string encoded in UTF-8, at all times occupying precisely N bytes, the unused trailing bytes being set to zero.
gregory5: then for parallelism you can change the definition of APestString to: a string encoded in pure ASCII, at all times occupying precisely N bytes, the unused trailing bytes being set to zero.
gregory5: if you don't mind being Laconian, there is no reason to mention "characters" in this section
asciilifeform: gregory5: racking brain, but can't see why contradictory
asciilifeform: can be e.g. 0, 1, etc. characters, but still occupies N bytes.
asciilifeform: in ascii or utfism, can fit at most N chars (which are distinct, in the case of the latter, from bytes)
asciilifeform not historically enamoured of utfism, but if can't paste e.g. ru string, as in trad irc, won't use
asciilifeform: the major departure from tradition here is that handles are forced into ascii. which imho is absolutely essential: permitting homoglyph impersonation of nicks is profoundly retarded and major weakness of classical irc
asciilifeform genuinely bbl
cgra: http://logs.nosuchlabs.com/log/asciilifeform/2022-01-16#1073393 << document type starting to really look like what the supposed trb documentation also could look like...
dulapbot: Logged on 2022-01-16 13:28:07 asciilifeform: gregory5: asciilifeform hesitated to do this, it will inevitably make for a more obese spec. but in the end decided it aint escapable. atm wrapping up a 'data types' section, will upload half-finished item shortly so folx can see what meant.
cgra considering publishing similar text re trb, to support further discussion of sync (have had such a draft around getting shape since the first months of study)
gregory5: http://logs.nosuchlabs.com/log/asciilifeform/2022-01-16#1073401 << it depends on whether "at most N characters in length" and "The string at all times occupies precisely N bytes." are interpreted as hypotaxis or parataxis.
dulapbot: Logged on 2022-01-16 14:01:33 asciilifeform: gregory5: racking brain, but can't see why contradictory
gregory5: that is, whether the second sentence is subordinate to the first, or whether the sentences work together in parallel.
gregory5: this might have been a trivial contention, sorry.
dulapbot: Logged on 2022-01-16 12:38:37 asciilifeform: http://logs.nosuchlabs.com/log/asciilifeform/2022-01-16#1073376 << actually asciilifeform is writing even nao a revised-again sect4 where third, 'output buffer', entirely as you described
jonsykkel: http://logs.nosuchlabs.com/log/asciilifeform/2022-01-16#1073386 << so far seems to make sense but i need to code first to be able to see any problems anyway
dulapbot: Logged on 2022-01-16 12:42:31 asciilifeform: jonsykkel, thimbronion , et al : lemme know if the new 'unified' sect4 algo makes 100% sense to you. imho the original one asciilifeform came up with was braindamaged and just about impossible to implement deterministically
jonsykkel: there're some details missing from algo that i can see: chek protocol version, invalid command, handling ignore pakets (specially with regard to rekeying), update peer packet stamp + key specific stamp + at entry
jonsykkel: order of operations of these not entirely self explanatory i think anyway
jonsykkel: 100%agre that eliminating handwaving > fat spec concerns
jonsykkel: otherwise guarnanteed to end up with buncha pestrons all doing slightly diff stuff
asciilifeform: jonsykkel: re invalid cmd, we already have 'packet is rejected' neh
asciilifeform: the timestamps thing is arguably not req'd to specify in protocol (it concerns strictly client ui) tho imho 100% oughta show last-valid-packet-time in AT view
jonsykkel: indeed
asciilifeform: still btw not certain what to do w/ versions
asciilifeform: (reject packets from later versions? or not?)
asciilifeform: prolly oughta, or at some pt may end up w/ nonsense when an ancient client tries to talk to later 1
jonsykkel: ye dunno wat makes sense, if do that will force entire net to upgrade at same time
asciilifeform: well you can't actually force folx to move, lol
asciilifeform: they'll move or not on own steam
asciilifeform: (we aint baking a prb, lol)
jonsykkel: u are correct of course
asciilifeform: http://logs.nosuchlabs.com/log/asciilifeform/2022-01-16#1073410 << prolly oughta rephrase; tho atm unsure precisely in which dir
dulapbot: Logged on 2022-01-16 14:44:12 gregory5: http://logs.nosuchlabs.com/log/asciilifeform/2022-01-16#1073401 << it depends on whether "at most N characters in length" and "The string at all times occupies precisely N bytes." are interpreted as hypotaxis or parataxis.
asciilifeform: http://logs.nosuchlabs.com/log/asciilifeform/2022-01-16#1073409 << would be a+++, and rlyoughta've existed 10+y ago
dulapbot: Logged on 2022-01-16 14:14:28 cgra: considering publishing similar text re trb, to support further discussion of sync (have had such a draft around getting shape since the first months of study)
asciilifeform: http://logs.nosuchlabs.com/log/asciilifeform/2022-01-16#1073416 << imho the new sect.4 is simpleenuff that oughta be 'walkable in head' (try)
dulapbot: Logged on 2022-01-16 15:19:38 jonsykkel: http://logs.nosuchlabs.com/log/asciilifeform/2022-01-16#1073386 << so far seems to make sense but i need to code first to be able to see any problems anyway
asciilifeform: i.e. the automagick 'knocking out' of hearsays when immed. copy of msg is received is now simple logical consequence of the latter going straight into the longbuffer
jonsykkel: tried and failed. depends very much on capacity of head in question imo
asciilifeform interprets this as a sign that thing still needs work
asciilifeform: the aim is 'worx obviously like kalash'
jonsykkel: dunno, always need code in front of face to "think clearly"
verisimilitude: http://logs.nosuchlabs.com/log/asciilifeform/2022-01-16#1073405 It's tangential, but the ideal computing system I wish to bring about would solve this nicely, by forcing such text to be in one language, and making it known which language were being used.
dulapbot: Logged on 2022-01-16 14:05:39 asciilifeform: the major departure from tradition here is that handles are forced into ascii. which imho is absolutely essential: permitting homoglyph impersonation of nicks is profoundly retarded and major weakness of classical irc
verisimilitude: The insanity of Unicode is thinking that languages, unlike people, can peacefully coexist, as a colourful way of putting it.
verisimilitude: An idiot would argue ``B-But now we need special software for each language.'', while his better would retort that it would be needed regardless, to have anything decent.
verisimilitude: Another retort I've seen from idiots is that sometimes languages are heavily mixed, but my system poses no threat to such mixing.
verisimilitude: No differently than having a sane computing system, idiots not only justify the current ways, they argue no other ways should be allowed to exist.
asciilifeform: verisimilitude: pre-unicodism things actually worked rather like this
asciilifeform: (w/out, of course, anyffin the way of a standard means to indicate that e.g. next 9000 bytes are in jp or whatnot, had to guess)
asciilifeform still has hdd fulla ru txts encoded in ancient 'koi8' etc
gregory5: to my understanding, part of the intent of Unicode was that program-authors ought to define their own subsets which the program can perfectly support.
asciilifeform: yeawell
gregory5: instead, Unicode became supported at the language-level (as with Java's built-in UTF-16 support) and programmers began working with it blindly.
gregory5: moreover, a sort of Political Correctness began to form around the standard. anyone not using Unicode was stigmatized.
verisimilitude: Regarding the first article, I'd someone tell me that he wished his co-workers understood how to think like that.
gregory5: http://logs.nosuchlabs.com/log/asciilifeform/2022-01-16#1073458 << good to see that you have made the same observations.
dulapbot: Logged on 2022-01-16 17:07:36 verisimilitude: Read the following.
verisimilitude: I joke with others about how ``If the program accepts X, handle X.'' passes for great wisdom nowadays.
gregory5: http://logs.nosuchlabs.com/log/asciilifeform/2022-01-16#1073462 << this is why languages which built Unicode into basic strings such as Python have caused more harm than good.
dulapbot: Logged on 2022-01-16 17:08:45 verisimilitude: I joke with others about how ``If the program accepts X, handle X.'' passes for great wisdom nowadays.
verisimilitude: Those languages were causing harm without anything like that.
asciilifeform still gotta move various sections around, and clean up whatever barf, but this is 90% of what was aiming for today
asciilifeform not esp. happy re the spec bloat (is over 100kb nao, in html format!) but imho over9000x moar concrete from implementer pov
asciilifeform cleaned up buncha rubbish, will leave along for hr or 2
asciilifeform: e.g. 'getdata' needs ~100% rewrite
asciilifeform: a complete list of KNOBs is needed...
asciilifeform lulz that no one caught 'keys are 512 bytes' typo in past N revisions of spec
signpost been working on the linux distro, fixed all known bugs caused by inanities like relying on file mtime to decide whether to perform magic with e.g. autoconf
signpost: or that v (rightly) doesn't preserve execute bits, things like that.
signpost: also removed all external-world deps from the bootstrap from vintage debian debs.
signpost: getting close to useful. next step is to get my own arse sitting on it.
signpost has a broader vision for the thing that resembles old "UCI" contemplations.
signpost: I want to be able to sit down at a bare computer, boot a kernel, and have the new machine up and processing jobs signed by my key within minutes.
signpost: swarm downloads, recent disussions of DHT-atop-pest, all figure into this of course.
signpost: dragged the thing closer also to binary-reproducible builds, removed a few sources of bit-churn. more yet to find and smash.
signpost: (this is important because with a fixed src-to-bin mapping, anyone can confirm for himself what the proper hash of a build should be. this allows him to trust his friendly peers to provide a cached build of an item, but verify independently)
signpost: it's what a decentralized distro looks like, rather than having trusted parties producing binary packages, and all sharing the same needle.
signpost: regarding "job", I define this as a particular filesystem (or filesystems layered atop one another) within which a command is to be run. that filesystem image can even be a rootfs from another distro, say a dulap.
signpost intends to use other distros in this way going forward, with said tongs.
signpost: (when needed)
signpost: I'll get this thing posted for feedback before long, oughta grunt out a proper spec for folks to read too.
signpost: in other lolz, pretty sure I have bad ram, getting intermittent segfaults when blasting make -j16 using tmpfs for disk
asciilifeform: http://logs.nosuchlabs.com/log/asciilifeform/2022-01-16#1073488 << eccless box ? (if ecc -- is there barfola in log?)
dulapbot: Logged on 2022-01-16 20:07:06 signpost: in other lolz, pretty sure I have bad ram, getting intermittent segfaults when blasting make -j16 using tmpfs for disk
asciilifeform: signpost: whole thing sounds a++ nifty; would be a+++ nifty if could bootstrap it from dulap-gentoo (iirc last time asciilifeform tried, no liftoff, needed concretely a debian of some kind)
asciilifeform: http://logs.nosuchlabs.com/log/asciilifeform/2022-01-16#1073479 << asciilifeform in fact achieved this, so to speak; simply turned out that the jobs would run at speed of 486...
dulapbot: Logged on 2022-01-16 19:53:18 signpost: I want to be able to sit down at a bare computer, boot a kernel, and have the new machine up and processing jobs signed by my key within minutes.
asciilifeform moved sections around in spec; afaik all that remains is to move the buffers sect., write up mechanics of output buffer (and 'getdata' with it, the issuance of getdata's is rightfully part of that sect.)
asciilifeform unlikely to get anymoar done today
← 2022-01-15 | 2022-01-17 →