snsabot: Logged on 2019-09-26 13:19:48 diana_coman: mircea_popescu: once the new voice model comes in, is the L1 redefined accordingly too? iirc there are some priviledges such as "can use feedbot & deedbot only if in L1", right? spyked or trinque might correct me here.
spyked: http://logs.nosuchlabs.com/log/trilema/2019-09-26#1938376 <-- hm. I suppose feedbot could do something along the lines of "talk (via PM) only to people who are already +v in the channels where bot is listening"; either way, there's currently no strict enforcement in place, I'll add something if/when it becomes necessary
mircea_popescu: aite, /me shall bbs on the travel console!
asciilifeform: BingoBoingo: granted you won't be building a secondary dc in the garage, with gsm, but oughta suffice, i suspect, for work
BingoBoingo: The local telecom's preferred solution for rural internet is cellular.
BingoBoingo: One other complication leaving the central municipalities C, B, CH, and E involves incomplete coverage of the residential fiber network.
asciilifeform: BingoBoingo: admitting that this puzzler is out of asciilifeform's paygrade -- imho getting even the smallest of these barracks would be a massive win.
BingoBoingo: Go one or two streets back and just about everything without a for sale or for rent sign lights up at some point during the week.
BingoBoingo: Most of the apartments along the short of the Rambla here in Pocitos are never lit, never show signs of life, suggesting absentee owners
BingoBoingo: asciilifeform: As of 2013 through a bilateral treaty the shelter part of the arrangement went out. From where I'm sitting it is hard to tell which way the Argentine winds will push
asciilifeform: aha but argentards back 'in the saddle' nao
BingoBoingo: The advertised asking prices have been inching down since I arrived. I suspect the former use of property in Uruguay as a shelter from Argentine taxes kept real estate prices artificially high.
BingoBoingo: asciilifeform: The widely reported years ago on numerous blogs conventional wisdom is that everything is for sale and ~40-60 percent negotiated off the asking prices was a common practice. This needs to be investigated if moving forward.
BingoBoingo: asciilifeform: Well, there's also taxi strikes ~every other month. Taxis don't run on holidays except by prior arrangement.
asciilifeform: BingoBoingo: 1-2k $ / m^2 seems astonishingly high for an orcistan. but then again i aint expert on subj
BingoBoingo: asciilifeform: They add up. The just under 10 km ride to Prado was 390 pesos going there, 340 pesos returning after traffic settled, for ~20 USD round trip.
asciilifeform: BingoBoingo: 1 immediate observation -- the cabs were so cheap , transport dun seem to be major problem ( from asciilifeform's admittedly brief expedition )
mircea_popescu: meanwhile it looks like ima have to review the trackback model also, as it turns out the new select style can spam trackbacks ( http://trilema.com/2019/the-freenode-issue/#comment-131559 and on)
diana_coman: mircea_popescu: once the new voice model comes in, is the L1 redefined accordingly too? iirc there are some priviledges such as "can use feedbot & deedbot only if in L1", right? spyked or trinque might correct me here.
snsabot: Logged on 2019-09-20 12:05:15 diana_coman: hanbot: you have broken trilema links on your blog as they still use polimedia.us eg those in http://thewhet.net/2012/11/shall-be-delivered/
hanbot: http://logs.nosuchlabs.com/log/trilema/2019-09-20#1937445 << oughta be fixed nao, ty for the notice!
diana_coman: mircea_popescu: he has http://thimbron.com/ for some reason; /me will catch up with the logs now
lobbes: and now that I know that sync-ism is not wanted with wp-mp logger, I have 0 doubts about this nixing. The wonders of the forum for clarifying courses of action never cease
lobbes: buut, I realise now this was kinda dumb. All the data (both "meta" and otherwise) is already inserted into the mysql db ffs. (i.e. if python crashes, I just... query mysql for last article updated and can get the post ID and last updated date, as well as read the contents field into memory). I will nix the flatfile-ism in short order
ericbot: Logged on 2019-09-26 07:31:58 mircea_popescu: and why the fuck flatfile intermediary anywya, what is this, medieval clockwork ?! what's it needed for ?!
lobbes: http://logs.ericbenevides.com/log/trilema/2019-09-26#1938178 << this was a crackpottery I initially came up with to deal with an imagined problem. The "problem" being that I did not want to keep things like "meta data" and the current day's log lines stored in python memory, lest it crashes and loses that data
asciilifeform: just as , i suspect, there were a coupla sane, honest folx even on sslism committee...
mircea_popescu: problem with smart, educated, english-speaking kids is that they'll believe anything.
asciilifeform: in the end, was 'medicine for corpse'.
asciilifeform: they did, i can see it
mircea_popescu: when you know, people still inexplicably imagined they could help
asciilifeform: asciilifeform's brother won't even read naggum : 'i hope that bastard is roasting in hell for helping with sgml'
mircea_popescu: chet was even on the committee back in like 2000
asciilifeform: mircea_popescu: ever look into the history of that liquishit ? i.e. ibm's 'sgml' ? no mega-surprise that the end product is what it is
mircea_popescu: anyway, back to it : html was written by people who had not even a cursory familiarity with what resource location might even mean. consequently, there is no actual way to locate subsets within resources, even if continuous. the only available primitives are, either spurious linefeed, or else partial selects.
snsabot: Logged on 2019-09-26 05:31:35 diana_coman: (and them mircea_popescu will write a mandatory-reading thread of more than 500 lines and my log-selection via log-raw will fail).
mircea_popescu: except of course all the old farts are completely moroning hellspawn a la jwz, rms, ers, etcetera
mircea_popescu: the sadness about this sorta job is that it's eminently not noobworthy.
mircea_popescu: they chiefly come in large drums of "everyone coming before was a fuckhead"
mircea_popescu: what can i tell you my bruther, we got problems.
asciilifeform: mircea_popescu: recall from what they were pre-written. ocean of regexp.
mircea_popescu: and if it doesn't, i daresay the problem's not your slaves.
asciilifeform: hell, it took asciilifeform 7+ weeks to only make the one for 'peh' !
mircea_popescu: what you ~could~ do for your usecase, usefully, would be to add a, say, ?a=alf switch, such that when a=alf, every paragraph gets a chapter anchor and every sentence therein gets a sentence anchor, and then alfblog.org/?a=alf#137:37 takes one directly to paragraph 137 verse 37
asciilifeform: well that's be quite useless. what i want is ~arbitrary~ text, as the occasion demands, whether 1 word in 1 para somewhere, or 6 para with start inside 1st and end in middle of last etc
asciilifeform: if it eats same url commands as other selectors , but able to light up arbitrary text, why would break expectations ?
mircea_popescu: moreover, if that's all you want, you could just add per-para anchors with some tiny tweaking of the mp-wp
mircea_popescu: the problem here is that select behaviour is social, and implementing a weird select for your blog is likely to cost more in that than in provides in whatever it provides
mircea_popescu: asciilifeform, if you select whole paragraphs and them only, you can get away with div at very minor cost (ie, there may be slight diff between "normal" para and selected on some browsers)
mircea_popescu: the thought manifestly never fucking occured to the muppets, "chief among the jobs of any uniform resource locator is to expose a manner to reference arbitrary subsets of the resource once located ; at the very least ONE CONTINUOUS SUBSET)
asciilifeform: mircea_popescu: as i understand, mircea_popescu's current selector worx a++ to select text in 1 para where no line breaks, and does not put in any spurious. but if you want to select multiple para, wai not put the table turd where there are already line breaks ? oughta be visually indistinguishable , neh ?
mircea_popescu: yes. "a one cell table" is just a clunkier div. and the problem with div (instead of span) is that it... breaks the line where it starts.
asciilifeform: a, the line feed.
mircea_popescu: asciilifeform, no, there's no way to do it with any number of tags. the problem is exactly as described : spans that can span any subset of the characters insert a spurious line feed ; spans that do not insert spurious line feed can not span any subset of the characters but only some.
snsabot: Logged on 2019-09-26 11:38:35 asciilifeform: when humans write responses to articles, typically the section being addressed goes ~above~ the reply. why not log renderer follow same form ?
mircea_popescu: http://logs.nosuchlabs.com/log/trilema/2019-09-26#1938308 << this isn't nearly as universal as all that. going by my own practice, sometimes i paraphrase, sometimes i footnote... but the whole discussion's more a case of "when bird flies, wings move, why not airplane". cuz not everything's a bird, what.
asciilifeform: mircea_popescu: i fully buy that there prolly aint a way to do it cleanly with ~1~ added tag.
asciilifeform: will have to try, i'd really like to be rid of the js, ~without breaking existing text~
mircea_popescu: but the problem's of primitives not wrappers ; nobody on the fucking w3c commission ever had any actual usecases in mind. just shit.
mircea_popescu: asciilifeform, font doesn't do what you think (not addressable) ; table is clunkier div. can do either of course, up to you.
asciilifeform: when humans write responses to articles, typically the section being addressed goes ~above~ the reply. why not log renderer follow same form ?
snsabot: Logged on 2019-09-26 08:29:31 mircea_popescu: http://logs.nosuchlabs.com/log/trilema/2019-09-25#1938084 << imo this sounds like a good idea in theory, but is not a good idea in practice. the reason for the difference is that the links as we use them now are ~fixed length~ AND ~fixed format~, something the quotes can never be. they are BETTEr like this than like that because the fix-* is the more important attirbute in thsi context.
asciilifeform: http://logs.nosuchlabs.com/log/trilema/2019-09-26#1938241 << suggestion here wasn't that speakers literally paste the quotes , as in heathen days, but that the ~current~ 'paste log link and see bot echo' get transformed ~by www viewer~ into the pictured form.
snsabot: Logged on 2019-09-26 08:12:38 mircea_popescu: http://logs.nosuchlabs.com/log/trilema/2019-09-25#1938065 << sadly this is a html breakage. the options are delineated in article : either use span and not cross paras, or else use div and end up with an extra newline.
asciilifeform: http://logs.nosuchlabs.com/log/trilema/2019-09-26#1938229 << i dun see why the server end could not simply add <font...> tags around all of the material being selected?
asciilifeform: mircea_popescu: anyone who wants can run the current logger and declare 'i aint syncing, mine writes to antifuse rom' etc. very easy to turn syncing into nonsyncing logger, just dun sync. but i think i get how and for what mircea_popescu's is supposed to work
snsabot: Logged on 2019-09-26 04:44:35 mircea_popescu: ie, if "selection doesn't work for me" "why not ?" "because what i want to select recurs", give some thought whether indeed "selection is broken and should be fixed". WHY do you want to select something THAT RECURS, and recurs so much it actually doesn't allow you to extend the context slightly, one character at a time, resolving your problem ?
mircea_popescu: i am pretty sure nobody here ever thought before to ~edit~ their irclogs for ~any reason~.
mircea_popescu: the situation where there's both unchanging loggers and syncing loggers is in fact more stable wrt that sort of challenge.
mircea_popescu: consider the complexities involved also. what if there's a serious dispute re log ? what do we all do, we each fall back on whatever private irclogs we kept, eys ? and conceivably even have a party split, x-y-z see Q, k-l-n see P.
mircea_popescu: that it is. and i would be concerned if there was no synced logger, too.
mircea_popescu: or to put it another way : a logger confronts the dilemma if whether to sync with history or with other loggers. i don't believe there's need for a rule, it can stand as such.
snsabot: Logged on 2019-09-26 08:42:44 mircea_popescu: http://logs.nosuchlabs.com/log/trilema/2019-09-26#1938136 << this is quite the idea. i have no expectation trilema log and whoever else's such log be machine-identical.
asciilifeform: http://logs.nosuchlabs.com/log/trilema/2019-09-26#1938261 << this appears to be the key : if mircea_popescu's item aint intended to be a syncable logger, then naturally it dun need to giveashit re dates problem etc.
mircea_popescu: that'd be the whole story.
mircea_popescu: the owner doesn't have to specify WHO is his lordship, because the bot just looks at who he rated 9.
mircea_popescu: i kept dithering back and forth in here because i stupidly never sat down to draw this all out properly and review history etc. but now i have, and that's the story.
mircea_popescu: in order for this system to retain its utility throughout the board, i believe the whole array should be offered to castle owners. thus for deedbot to manage voicing in a castle the owner would have to specify the castle name, what x value he wants, whether lobe 2 is on and if so what threshold y to be used.
mircea_popescu: 1 must be on for there to be a voice model in the first place. #trilema was traditionally 1 on, 2 on, x = 0, y = 0. with this castle update #trilema is moving to 1 on, 2 off, x = 1, y = undefined.
mircea_popescu: there's two lobes to qualify for voice in a castle : EITHER 1.(rating above x by the castle owner) OR 2.(sum rating over y by his lordship as he defines it.)
mircea_popescu: trinque, here's the complete story :
snsabot: Logged on 2019-09-26 00:54:44 asciilifeform: it's possible that mircea_popescu intends his item to be standalone, rather than pilot plant for all 'canonical' loggers. then dun matter so much whether calendar etc.
mircea_popescu: http://logs.nosuchlabs.com/log/trilema/2019-09-26#1938169 << quite. i intend it to be pilot plant for standalone, i'm specifically curious how bad "the bad" is gonna turn out to be.
snsabot: Logged on 2019-09-26 00:42:50 asciilifeform: asciilifeform, for instance, likes to think that his hands grow from shoulders, and not from arse, and still evidently produced these.
mircea_popescu: http://logs.nosuchlabs.com/log/trilema/2019-09-26#1938159 << you have a certain sort of impulsive dr hyde lying inside, prone to all sorta these out of a specific set.
mircea_popescu: i dunno that i'd want this ; but the experimentally inclined may find it interesting to explore
snsabot: Logged on 2019-09-26 00:30:46 asciilifeform: can be anyffin, even line count. so long as it can be mechanically derived from the actual lines, rather than clocks.
mircea_popescu: http://logs.nosuchlabs.com/log/trilema/2019-09-26#1938151 << there is such a thing implemented, specifically in the auctionbot spec for #eulora (say your line on the hour but only if there's been chatter since last saying). something very similar can also work here, "a new day starts when there's been silence for x time"). while clocks drift, they don't also diverge, not on the sort of scales here contemplated.
mircea_popescu: if i make the log say X on my blog, that's my fucking priviledge, it exists on my authority.
snsabot: Logged on 2019-09-26 00:14:20 asciilifeform: you gotta break the pages ~somehow~. q is how. if you do it by 'day', you end up stepping on same mine.
snsabot: Logged on 2019-09-26 00:13:24 lobbes: re: two or more mp-wp loggers agreeing with each other; I'm not sure they would ever need to agree with each other in terms of links... IF not indexing. In other words, each instance of a mp-wp logger and its output becomes indistinguishable from any other blog post on said blog. (in other words, if you don't care to sync, or don't care if "missing" lines, then not an issue)
mircea_popescu: http://logs.nosuchlabs.com/log/trilema/2019-09-26#1938136 << this is quite the idea. i have no expectation trilema log and whoever else's such log be machine-identical.
snsabot: Logged on 2019-09-26 00:12:38 asciilifeform: but if you calendarize the pregenned-pages thing, you end up baking the 'when does day start' idiocy right into the db. where cannot even be easily later fixed...
mircea_popescu: http://logs.nosuchlabs.com/log/trilema/2019-09-26#1938134 << can be "easily fixed" in theory because it was dumped in mechanically ; but won't be easily fixed in practice because fu, i'm not changing trilema articles once published.
snsabot: Logged on 2019-09-25 23:49:18 lobbes: in the mircea_popescuine logger, though, all links will use the server-side-selectory-style. By definition those links will never agree with the traditional loggers since the latter can't span arbitrary lines / text
mircea_popescu: http://logs.nosuchlabs.com/log/trilema/2019-09-25#1938119 << even there, you'll have to name the days ~something~, which reduces to naming them the date, meaning that your select will only work if the line is IN tjat date.
snsabot: Logged on 2019-09-25 23:29:24 asciilifeform: nor is there likely ever gonna be any such agreement. use of reich time gotta be made 100% decorative, or to go away entirely.
mircea_popescu: http://logs.nosuchlabs.com/log/trilema/2019-09-25#1938099 << hey, trilema has a romanian chunk. ~doesn't bother anything~.
mircea_popescu: http://logs.nosuchlabs.com/log/trilema/2019-09-25#1938096 << altogether not clear to me any actual retardation's in fact been identified, which may have something to do with it.
snsabot: Logged on 2019-09-25 23:01:58 asciilifeform: if finally throwing out the ancient kakobreklic format for logger display, wai not go full throttle.
mircea_popescu: s 60% the length ; but this is not the be-all end all of intellectual accessibility.
mircea_popescu: a car-plane is not more compact than a set of car, plane ; nor is it "more intellectually accessible". it's true the manual
mircea_popescu: http://logs.nosuchlabs.com/log/trilema/2019-09-25#1938084 << imo this sounds like a good idea in theory, but is not a good idea in practice. the reason for the difference is that the links as we use them now are ~fixed length~ AND ~fixed format~, something the quotes can never be. they are BETTEr like this than like that because the fix-* is the more important attirbute in thsi context.
snsabot: Logged on 2019-09-25 22:43:04 asciilifeform: incidentally, if yer going full throttle w/ selectables , the Right Thing imho would be to take it all the way and make items like this actually display (if reasonably compact.. config knob?) the linked text. a la old man ted nelson's 'transclusions' concept .
mircea_popescu: http://logs.nosuchlabs.com/log/trilema/2019-09-25#1938080 << there's no good way to "knob" this.
snsabot: Logged on 2019-09-25 22:38:25 asciilifeform: the other thing, the adhoc traditional line select is human-friendly, very fast to grab a line url, whereas mircea_popescu-style selector less so, gotta find unique start/end text, when citing from trilema it usually takes asciilifeform 2-3 shots to nail down the correct snip
snsabot: Logged on 2019-09-25 22:29:07 asciilifeform: http://logs.nosuchlabs.com/log/trilema/2019-09-25#1938050 << if you dun actually store the raw irc lines somewhere, sync ~from~ your logger becomes suddenly quite nontrivial.
mircea_popescu: http://logs.nosuchlabs.com/log/trilema/2019-09-25#1938067 << so the idea is that the flatfile is part of the orig logger ? i misread, i thought it's novel addition.
mircea_popescu: until html replacement, this is the limit
mircea_popescu: http://logs.nosuchlabs.com/log/trilema/2019-09-25#1938065 << sadly this is a html breakage. the options are delineated in article : either use span and not cross paras, or else use div and end up with an extra newline.
mircea_popescu: alternative schemes may be devised, but it's not a simple "let's just mechasmush everythin together" sorta level of consideration.
mircea_popescu: which is why generally when rendering log i tend to preserve nametags per line and the original lines (though not always) and instead insert extra line breaks now and again to paragraph it up.
mircea_popescu: unlike "poetry" ie the socialist shit, where "things look a certain way" with 'verses" (and i'm stuck ha;lf the time dropping it altogether, because it's better absent than fucked), our prosody is actually interesting.
snsabot: Logged on 2019-09-25 22:23:37 asciilifeform: ( and: if yer not indexing by line , why not actually merge the text when speaker is contiguous ? )
mircea_popescu: http://logs.nosuchlabs.com/log/trilema/2019-09-25#1938061 << this is a long standing problem (not kidding, i been pondering "how to display logs" for a LONG time). generally it seems to me preferable to preserve the original prosody of the author (because yes, this is what this is, "how to break into lines", duh, ancient greek issue)
spyked: diana_coman, from what I can see, raw-log display also fixes the "when to cross page boundary" problem, heh. e.g. http://logs.ossasepia.com/log-raw/trilema?istart=1938114&iend=1938120 displays fine
mircea_popescu: will need moar beard for crossing the 500 line threshold.
diana_coman: (and them mircea_popescu will write a mandatory-reading thread of more than 500 lines and my log-selection via log-raw will fail).
mircea_popescu: no, tis outta respect. i will take the time to type something out, an old poem, whatever, for similar reasons : some things are worth it. and they get worth it through occasionally, ever so occasionally, doing you that rare and radiant wonder of a favour, where "it turns out" aka you finally deign to fucking notice you were doing something stupid.
mircea_popescu: in fact, the reason i take the time to explain how various shits work, unpopular as they may be, is because i put the time in to actually understand how they do in fact work ; and i did that not because i was bored, or because i was desperate to find something to pour some of this time into, i have like a well of time at home and it's overflowing so you gotta put some in all availavble receptacles or else it floods the hou
mircea_popescu: ~the dichotomy is creative~. we don't have many (in the sense of : not any) alternative leverage points available. gotta use what's there.
mircea_popescu: yet strangely mp doesn't mind nearly as much as the bottom centile of the "red pill" crowd. how the fuck ?! is it because he's a jew that he doesn't suspect jews are in charge ? what the fuck's going on here ?!
mircea_popescu: yet mp is also the ~least~ angry at "the system". how the fuck could this be ? is mp insane ? if everyone's so fucking pissed off over epsilon, "they gave us 9000 things but we're missing like... three. burn it all down", what the fuck's mp supposed to say, he has to have his fucking food flown in / slave made, and clothes flown in across the world and everything else.
mircea_popescu: erryone's happy with a girl to worship at home, just put her into this one shrineroom and worship there now and again ; mp apparently needs multiple (what! how!) and ~somehow even manages this~, for years, decades, it's not a wish-perceived fantasy, he lives like this ?!
mircea_popescu: how it all relates to the system.
mircea_popescu: and it can not even be explained JUST HOW!!! creative that thing that's creative actually is. let's look at one example : mp is, as a factual matter, the one who needs most things not put in. mp apparently needs to walk his bitches on a leash downtown, which romania doesn't even have words for, every newspaper in the country must drop whatever it was doing, selling cheap chinesiums, to talk about mp's weird needs now, and
mircea_popescu: it could be, of fucking course, there's no denying that. but there's another step in there, and not to be skipped over.
mircea_popescu: in fact, all it really takes is this conviction that a problem once perceived's thereby an' therefore also a problem that needs solving. NOT SO.
mircea_popescu: and participation ~doesn't take all that much~. that's why it exists, after all, because it's easy, because it is in fact even easier than the very little use them powerful old men behaving badly might put a dumb wench towards.
mircea_popescu: the subjective paralysis is actually the problem. participating in the tavern wench chorus as to how bad your owner is prevents you from fucking thinking, or growing the fuck up (the two being related).
mircea_popescu: this isn't actually the problem though ; as well documented on trilema, the effect on powerful old men an' their behaviour of the tavern wench consensus is still the nil it's ever been.
mircea_popescu: if one just goes by the persuasion tradewinds, this lever's permanently stuck to position one ; and ridiculous nonsense flows downstream, superficially visible as "old men in power have been behaving badly" or w/e currently fashionable cri de guerre of the retard camp.
mircea_popescu: the dichotomy between the subjective life of the subject and the demands of the outside structure ~is creative~. this is specifically the mechanism through which it is creative : "i wonder why it is i want something that these idiots didn't put in". one possible outcome of a correct such evaluation is, indeed, "jesus fuck they're idiots". however, ANOTHER possible outcome is the ~EXTREMELY~ valuble bootstrap out of d-k rec
mircea_popescu: the persuasive universe is actually very much like a fungal infection of the mind : it grows nicely, i'm sure, but in so doing it misuses a fundamental other thing, that actually has a function independent of ever-growing fungal blather.
mircea_popescu: w far that gets you and bother whoever cares about it.
mircea_popescu: there's exactly no need to "stop rape", for instance. "being special" (defined as, "never being raped") is not some kind of "universal right of womanhood". no woman's born entitled to be special, she's born to be used, like any other blade of grass on this here GREEN earth. that specialdom may arise, as a mist, RETROSPECTIVELY, is one thing. but the problem needs no "solving" -- you don't wanna take the cock, don't. see ho
mircea_popescu: nevertheless -- not all problems one can persuade himself into perceiving are actually worth solving.
mircea_popescu: the only available support for such bootstrap is this kind of indicia, "why are my self-perceived needs conflict with my betters' designs". it could, surely, be the case they're fucked in the head. i'm sure they often are.
mircea_popescu: god put ~exactly no clues~ permitting one to bootstrap out of the ~necessary~ d-k recursion. god put no such clues in ~deliberately~, and god also made it necessary in the first place. because god's an asshole.
mircea_popescu: ie, if "selection doesn't work for me" "why not ?" "because what i want to select recurs", give some thought whether indeed "selection is broken and should be fixed". WHY do you want to select something THAT RECURS, and recurs so much it actually doesn't allow you to extend the context slightly, one character at a time, resolving your problem ?
mircea_popescu: http://logs.nosuchlabs.com/log/trilema/2019-09-25#1938058 << this is an insane problem to solve. if all you're trying to delineate is "mircea popescu", then how about linking to the whole page ; otherwise, it was ~something the man said~, which gives rise to both the need to point to it specifically, and to the method of doing just that.
mircea_popescu: this has easily been the weirdest wtf... uncleaned coffee mugs invaded by ergot or what the hell's going on here
snsabot: Logged on 2019-09-25 20:37:34 asciilifeform: imho a logger that can't resync at all, aint much of a logger. but at the same time i dun see why a generate-statics logger couldn't be made to resync. simply a bit moar complicated.
mircea_popescu: http://logs.nosuchlabs.com/log/trilema/2019-09-25#1938046 << are you fucking kidding me, how the fuck are old logs not read-only even anyway ?!
mircea_popescu: and why the fuck flatfile intermediary anywya, what is this, medieval clockwork ?! what's it needed for ?!
snsabot: Logged on 2019-09-25 20:21:24 lobbes: indeed. Would just be a buncha "update" statements that fill in the proper lines. As it currently is designed, I have it spitting lines into a flat file first, and that file is used to update the "current day's log" post as new lines are seen
snsabot: Logged on 2019-09-25 20:17:27 asciilifeform: lobbes: as i understand, with that type of logger you'd have to regen all of the pages that include or follow the gap, in order to close a filled gap.
mircea_popescu: http://logs.nosuchlabs.com/log/trilema/2019-09-25#1938040 << this made entirely 0 the fuck sense, what the fuck, if you miss a line thereby the int'l dateline changes ?!
asciilifeform: ( if he did -- presumably manually fixed, and didn't bother to make a note of it )
asciilifeform: it's possible that mircea_popescu intends his item to be standalone, rather than pilot plant for all 'canonical' loggers. then dun matter so much whether calendar etc.
lobbes: I must digest these threads some more, methinks (though I thank asciilifeform for the edification). I'll bbl; sleep
ericbot: Logged on 2019-09-26 03:46:04 asciilifeform: imho a pregen-page logger must split on something ~other~ than calendarism.
lobbes: http://logs.ericbenevides.com/log/trilema/2019-09-26#1938150 << nevertheless, things like this are harder to implement once the concept of dates are thrown out the window
asciilifeform: ( they are regularly found ~not~ to )
asciilifeform: this is not even to touch on the q of whether a particular shitlang's regextron even worx as specified.
asciilifeform: asciilifeform, for instance, likes to think that his hands grow from shoulders, and not from arse, and still evidently produced these.
snsabot: Logged on 2017-02-09 12:53:33 asciilifeform: incidentally -- there is more to this crapolade than meets the naked eye. regexp is a fundamentally braindamaged concept -- not only in the way described by naggum (how do you infer false positives?) but also in the haskellian idiocy where 'reasonable' and 'thermonuclear self-annihilation' programs look nearly identical, and distinguishing them is np-hard
snsabot: Logged on 2019-09-26 00:07:47 asciilifeform: the browser is unable to send to '#1938120' to the box.
asciilifeform: can be anyffin, even line count. so long as it can be mechanically derived from the actual lines, rather than clocks.
asciilifeform: imho a pregen-page logger must split on something ~other~ than calendarism.
asciilifeform: if you introduce hard splits by date -- the example differential pair nao will not only not load identically, but also not echo. and worse, there will be no decidable algorithm for even determining what the link made for 1 logger, corresponds to on other.
asciilifeform: the current logger only is able to get away with disregarding dates when echoes, because the pages are dynamically genned
asciilifeform: the sooner we find a pill for this, the fewer speshulcase atrocities will have to be committed to make old log readable in new formats.
asciilifeform: the date -- gotta go
asciilifeform: even if they're not indexed by line, but take the form ...../chan/date?start=foo&end=bar
asciilifeform: if the links contain dates, and the date is in any way significant to the machine -- which, if you cut by 'day' the pages, it will be -- yer stuck with this horror.
asciilifeform: lobbes: understand also that the problem aint limited to '2 loggers agree'. clocks -- drift. arbitrarily. needs also to agree re links with ~past self~
asciilifeform: you gotta break the pages ~somehow~. q is how. if you do it by 'day', you end up stepping on same mine.
ericbot: Logged on 2019-09-26 00:54:16 mircea_popescu: whereas on trilema you have the whole power of the mp-wp selectylanguage at your disposals, you don't need the line numbers because you can just select w/e actually interests you
lobbes: Now, to revisit upstack, if you are intending to *sync* one mp-wp logger with another mp-wp logger, then I guess you'd run into the same issue as the trad. loggers. I.e. In order to sync you *would* need to index.
lobbes: re: two or more mp-wp loggers agreeing with each other; I'm not sure they would ever need to agree with each other in terms of links... IF not indexing. In other words, each instance of a mp-wp logger and its output becomes indistinguishable from any other blog post on said blog. (in other words, if you don't care to sync, or don't care if "missing" lines, then not an issue)
asciilifeform: but if you calendarize the pregenned-pages thing, you end up baking the 'when does day start' idiocy right into the db. where cannot even be easily later fixed...
ossabot: Logged on 2019-09-26 00:05:26 lobbes: but I see the issue re: the traditional loggers agreeing with each other
snsabot: Logged on 2019-09-25 23:49:40 lobbes: but I see the issue re: the traditional loggers agreeing with each other
asciilifeform: try to click on the snsabot link.
asciilifeform: the bots know what to do ( i wrote so it loads by index strictly ~when parsing for echo~ cuz ~possible~ there ) but yer browser does not !
ossabot: Logged on 2019-09-26 00:05:26 lobbes: but I see the issue re: the traditional loggers agreeing with each other
snsabot: Logged on 2019-09-25 23:49:40 lobbes: but I see the issue re: the traditional loggers agreeing with each other
asciilifeform: for so long as concept of 'day' is used at all, 'the hole through which the night goes in'(tm) remains open
lobbes: in the mircea_popescuine logger, though, all links will use the server-side-selectory-style. By definition those links will never agree with the traditional loggers since the latter can't span arbitrary lines / text
snsabot: Logged on 2019-09-25 23:27:50 asciilifeform: ericbot: i'ma restate compactly. atm we index lines. by a monotonic # . the timestamps are stored as unix epochal times. the urls, however, demand a e.g. 2014-07-16 , and always did, and these both now and then were parsed via what the particular box thought local time was. resulting in headache, because machine timekeeping was , and remains, shite.
asciilifeform: with ntpism they give up not only sovereignty (washington gets to tell you 'what time it is') but monotonicity !
asciilifeform: the heathens pretend that they 'solve' this via ntpism. but, characteristically of heathenisms, it dun fucking solve anyffin, is a 'can't believe it aint butter!' pressed sawdust 'solution'.
asciilifeform: the historic epochtimestamps are incidentally all over the place, try setting a scratch box logger to sort ~by time~ some time and weep. ( from when phf , for instance, imported his znc, evidently wasn't 100% synced to the primary logger ; and elsewhere )
asciilifeform: so , say you have a logger of whatever type (either traditional or mircea_popescuine) and 2 of these dun agree re when a day ended -- bang, you get links that dun lead to the desired text. cuz they try to use the fucking date.
asciilifeform: even to actually use the unix epochal time, is dodgy. cuz, again, no 2 tmsr boxes are ever likely to come to an accurate agreement re what unix.epochtime it is 'now'.
asciilifeform: nor is there likely ever gonna be any such agreement. use of reich time gotta be made 100% decorative, or to go away entirely.
asciilifeform: this problem, incidentally, if not cured, will persist in a hypothetical mircea_popescu-style pregenned-pages logger.
asciilifeform: ericbot: i'ma restate compactly. atm we index lines. by a monotonic # . the timestamps are stored as unix epochal times. the urls, however, demand a e.g. 2014-07-16 , and always did, and these both now and then were parsed via what the particular box thought local time was. resulting in headache, because machine timekeeping was , and remains, shite.
snsabot: Logged on 2019-08-27 11:35:18 asciilifeform: http://logs.nosuchlabs.com/log/trilema/2019-08-27#1931353 << phf yer missing the point , i need so that http://btcbase.org/log/2014-07-16#758070 AND http://btcbase.org/log/1945-07-16#758070 go to same thing !
asciilifeform: (couldn't agree on translation of n-1th ea. time they got hard-reset)
asciilifeform: ancient problem, sorta how the j00z ended up with a talmud in 4 langs
asciilifeform: ftr we still haven't a pill for the url clocks problem.
lobbes: aha, I was just about to mention the bot echos
asciilifeform: this, rather like the 'url clocks' problem, demands a 'smart' transmutation of old l0gz, somehow, in order to work 100% correctly.
asciilifeform: would have to make the wwwtron avoid displaying bot echoes tho ( theoretically one'd still want bot echoes , to see wtf is going on when reading via live irc session rather than www )
asciilifeform: if finally throwing out the ancient kakobreklic format for logger display, wai not go full throttle.
asciilifeform: for folx who give a shit 'via what bot? was the citation' -- can put that in 'hovertext' .
snsabot: Logged on 2019-09-25 21:41:56 mircea_popescu: consider both the positive (no convergence language'd ever have emerged out of "oh, phf did excellent job with logger") and the negative (without "we have these loggers, how do we sync them" nobody'd have ever understood why mp-wp is a better display mechanism than "this shit i just brewed").
asciilifeform: result is imho not only moar compact ( why throw the ' http://logs.nosuchlabs.com/log/trilema/2019-09-25#1938055 ' to the screen ? ) but moar readable .
asciilifeform: ^ in this example, as we're using line-based selector, cites whole line. but if used fine-grained selector, could cite the selection.
asciilifeform: another thought : the display of literal machine-readable links verbatim in log, is quite wasteful, and if yer doing a whole-page transform, quite avoidable. why not make the cited text the displayed part of the clickable link (and orig. url -- the machine part when same is clicked) .
asciilifeform: incidentally, if yer going full throttle w/ selectables , the Right Thing imho would be to take it all the way and make items like this actually display (if reasonably compact.. config knob?) the linked text. a la old man ted nelson's 'transclusions' concept .
snsabot: Logged on 2019-09-25 21:37:54 mircea_popescu: loggers as extant suffer from being built manalone style, from scratch, as such the only possible in-page selection is an ad-hoc (but very functional, as it happens) by-line.
asciilifeform: the other thing, the adhoc traditional line select is human-friendly, very fast to grab a line url, whereas mircea_popescu-style selector less so, gotta find unique start/end text, when citing from trilema it usually takes asciilifeform 2-3 shots to nail down the correct snip
snsabot: Logged on 2019-09-25 22:23:37 asciilifeform: ( and: if yer not indexing by line , why not actually merge the text when speaker is contiguous ? )
snsabot: Logged on 2019-09-25 05:08:58 mircea_popescu: billymg, did you ever find the trilema article re how to re-do your missed pingbacks ?
snsabot: Logged on 2019-09-25 21:41:56 mircea_popescu: consider both the positive (no convergence language'd ever have emerged out of "oh, phf did excellent job with logger") and the negative (without "we have these loggers, how do we sync them" nobody'd have ever understood why mp-wp is a better display mechanism than "this shit i just brewed").
asciilifeform: http://logs.nosuchlabs.com/log/trilema/2019-09-25#1938055 << iirc he described , 'cheated' by running a znc somewhere and so happened that at no point both fell down simultaneously ( just as e.g. snsabot and ossabot not yet fell down together )
asciilifeform: can't think of any reason not to ~display~ them as mircea_popescu described tho
snsabot: Logged on 2019-09-25 21:37:00 mircea_popescu: http://logs.nosuchlabs.com/log/trilema/2019-09-25#1938036 << nor do you give a shit, seeing how the lines aren't indexed.
asciilifeform: http://logs.nosuchlabs.com/log/trilema/2019-09-25#1938050 << if you dun actually store the raw irc lines somewhere, sync ~from~ your logger becomes suddenly quite nontrivial.
asciilifeform: ( and: if yer not indexing by line , why not actually merge the text when speaker is contiguous ? )
snsabot: Logged on 2019-09-25 21:38:22 mircea_popescu: whereas on trilema you have the whole power of the mp-wp selectylanguage at your disposals, you don't need the line numbers because you can just select w/e actually interests you
mircea_popescu: the problem with people isn't a problem with people, but with infinities : realia is somewhat infinite ; idealia is MUCH more infinite than that (whole thing neatly mirrors the cardinality of power sets, actually).
mircea_popescu: consider both the positive (no convergence language'd ever have emerged out of "oh, phf did excellent job with logger") and the negative (without "we have these loggers, how do we sync them" nobody'd have ever understood why mp-wp is a better display mechanism than "this shit i just brewed").
mircea_popescu: this whole pile is however a massive lesson in evolution-vs-design thematics & tropes ; that "the function creates the organ" is musky throughout ; the blogger's retrospect ("had i not done x it'd never have occured to me to do y") omnipresent ; and so following.
mircea_popescu: whereas on trilema you have the whole power of the mp-wp selectylanguage at your disposals, you don't need the line numbers because you can just select w/e actually interests you
mircea_popescu: loggers as extant suffer from being built manalone style, from scratch, as such the only possible in-page selection is an ad-hoc (but very functional, as it happens) by-line.
snsabot: Logged on 2019-09-25 20:12:27 lobbes: hmm this makes me rethink my current design re: mp-wp bot as well. As it stands, there is no way to "re-sync" since it just spits the lines into a blog post.
mircea_popescu: http://logs.nosuchlabs.com/log/trilema/2019-09-25#1938036 << nor do you give a shit, seeing how the lines aren't indexed.
snsabot: Logged on 2019-09-25 20:02:13 asciilifeform: was thinking re how oughta do auto-syncs. one possible method, is for bot to take command via pm, e.g. !q sync 1000 http://logs.ossasepia.com ; would then walk last N (here, 1000) ln, and offer 'identical', 'diverges prior', or 'diverges at I', I is index, and offer alignment, operator (set in config who) can then confirm or reject
mircea_popescu: http://logs.nosuchlabs.com/log/trilema/2019-09-25#1938022 << emerging sync language from the fact we use bots, rather than not. mindblowing.
asciilifeform: it'd be a 1st class bitch to resync ~from~, however. but fortunately we have other types of loggers that know how to emit lines x..y for given x/y
asciilifeform: imho a logger that can't resync at all, aint much of a logger. but at the same time i dun see why a generate-statics logger couldn't be made to resync. simply a bit moar complicated.
lobbes: in theory, could index the lines in that file, so as to preserve ordering, as well as keep more than one file (currently, this one file is overwritten on each new day)
lobbes: indeed. Would just be a buncha "update" statements that fill in the proper lines. As it currently is designed, I have it spitting lines into a flat file first, and that file is used to update the "current day's log" post as new lines are seen
asciilifeform: ( and to do it very quickly, else someone could easily speak during the regen, and leave you unsynced again )
asciilifeform: lobbes: as i understand, with that type of logger you'd have to regen all of the pages that include or follow the gap, in order to close a filled gap.
lobbes: yeah, I'll guess I'll let the man speak to if he wants sync capabilities in the mp-wp-tronic branch
asciilifeform: lobbes: it's 1 of the reasons why asciilifeform did not like mircea_popescu's original tip re how to bake logger ('just pipe it into a wp') ; but mircea_popescu did specifically ask for 1 that does exactly that, for own www, i presume he knows what he's doing
asciilifeform: hence why imho resync ~must~ be a semi-automatic, rather than automatic, process.
lobbes: hmm this makes me rethink my current design re: mp-wp bot as well. As it stands, there is no way to "re-sync" since it just spits the lines into a blog post.
asciilifeform: mechanical diffing, meanwhile, is tricky because ordering is already known to differ ( almost guaranteed to differ when bots are speaking, one's own bot's output ~always~ enters log before that of other bots ; but also can differ elsewhere on acct of fleanode weather )
asciilifeform: so far all the gaps have been small enuff to visually examine, but this will not necessarily remain the case
asciilifeform: thing is -- imho it is poor practice to blindly pick another logger and overwrite massive chunk of own db with it w/out mechanically testing that they diverge strictly in favour of the other
asciilifeform: lobbes: 'tail' here being whatever lines come after the gap, and thereby have erroneous index #s
asciilifeform: lobbes: not if it knows how to drop the broken tail
lobbes: in general auto-sync would be nifty. but just like the manual method, would need to wait until "dead time" to prevent new lines appearing during the sync wouldn't it?
asciilifeform: another method would be that periodically tests against loggers given in config, and pm's operator if and only if finds divergence.