log☇︎
2900+ entries in 0.001s
ossabot: (eulora) 2017-02-05 hanbot: fwiw diana_coman i have a blender-made animated character guy finished and am gonna try out the crystal space exporter thingy as per http://www.crystalspace3d.org/docs/online/manual/Blender.html , prolly tomorrow. if you have any tips/pointers/etc in the meantime pls to schpiel at me
hanbot: yeah there's some #eulora refs like http://logs.ossasepia.com/log/eulora/2017-02-05#940257, i sincerely dun recall properly documenting tho', probably part and parcel of the overwhelmed at step 10 of 10 thing. mea culpa, i oughta know better.
diana_coman: re maintaining the infrastructure, I don't see how it can be really justified as such anyway given that uhm, those doing something with it are anyway essentially non-existent currently so can't maintain so they "do" anything; and if it is to *also* build them up, then well, they'll fit whatever infrastructure is provided, not the other way around.
mircea_popescu: i recal lpictures of the slime guy in it!
diana_coman: mircea_popescu: I had the article on the toolchain to help her as I knew she was having a go at it based on some discussion in #eulora I think; maybe that?
mircea_popescu: hanbot, why the fuck didn't you write it out ? i remember reading SOMETHING.
mircea_popescu: diana_coman, the problem with the foregoing is that i can kinda ballpark the COST of maintaining all the infrastructure for monkeys to typewrite at their leisure. so can you, i expect.
hanbot: nah, i committed a major sin there, possibly i'll have to pay by starting over and properly documenting.
mircea_popescu: where was the article btw, got a link ?
mircea_popescu: so you're saying the unusable tools part was not the making but the exporting, as it were ?
hanbot: i had a lot of fun modeling a slime guy in blender, then discovered getting it to a usable form via "baking textures" etc 5 or so layers deep was ... i guess i'd call it beyond my attention span.
mircea_popescu: hey hanbot how did that go ?
mircea_popescu: except, of course, they don't. and when hanbot TRIED to do it, she died in a flaming mess of deeply inadequate tools. she almost made a slime. ALMOST.
diana_coman saw this coming
mircea_popescu: just so these idiots can alter a pixel here or there.
mircea_popescu: so it's worth making a thing like blender, and then making a thing like the inexistent exporter, and then making a thing like cs, and all the xml wrapping and etcetera,
mircea_popescu: so hard and difficult and unapproachable and scary and etcetera is this question of splittign the space, that it is worthwhile to go to all the trouble of farming a bunch of morons, because their crap/noncrap decision is tantamount to fucking holy, and no deployment of anything but honest to god THE dude from big lebowsky can possibly cut it.
mircea_popescu: now the proposition here is, as i understand it, the following :
mircea_popescu: but mostly, there's going to be empty space. or rather : "crap" is the largest equivalency class.
mircea_popescu: alright. out of this space of 3bn units, once applying the rule of "looks like a game character" (or "mob" or whatever forumation of "is useful for eulora") there's going to be a few equivalency classes (most visible on say nintendo, bcause most annoying on nintendo, really, the wolf+1 gets blond hair ?)
mircea_popescu: a cube a thousand pixels wide will take, to be naively described point by point, about 3*256 Gb. this means there's about a trillion such cubes. yes ?
mircea_popescu: now let's see here : leaving aside for the sake of discoursive coherence the correct vectorial representation, and making do with a purely catesian approach (aka bitmaps) for the time being, on the expectation that while this rather than that allows much easier quantification, that rather than this doesn't magically pack much more complexity, greeks be damned :
diana_coman: that yes, I do; though I am partial to an sukhoi more than to an AI because at least it's a concrete beast I guess.
mircea_popescu: do you see the argument that "an sukhoi" is worth exactly the same as "an ai", in the sense that they're the ash of a cigar someone else smoked, the marblecake of an anal cavity someone else fucked, and, to quote the quite prophetic mr mel again, what's the point of a program that can't rewrite its own code ?
diana_coman: as above, it's negative value, below that "not much" for the sukhoi.
diana_coman: rather: unclear what *useful* thing it does
diana_coman: what, for buying an AI system that is "very powerful" when ran through the proprietary bundle & otherwise unclear what it does?
mircea_popescu: now, what's your scalar in this new context ?
mircea_popescu: otherwise, it... i dunno, it makes a skymap if you print the 1's as stars i guess.
mircea_popescu: if run through the proprietary bundle, it beats any human at go.
mircea_popescu: now, suppose we change out the sukhoi, and chande in instead "an artificial intelligence", specifically, a few TB of w/e worth of mostly 0s, binary values.
mircea_popescu: a very simple thing : systems of really really many linear equations with really really numerous variables hidden behind very small parameters.
diana_coman: myeah; I fell in love with it at 17 (when read the promise) and then promptly barfed by 19 (when encountered the full extent of the "practice")
mircea_popescu: n, and such others as lost themselves in that whale,
mircea_popescu: artificial intelligence as actually extant today is, leaving aside all sorta obscure lulz alf used to like referencing about expert systems and nato wargames, and the assorted lulzpile of neural networks and etc from the very dead 70s, and also leaving aside the whole life OF MINSKY, and then sussma
mircea_popescu: "artificial intelligence", like "flight" or whatever else, is as such and in the abstract a long cherished dream of humanity.
diana_coman: oh huh, only one? dunno, perhaps the idea is to reverse engineer, that might be about the only possible value I can see but it's dubious for all sorts of reasons.
mircea_popescu: i suspect the item as discussed is actually worth 0, yes, making complete idiots of the management of all the us' "allies".
diana_coman: if you are asking "what is this item worth to *you*", well, not much; the fuck do I do with this sukhoi.
mircea_popescu: well, there's one overarching bit of context here : the offer is to sell ~ONE~
diana_coman: because "what this item is worth" depends on more than just the item in itself and I have no idea even on the item really, let alone the context.
diana_coman: ugh, this seems out of my advising capabilities really.
mircea_popescu: so, the sukhoi 34 is a pretty cool plane ; in any case the basis of the russian defeat in the field of usg's pretense to air participation. now suppose for the sake of argument someone comes offering to sell one ; and suppose further that you're to advise on the purchase by providing a single scalar value : what the item is worth iyo.
mircea_popescu: bear with me here, because this is going to be lengthy.
diana_coman: the sort of accumulating accidental complexity, to link it in with trinque's thread re OS.
diana_coman: hm; I keep thinking that "perhaps I don't know enough about them to find the value" ; as I see them now, they are more accumulations of trial and error/overfitting/tinkering though so they seem of very little - or indeed negative - value tbh.
mircea_popescu: perhaps it's why you're asked then.
diana_coman: oh boy, I'm possibly the last person to ask for an argument pro "valuable" on those.
mircea_popescu: explain this to me, can you ? why are these valuable ?
mircea_popescu: the other part is that, well... why exactly is a texture / skin / shader / model / anything else in compgfx even valuable in the first place ?
mircea_popescu: so i don't even know on what basis i'm supposed to decide what, here.
mircea_popescu: ists worth the mention" discussed in the article and so on.
mircea_popescu: one is that a coupla years of diligent work have gotten eulora to this position where it actually outgrew the intellectual basis of the community such as it is in pretty much every respect. many things nominally upstream would be inputs for this decision spot but are sadly absent, from "well... what about scheme then ?" coincidentally brought up avbove to the self-obvious "how did the western world produce 0 graphic art
mircea_popescu: there's two portions to the problem.
diana_coman: mircea_popescu: hm, thinking now about it, I think there might be, namely the more directly game-relevant parts that are also not yet fully spec (eg character, structure/item etc); I need to go over it in this light, put the graphics to the side for now and see from there.
mircea_popescu: were there more chunks you could be working on ?
mircea_popescu: diana_coman, let's restate this, so currently work on nailing down the comms protocol is stalled on a definitive universal data model, which is stalled on graphical use in the client, that about it ?
ossabot: Logged on 2019-12-28 02:22:16 mircea_popescu: so listen... while you were away there's been a lot of progress ongoing in the republic, these guys now have blogs and stick to publishing planning schedules and such wunderbar alien techs. you gonna catch up with the group ?
Mocky: http://logs.ossasepia.com/log/trilema/2019-12-28#1956120 << yes I'm going to catch up. I have a lot of catching up to do
mircea_popescu: diana_coman, motherfucker, they force-inherited us with their idiotic everything-depends-on-everything-else model, did they!
mircea_popescu: pretty much the only lisp i use how shall we put this, on a day to day basis as opposed to some kinda ceremonial function
mircea_popescu: hanbot, a, that's actually a good point -- we might want to look into how gimp uses scheme for scripting
ossabot: Logged on 2019-12-29 09:56:30 mircea_popescu: neways, as to the burning "OTOH, I wonder if things like Apache or imagemagick get installed, how will the package management system work out, and how comprehensible will system stay?" question -- i see the merit of using the clean spot as a fixed point to attempt expanding cleanliness. so, it would work by apache becoming tmserv or w/e, and not sucking anymore.
hanbot: http://logs.ossasepia.com/log/trilema/2019-12-29#1956198 << i thought i got rid of the need for imagemagick with gimp's built-in scheme thing: http://thewhet.net/2019/11/a-little-bit-of-tinyscheme-a-lot-of-cozonac/?b=It%20turns%20out%20the%20box&e=thereby#select
diana_coman: oh; I was blissfully unaware of the extent of xml-shading, lolz.
mircea_popescu: 56 stars ans 86 forks, if this ain't the mark of the times i can't imagine what'd be
ossabot: Logged on 2019-12-30 07:02:12 mircea_popescu: http://logs.ossasepia.com/log/trilema/2019-12-29#1956250 << i honestly think it's the right approach. "identify an useful iota of acceptable internal structure and measure everything in that". next step self-obviously being, "once pretty much everything's within the first sigma, take another look at what you mean by acceptable internal structure".
ossabot: Logged on 2019-12-29 18:21:12 trinque: whaack: starting at beliefs is going to be a shaky foundation, but fwiw I "believe" most folks writing lisp today are compulsive masturbators like gabriel_laddel
mircea_popescu: http://logs.ossasepia.com/log/trilema/2019-12-29#1956260 << it seems indeed well borne by experience that the only thing "uses lisp" codes for nowadays is mental derangement.
mircea_popescu: it is, in fact, the foremost worth.
ossabot: Logged on 2019-12-29 18:20:02 trinque: which is not just a bunch of bawww. it's worthwhile to question wtf we use to value things within which *whole lifetimes* can be lost, wasted.
ossabot: Logged on 2019-12-23 02:15:42 mp_en_viaje: wut is this java bullshit, "oh, i've been fucking three women". mocky.org/three-nude-sluts when!
mircea_popescu: in short, we might be one or two "cultural aspects" away from anything like a useful or usable version of common lisp. now go fuck some teenie boppers on cam ; and when you're done have them eat your ass. ON CAM. and then maybe we lisp together, later.
mircea_popescu: nevertheless, all solved problems begin their life as solved problems through first being closed ; a refusal to close problem and a tendency to "keep problems open" is in the end specifically what neoteny even is.
mircea_popescu: but you don't ~like~ closed problems, or rather, you deem the putative existence of such a thing as closed problems as such a threat to your misconception of identity, it's almost like it's the koschei stealing the above-linked "cultural aspects" of young female chest away from you.
mircea_popescu: and what i mean by "there's no more space" is rather a lot in the vein of "you lot chafe and get massively butthurt at even the faintest version of what that world would look like anyone [by which we mean me, for lack of alternatives] can come up with". lisp is a tool for resolving closed problems, not for floating about the endless uncharted seas [http://trilema.com/2014/i-love-rochester-so-will-you/?b=try%20to&e=swim#
mircea_popescu: the main problem (and the reason i don't come across as much of a lisp fan) is that since the 80s complexity exploded significantly ; there's no more space for proto lisp as represented by mel ; nor for the classical lisp as represented by the defunct moron club.
mircea_popescu: THAT is the great value of lisp, in its classical form as understood by they regarded as luminaries by the "cultural aspects" crowd : that once you're done mentally digesting your problem, the peristalsis is as short and uncumbersome as it can possibly be.
mircea_popescu: does indeed dance with these wolves) ; and then once THAT is complete the solution takes five minutes of notation.
mircea_popescu: traditionally lisp coding consists of spending some time to ~well~ understand your problem ; the internal ast of this digestion is not well fit for language (even though yes, the bleeding edge of human technologies & understandings as represented by trilema
ossabot: Logged on 2019-12-29 18:19:37 whaack: diana_coman: if i search for a piece of lisp code to do something, I have a (perhaps naive) belief that I am more likely to find something well written, since there are fewer people using the language
mircea_popescu: ral carefully disolving every grain of actual thought into many galons of moosauce (which is what the "go playing" moo-engine even uses to identify "similar" bits "that might interest you")
mircea_popescu: http://logs.ossasepia.com/log/trilema/2019-12-29#1956258 << except for the part where lisp code production was never this paste in google / paste from github. this works well for ineptly structured languages like the moroncrowd favours, that are lenghtly verbose and cubical-dweller optimized, loaded with wooden tongue constructrs and in gene
ossabot: Logged on 2019-12-29 18:19:12 trinque: all my tmsr work to date could've been php or pyshits for all the difference it would've made.
mircea_popescu: we gotta think this through. and also let's see what jfw says and so on.
ossabot: Logged on 2019-12-29 18:13:10 diana_coman: re paster though tbh this attempt (coming as it does on top of the previous mess with python and its flask) left me looking again for not using either.
mircea_popescu: http://logs.ossasepia.com/log/trilema/2019-12-29#1956252 << the only practical approach to evicting python for the active diatesis of the logger is to rewrite that part in bash ; which, without an actual decision to do this from now on is imo inexcusably gnarly.
ossabot: Logged on 2019-12-29 18:11:26 diana_coman: ahaha, by now I see that trinque will measure everything in multiples of busybox.
mircea_popescu: http://logs.ossasepia.com/log/trilema/2019-12-29#1956250 << i honestly think it's the right approach. "identify an useful iota of acceptable internal structure and measure everything in that". next step self-obviously being, "once pretty much everything's within the first sigma, take another look at what you mean by acceptable internal structure".
mircea_popescu: not to tar you with a well used brush, but it HAS been the experience to date.
ossabot: Logged on 2019-12-29 18:12:50 whaack: trinque: First, I find it more pleasant writing in lisp. Second, I like the cultural aspect of a (present day) more niche language. Then for problems with python: i find the single line lambdas a strange requirement, and not having tail call optimization is lame (although i admit i can't think of many/any cases where that's been a problem)
mircea_popescu: http://logs.ossasepia.com/log/trilema/2019-12-29#1956251 << pleasant nothing. i bet you if i sit down to read what you've written "in" lisp ima in short order discover you wrote it in python anyway.
ossabot: Logged on 2019-12-29 18:07:26 trinque: which includes a shell
mircea_popescu: http://logs.ossasepia.com/log/trilema/2019-12-29#1956245 << indeed, ash/bash prolly the way to go. otherwise, "tcl/tk in python as tinkler" hurr durr, god help us.
ossabot: Logged on 2019-12-29 18:02:34 trinque: if we were to retain lisp, I'd say pick one, don't have a python alongside it, and don't expect to use much "open source" to help you.
mircea_popescu: http://logs.ossasepia.com/log/trilema/2019-12-29#1956242 << more probably, something like "lisp castrated into scriptlang"; thoguh there's also lua or say tcl...
mircea_popescu: naggum for that mattter is being whipped upside down in hell even as we speak, specifically for "wtf did you do with your calling, you schmuck, killed yourself before hearing it ? TAKE IT LIKE A LITTLE BITCH".
mircea_popescu: one irresponsible twit gave himself fatal ulcer, the other and so on.
ossabot: Logged on 2019-08-26 13:49:11 phf: you ought to consider moving here, or at least coming for a visit. the place is a lot livelier than dc, food is better, girls are prettier, things are generally cheaper, etc. etc. etc.
ossabot: Logged on 2019-12-10 15:14:42 mp_en_viaje: and in general -- the absentee, the insufficient, the "otherwise busy", the butthurt, physically or mentally -- better pray they're in fact quite as inconsequential as they seem.