log☇︎
9700+ entries in 0.009s
ossabot: Logged on 2019-10-22 16:28:01 bvt: some thoughts: currently feeder app takes 3% CPU @ ~2.4 kb/s when feeding data into O through HG, because the bottleneck is in FG reading, and lots of overhead seems to be coming from retarded tty interface, which forces reading of individual bytes.
mp_en_viaje: http://logs.ossasepia.com/log/trilema/2019-10-22#1947723 << imo 3% is already exaggerate. i'd be thinking more like a half percent, ideally.
ossabot: Logged on 2019-10-22 15:03:50 asciilifeform: for intervals of expansion : would also help if had any way to gauge demand. ( atm i have 1 ~maybe~ dulap-install being considered by diana_coman & mp. anyone else ? )
mp_en_viaje: http://logs.ossasepia.com/log/trilema/2019-10-22#1947675 << gauge nothing. glaringly absent from your alleged plan is the 200 hour / month time allocation for creating that demand, with clear by day of week scheduling. is it gonna be 8am to 5pm on tuesdays and 4pm to midnight on wednesdays or what ?
BingoBoingo: bvt: I'm sure they can try to work something out aroung your schedule.
ossabot: Logged on 2019-10-22 06:31:20 mp_en_viaje: incidentally, mylord verschlimmbessert : how's the http://logs.ossasepia.com/log/trilema/2019-10-22#1947499 notion sound ? either take a girl from home, have her do some nude work on the world's greatest beaches to spiffy up your blog illustration-wise ; or else i guess something can be provided locally.
bvt: http://logs.ossasepia.com/log/trilema/2019-10-22#1947565 << ty for offer, unfortunately something like this would be possible for me only in feb.
lobbesbot: BingoBoingo: The operation succeeded.
BingoBoingo: !Qlater tell billymg http://paste.deedbot.org/?id=9r9U
ossabot: Logged on 2019-10-22 06:42:27 mp_en_viaje: anyway, i suppose the logical next step is for the remarkably productive bvt to do some benchmarking re speed of possible candidates (a list including atm the chacha and serpent -- knowledgeable folk feel free to propose more candidates) so as to have some practical basis.
bvt: http://logs.ossasepia.com/log/trilema/2019-10-22#1947578 << yeah, i will definitely test, including keccak (can pick c impl. for tests)
ossabot: Logged on 2019-10-22 06:39:48 diana_coman: http://logs.ossasepia.com/log/trilema/2019-10-22#1947561 - I can see that; my thinking was not towards using sha1 but more towards permitting other-than-serpent mainly because 1. serpent is still snake-oil and only adopted-for-lack-of-better option afaik b. maybe tmsr makes its own hash function next year or something (ha!)
bvt: http://logs.ossasepia.com/log/trilema/2019-10-22#1947572 << allowing arbitrary hash functions would create more bloat -- either i'd have to use some generic crypto abstractions, or hack up the build system and unconditionally enable all the accepted crypto algos at build time to use them directly. it's only chacha and sha1 that are located among e.g. memcpy in lib/ and available uncoditionally. the rest
ossabot: Logged on 2019-10-22 05:46:09 diana_coman: mp_en_viaje: why mandate a specific hash function (esp given that there isn't any clear way to base such mandate on "this is best hashing")?
bvt: http://logs.ossasepia.com/log/trilema/2019-10-22#1947556 << well, as long as selection procedure is rational, can always explain why have chosen this one and explicitly say that this is not an endorsement.
ossabot: Logged on 2019-10-22 02:58:56 mp_en_viaje: and thanks bvt, megaworks there.
bvt: http://logs.ossasepia.com/log/trilema/2019-10-22#1947554 << and ty, mircea_popescu, diana_coman!
ossabot: Logged on 2019-10-22 02:32:27 mp_en_viaje: i also kinda like the idea of permitting each user to pick his own rng-hash-crypto key.
bvt: http://logs.ossasepia.com/log/trilema/2019-10-22#1947552 << this is a nice idea, i will implement it
bvt: but how much of cpu percentage would we want to dedicate to good hashing? if e.g. keccak is in place, what could happen (but only experiments will show) is that the bottleneck would be in the CPU -> if the system runs low on entropy somehow and self-hashes (which goes through the HG now, should it go through HF? (would also make sense from Size/2 decision rule)), we'd have an easy DoS vector against the
bvt: some thoughts: currently feeder app takes 3% CPU @ ~2.4 kb/s when feeding data into O through HG, because the bottleneck is in FG reading, and lots of overhead seems to be coming from retarded tty interface, which forces reading of individual bytes.
ossabot: Logged on 2019-10-22 02:29:31 mp_en_viaje: http://bvt-trace.net/2019/10/fg-fed-linux-rng/ << the most important q here is, are we going to mandate serpent ? or are we going to permit legacy sha1/chacha ?
bvt: http://logs.ossasepia.com/log/trilema/2019-10-22#1947549 << i certainly liked serpent most, because it went through at least some of form republican investigation, but decided not to hurry too much putting it everywhere.
BingoBoingo also still not at all inclined to evacuate without a better landing place.
BingoBoingo: If they don't settle at all they next course is weighing which action imposes the most cost on them at the least expenditure.
BingoBoingo: The primary ground for damages in the local legal environment once again, are based on my lost income which... not going to generate an incredible huge number.
BingoBoingo: The damage a judgment does would be constrained in the local environment.
BingoBoingo: And derps still spend in their defense
BingoBoingo: asciilifeform: Well, the Fiscalia spends to prosecute that beyond the initial complaint.
asciilifeform: BingoBoingo: while it'd be quite satisfying to see the derps jailed, it dun do much for the filling of our crater afaik
BingoBoingo: asciilifeform: Further action in that direction by the damned leads to the accumulation of criminal fraud charges.
BingoBoingo: I do have a meeting with different counsel tomorrow afternoon as well.
asciilifeform: BingoBoingo: supposing they give rat's arse re 'longterm', i suspect latech is in the 'take the money & run' phase of scamlifecycle
BingoBoingo: Ending up in the consumer protection docket at all imposes long term costs on them.
BingoBoingo: asciilifeform: Even the less than full press in the consumer protection docket, which is less expensive piles costs on them faster than myself as its a mostly diy job.
mod6: I'll have to admit, I don't care about all the rest of the noise, or whatever, my singular goal is to avoid paying out the rest of the contract.
asciilifeform: ( iirc mp already detailed this. )
asciilifeform: but does req. a credible threat
asciilifeform suspects that settlement is the only way this could come out +ev
BingoBoingo: asciilifeform: This is why I remain open to settlement while applying maximum pressure.
asciilifeform: 8k granted may be their budget for fighting plain termination of the contract
BingoBoingo: asciilifeform: The cap on their budget is bounded by their maximum upside by repairing the contract.
BingoBoingo: The damned's budget is 8k USD or less unless they suck at math. Their spending increases faster than ours because up until this enters the courtroom /me is an active legwork participant.
asciilifeform: but the decision is ultimately on mod6's desk
asciilifeform: mod6: i'd even be ok w/ spending a larger (so long as ~bounded~) sum, if we had some notion re the prospects
mod6: asciilifeform: that's pretty much where I was at as well. .25 BTC to evac. which is about ~$2k in today's prices.
asciilifeform pictures latech reading the l0gz, and thinking 'haha, suckers, we'll spend N+1'
asciilifeform: mod6: imho makes sense to spend a sum that is strictly < than the cost of evacuating BingoBoingo . but not moar.
mod6: Do we have a specific budget in mind here for what we're willing to spend? Because, for me, it's not very high.
mod6: It totally agree that costs will soar if this thing goes to court.
asciilifeform: mod6: plox to do BingoBoingo's 1st, his fire is moar immediately hot
mod6: Yes, to both.
BingoBoingo: mod6: Did you get a chance to read the lawyer meeting notes yet?
asciilifeform: mod6: didja get a chance to read draft 1 btw ?
asciilifeform: mod6: ty, noted
mod6: asciilifeform: I think it sounds like a good place to rack the foundation's server (as previously discussed).
asciilifeform: ( all folx considering emplacing ~anyffin~ in asciilifeform's rack -- plox to write in. no commitment is implied, would like to simply gauge the immediate prospects. )
asciilifeform: w/ the new philosophy of tmsr iron being spread out worldwide, i dunno what kind of occupancy can expect in the immediate term. which is why began with dwarf rack rather than shelling out for the full orchestra immed.
asciilifeform: for intervals of expansion : would also help if had any way to gauge demand. ( atm i have 1 ~maybe~ dulap-install being considered by diana_coman & mp. anyone else ? )
asciilifeform wonders if he's thinking about this backwards, and oughta instead model the upper half of (expanded) tower as a separate numeric system, w/ same constants but Te == 200.
asciilifeform: BingoBoingo and mp were right, btw, it appears that a well-populated rk fixture (and, possibly, shared-machine) are the only way to actually cross that bridge w/out further raising subscriber colo charge.
asciilifeform: ( ^ in the simplest formulation, a full occupancy that brings in net >200 $ (the added expense of upgraded tower) w/out eating into irons amort. budget. )
asciilifeform will post in #2 example scenarios for 'full occupancy' and what is required to permit expansion of plant.
asciilifeform: ( elementarily, the irons ~must~ amortize in the given time spans, to cover cost of maintenance/replacement )
asciilifeform: i'ma have to recalculate w / machine-amort. as separate figure.
asciilifeform: ( in the black revenue-wise, that is; to cover the outlay for the machines per se would take at that rate 10y, supposing an improbably static btc exch rate )
asciilifeform will begin proper rework of costs draft #2 tonight. prelim. spreadshit however suggests that w/ 45% margin, three leased 'dulaps' already would bring in enuff to upgrade the tower (+20u, 10x pipe, 2x amperage) ~and~ remain in the black...
ossabot: Logged on 2019-10-22 13:26:30 BingoBoingo: asciilifeform: Your way is a clean way to handle WP given you have had other things you'd rather do than clean the messes, a product of its time.
asciilifeform: http://logs.ossasepia.com/log/trilema/2019-10-22#1947653 << ftr would have fixed, that wasnt the issue; simply wasn't literate enuff in phpism to do a proper cleanup. ( arguably still even nao , php to asciilifeform is ~= cobol )
asciilifeform: DanielBTC (~macaxeira@unaffiliated/danielbtc) has joined #trilema << anyone know whothefuq that is ?
asciilifeform found that there ~remain~ 'microshit quotes' in certain posts, where text had evidently been pasted from wp output. will have to write a sqlism to remove'em, but prolly not today an' not this week.
asciilifeform: when moved, will finally have that static ip...
ossabot: Logged on 2019-07-18 21:21:31 asciilifeform: sooooo... loox like all things die. and nfs hosting co finally went fullbore stupid. asciilifeform's www is hosed, they 'upgraded' php forcibly.
asciilifeform: ( how retarded? see also thrd. )
BingoBoingo: Qntra-wp similarly legacy system, trapped for the substantial overhead necessary to port the database holding the content. Qntra however is less blog and more wire service.
asciilifeform: the carrier where it lives, however, is terminally retarded, and it's to be moved as soon as i have a free hand.
asciilifeform: fwiw it has never been detectably vandalized, to date.
asciilifeform: BingoBoingo: my wp is squarely a 'legacy system', analogous to e.g. 'mpb'. i dun propagandize its use, nor intend to separately genesis it, imho mp has the wp question 100% covered .
BingoBoingo: asciilifeform: Your way is a clean way to handle WP given you have had other things you'd rather do than clean the messes, a product of its time.
ossabot: Logged on 2019-09-05 20:04:50 asciilifeform: back in '07 sewed his php wp with 'fascism' , it can only write to db, not to disk directly ; so was never able to use the bitmap processing , www uploader, related, features
BingoBoingo: asciilifeform: Well, the way wordpress updates, if soemthing gets updates whatever files you customized get replaced. Same if the theme "updates"
asciilifeform: hell knows. seems to remove the mutilation tho, w/out breaking anyffin (if anyone sees a breakage from this patch on my www, or remaining mutilated quotes -- plox to write in)
asciilifeform: BingoBoingo: ty, that actually worked. ( tho, funnily, i had to correct the author's own piece, on ~his~ wp evidently did NOT work, his contained microshit-quotes!! )
asciilifeform: the human-readable urls thing in particular, in mpwp, is quite appetizing.
asciilifeform: BingoBoingo: i expect to move entirely to mp's wp when i get to moving www finally. but atm hands quite full. ( it'll be somewhat tricky, will have to port the coad printer extension thing to it, and various other things i changed over 12yrs+ of my www)
BingoBoingo: mp-wp kills em already, and Qntra-wp code hasn't much changed since the cazalla days
asciilifeform writes all articles by hand, tags an' all
asciilifeform: fwiw i dun enter text via the 'pretty print' wp editor, but as raw htm paste, so it prolly aint the editor that mutilates
asciilifeform: ( will guess that it's 1 line somewhere in that war crime.. but where? )
asciilifeform: i'd like to finally patch mine
asciilifeform: tangentially, other mp_en_viaje point -- anyone readily recall how to banish mutilation of quotes from wp? ( aka 'microshit quotes' )
asciilifeform: diana_coman & mp_en_viaje are the king/queen of the prospective customer base, by weight, and so if no one adds anyffin in next few hrs, i'ma work w/ what's been given so far for draft 2.
asciilifeform: would like to accumulate all possible nitpicks of 1 before setting out tho.
diana_coman: I'll wait patiently for draft 2, there's no need for sneak previews in the logs or anything.
asciilifeform: diana_coman: moar in brief re expansions (will address pedantically in draft 2) -- they will have to be financed by actual revenue, and BingoBoingo-style auctions, i do not have massive capital (and what i have, is earmarked for physical irons, already ordered 'dulaps' and will be purchasing yet-moar of'em once it becomes clear that asciilifeform won't live alone in the rack)
asciilifeform: on top of this, i have a gentoo for'em; an automated system for provisioning; and rk is 'most cpu for the watt' box that i'm currently aware of, is imho excellent fit for a wattage-constrained initially small tower.
asciilifeform: diana_coman: re rk -- i included rk plant from the following logic : 1) it was best-selling product in era of piz 2) i am sitting on a crate of brand-new rk that did not get invoiced to piz or get to fly 3) pre-existing interest ( e.g. diana_coman's ) in inexpensive www mirrors.
diana_coman: asciilifeform: please do, thank you.
asciilifeform: diana_coman: 'fully populate' i expect will happen quickly when l1 finds prospectus satisfactory. e.g. 4 'dulap'-type boxen and 1 shared-www per BingoBoingo's schematic already exhausts the energy supply. at which pt i'ma expand to full tower. and diana_coman is right imho, this oughta be detailed in 2nd draft.
diana_coman: take all the space you need, sure; I just want to see it or whether there will be any of it.