tmsr - Search: phf

1900+ entries in 0.079s

a111: Logged on 2018-07-18 12:32 phf: http://btcbase.org/log/2018-07-18#1835696 << i believe at least in the case of allegro the ownership is the dks/symbolics model (or perhaps soekris model), guy who made it sells it and there's enough sales money to periodically fund developers, add new features, etc (considering that some of the contracts are finance and gov)

asciilifeform: phf: for all i know, ~today's~ spaceprobes fly with java on winblowz. i am speaking of the 'golden age' 1980s-1990s items.

asciilifeform: phf: so i explicitly rejected the standard's routine.

asciilifeform: phf: observe however that it is impossible to make use of your approach re cmdline args. the standard unambiguously mandates variably-lenghted strings ( i.e. dualstackism ) for that. ☟︎

asciilifeform: phf: when i started out, i had no notion that anyone would help me by producing a working libc-less gnat.

ave1: phf, yes that's the right approach

asciilifeform: phf: it will require moar massage to actually produce a compact binary, and so i did not go with it; but it is still prolly The Right Thing long-term

asciilifeform: phf: i can't say this is a wrong approach

ave1: phf, asciilifeform's ffacalc uses only a few C functions all for IO / exit and command line arguments. I just have to provide the right functions in the ZFP library for ffacalc (currently first on conveyor but also some real life priorities these weeks)

asciilifeform: phf: iirc i used cism for put/get on stdio, and for cmdline.

asciilifeform: phf: first it must work with asciilifeform's item . and only then i can dare to speak of 'enuff for erryone!'.

asciilifeform: phf: my specific motivation in asking ave1 for this, was to enable building e.g. ffa where binary is small enuff to manually disasm and audit.

asciilifeform: phf: adacore actually distributes a compiler like this. but it is commercialware, and i've never seen it.

asciilifeform: phf: noshit, most c proggy won't build on a gcc with no libc!

asciilifeform: phf: are we speaking of http://ave1.org/2018/building-gnat-on-musl-updated-tar-line/ or http://ave1.org/2018/gnat-zero-foot-print-take-2-no-c/ ? these are separate items

asciilifeform: phf: ave1's item is simply a frozen snapshot of (iirc 2016) gnat.

asciilifeform: phf: why would either of these belong in a compiler ?!

asciilifeform: phf: say what you will re ada standard, but e.g. ffa is ( afaik! ) a nontrivial and at same time ada2012-compliant proggy. ☟︎

asciilifeform: phf: as you can prolly tell, asciilifeform has an old-fashioned, ГОСТ-like view of standards. in that a standard really ~must~ specifically make vendor lock-in impossible. and if it fails to do this, it is broken and oughta be corrected ( or thrown out. )

asciilifeform: phf: i do know, yes, that the standard was stunted at birth.

asciilifeform: phf: but asciilifeform is a subscriber to the 'what is not mandatory is forbidden' school of prog lang standard.

asciilifeform: phf: i cannot show you, as it does not mention every possible abuse, lol

asciilifeform: phf: holyfuq, what is this if not breaking standard ? standard permits no such horror

asciilifeform: phf: how not ? seems like if you add a feature that 'must be in compiler', you ipso facto broke standard

asciilifeform: phf: sounds like a , what, 100ln proggy

asciilifeform: phf: i suspect that the diff could be accounted for by diff in problem domains; but i did not find the advantage to be 'immense'

asciilifeform: phf: in commercial work i use things that would turn the strongest stomach. now, as always, speaking of civilized work. ☟︎

asciilifeform: phf: seems like that was lispworks, not allegro

a111: Logged on 2018-07-16 15:44 mircea_popescu: thoughts plox! (and i specifically want everyone to say at least an ack, so let's page asciilifeform ave1 ben_vulpes BingoBoingo danielpbarron diana_coman hanbot lobbes mod6 phf spyked trinque )

asciilifeform: ( recall phf's story of 'cipherpunks culture' )

mircea_popescu: phf a ok.

asciilifeform: mircea_popescu: as i suspect the folx who stole phuctor v.1 and v.2 src discovered, and prolly not much surprise to anybody else, phuctor is not a fully automatic mechanism. it requires asciilifeform's hands , applied on fairly regular basis, to function correctly. sorta like phf's logtron, for instance.

asciilifeform: phf: the pov that programmingism is a kind of tard reservation ~specifically because~ of the ease of reusing ancient src, is imho defensible.

asciilifeform: phf: in my particular case, phuctor backend is unpublished specifically because i do not want to encourage people to build on top of it. when i am killed, i'd hope that if folx still feel a desire for a phuctor they will take what asciilifeform ~publicly~ wrote on subj, and make a superior one, rather than trying to maintain the old ball of hack.

asciilifeform: phf: observe that using mechanisms created by dead people is ~very~ expensive, and seldom worth it. and they do not magically return to life by virtue of other folx having a raw copy of sores.

mircea_popescu: phf i do not believe the bus factor is even remotely important there.

asciilifeform: phf: phuctor was stolen, wholesale, complete with iron it ran on, no fewer than twice. usg still dun have , afaik, a working clone.

mircea_popescu: thoughts plox! (and i specifically want everyone to say at least an ack, so let's page asciilifeform ave1 ben_vulpes BingoBoingo danielpbarron diana_coman hanbot lobbes mod6 phf spyked trinque ) ☟︎☟︎☟︎☟︎

a111: Logged on 2018-07-13 21:41 asciilifeform: phf, trinque , danielpbarron ^

asciilifeform: phf, trinque , danielpbarron ^ ☟︎

mircea_popescu: phf this sounds great, and works fine for as long as you stick to pulp fiction (this ~particular~ narrowness of viewport is why dumas never counted as an important writer). try and step outside of the threemusketeering and see if it can still be meaningfully applied.

a111: Logged on 2016-09-14 19:35 phf: madam blavatski and her followers did a lot of damage for years to come. it was literally in her writing that original meaning doesn't matter, as long as the translation sounds cool (there's a rationalization for why that somehow supposed to make sense)

mircea_popescu: inb4 phf misreads "pdf"

asciilifeform: mircea_popescu: conceivably latex can be cured. iirc phf has a slow-burning march in that direction.

asciilifeform: phf also documented some other corruptions of latex in recent decade, but the details currently escape me, iirc in the log somewhere

a111: Logged on 2018-07-11 18:40 phf: i'm unconvinced there can be such a thing as hygienic latex, possibly there can be ghostscript if you pull out the pdf support (?)

a111: Logged on 2018-07-11 18:39 asciilifeform: phf: afaik there does not currently exist anything like a hygienic latex or ghostscript recipe

asciilifeform: phf: afaik there does not currently exist anything like a hygienic latex or ghostscript recipe ☟︎

asciilifeform: phf: why ?

asciilifeform: phf: the ones with xorg -- yes

diana_coman: hanbot, right, it's keccak hashes so ofc v.pl barfs although it's weird that it first presses so much of it and only then randomly throws up; at any rate, on different arch, phf's vpatch pressed the mp-wp genesis so at least the genesis itself seems all right

a111: Logged on 2018-07-11 03:49 esthlos: http://btcbase.org/log/2018-07-10#1833105 << fwiw phf, esthlos-v resolves the graph in the same manner as btcbase, and can press based at any node. of course, without checking the hashes, currently...

hanbot: diana_coman the regrind presses here without errors with phf's vpatch. i didn't try pressing with v.pl, possibly they're incompatible...?

a111: Logged on 2018-07-10 14:17 phf: like i said couple of days ago i'm going to forward merge that whole right side, right now the only advise is to delete the right hand side because none of the extant V's can resolve that graph, obviously a suboptimal suggestion. i'm working on a better grapher, but until then..

esthlos: http://btcbase.org/log/2018-07-10#1833105 << fwiw phf, esthlos-v resolves the graph in the same manner as btcbase, and can press based at any node. of course, without checking the hashes, currently... ☝︎☟︎

asciilifeform: btw, phf, dollars to doughnuts you have a loose battery connector. the whine and reset unixtime is dead giveaway

asciilifeform: PeterL: iirc phf had some recipe for these

BingoBoingo: <mircea_popescu> maybe if we keep going in this vein we can get the un to sponsor trilema. << Phf hasn't turned the chromebook into ladypants yet

asciilifeform: phf: possibly pertinent diff, is that i regreased it

asciilifeform: phf: in graveyard nao, but i had it running continuously off the dc brick for coupla weeks

asciilifeform: phf: didja take the heatsink off ? if yes, gotta regrease

a111: Logged on 2018-04-02 22:04 asciilifeform: trinque, phf , other vtronicists : http://therealbitcoin.org/ml/btc-dev/2018-April/000296.html

asciilifeform has not eaten phf's proggy yet

mircea_popescu: in my own notes phf had written a replacement!

a111: Logged on 2018-07-10 14:43 phf: so policy wise, like you say "no cyclical graphs" you can say "no identical changes in more than one v patch"

asciilifeform: phf: right, i grasped this, hence why i wrote a demo impl of trinqueian algo ( http://therealbitcoin.org/ml/btc-dev/2018-April/000296.html ) which actually does solve it

asciilifeform: phf: didja try it with my ancient v99 ? how does it die ?

spyked: that oughta work. I can't find the discussion where phf recommended this approach of keeping just one side of the tree in the patchset.

diana_coman: phf, I'm having trouble with latest vtools it seems (gcc and gnat version at the end of paste): http://p.bvulpes.com/pastes/WJios/?raw=true

a111: Logged on 2018-07-09 14:46 phf: i'm remembering when i was in costa rica there was a couple of days when it was raining non stop, and it was "muggy" and everything was constantly wet. visions of tropical diseases from the age of exploration books

a111: Logged on 2018-07-09 16:03 mircea_popescu: http://btcbase.org/log/2018-07-09#1832943 << same here, and generally throughout the logs. even phf mostly yielded ye olde box.

mircea_popescu: http://btcbase.org/log/2018-07-09#1832943 << same here, and generally throughout the logs. even phf mostly yielded ye olde box. ☝︎☟︎

BingoBoingo: phf: Muggy in a "cold" way, but less consistently

BingoBoingo: <phf> ave1: i believe we're beyond linux haters now, this is the uncharted territory of everyone being an idiot << I am pretty sure this line was crossed irrevcably sometime between 2011 and 2015. Now we are just hitting new frontiers of idiocy

esthlos: phf I fixed those files if you'd like them, at http://files.esthlos.com/crypto/esthlos-v/v_genesis.vpatch and http://files.esthlos.com/crypto/esthlos-v/v_genesis.vpatch.esthlos.sig

a111: Logged on 2018-07-07 23:58 phf: esthlos: your v_genesis patch is broken, http://btcbase.org/patches/v_genesis/file "contact" etc.

a111: Logged on 2018-07-07 23:27 phf: i've been using vpatch/vdiff without full blown v, because i can order the patches by hand (and there's now an explicit ordering provided by manifest), and vpatch verifies the hashes for me. it would be handy if i could also press existing sha patches with `vpatch -a sha` or whatever

asciilifeform: phf: at this point, if something dun build under musltronic gnat, the problem is with item, not with the gnat

asciilifeform: phf: theoretically it oughta run there without modification

asciilifeform: phf: are you in the midst of planting the thing on c101pa box?

asciilifeform: phf: ave1's item is pretty great, it resulted in tarballs containing working bins of amd64 and arm64 gnat; the latter worx great on my desk rockchip. the only item that dunwork of yet is the repeat of said process ~on~ arm64 (to crossbuild amd64 gnat there)

asciilifeform: phf: as discussed in the recent diana_coman thread, if you want systemwide musl-anythings, gotta have a pure musl box.

asciilifeform: phf: only on a system where the systemwide crapola is traditional (glibc) and your musl item is a homedir-local build, a la rotor.

asciilifeform: phf: ave1's recent breakthrough was specifically this -- a rotor-style cross-compile process that takes an existing amd64 gnat, and builds a particular other gnat for whatever arch

ave1: phf, link was wrong should have been: http://ave1.org/tarpit/tmsr-pgp-genesis.vpatch.ave1.sig. I updated the link in the post.

lobbesbot: ave1: Sent 1 day, 4 hours, and 57 minutes ago: <phf> ^

mircea_popescu: now, politically i was going to take a soft approach to it, but since phf decided to make a challenge out of it for no reason i can imagine, what can i say, let's make it a big deal.

a111: Logged on 2018-07-07 01:20 mircea_popescu: phf define slow adoption!!

mircea_popescu: i'm not suggesting either ; i'm just saying that you have two ways you could proceed. either identify among the code you wish to reuse a tree to nestle among, and then your vtron would be an eucrypt downstream item, or a phf vtron item or whatever you pick ;

mircea_popescu: also, can integrate phf's code, same way

esthlos: so two things I see are: 1. what to do for hasher? somehow integrate phf's item into my vtron? 2. what do to for diff/patch? lisp McIlroy?

mircea_popescu: now, shelling out to diff / patch is one thing (though somewhat iffy for reasons discussed in a long thread, which phf mostly fixed). but shelling out to hasher is no longer possible because keccak implementation isn't shell.

a111: Logged on 2018-07-06 23:40 phf: http://btcbase.org/log/2018-07-02#1831298 << unless you have particular preferences, i'm going to throw it into the same patchset as ircbot. it'll follow the grand ircbot tradition of a genesis that never the less relies on the adjacent patches

spyked: http://btcbase.org/log/2018-07-06#1832410 <-- ty phf! yeah, that works. though eventually it will be reground into a single tree, as per http://btcbase.org/log/2018-05-25#1818999 ☝︎☝︎

mircea_popescu: phf define slow adoption!! ☟︎

a111: Logged on 2018-07-02 12:07 spyked: ^ phf, can you pl0x add this to the btcbase patch list? and http://lucian.mogosanu.ro/src/trilemabot/ ?

lobbesbot: phf: The operation succeeded.

a111: Logged on 2018-07-06 17:06 mircea_popescu: phf do you remember why ave1 's thing isn't in http://btcbase.org/patches ?