log☇︎
261500+ entries in 0.169s
trinque: in no small part because V is utterly useful as a tool on its own
trinque: http://btcbase.org/log/2016-09-09#1537866 << aha, evolving from earlier build scripts but has to go ☝︎
asciilifeform: (though laser paper burns atrociously, it is treated with fire retardant so that to survive the fuser)
phf: one day orcs perhaps will use it to last through the night, as a kindle
asciilifeform: naggum even noted this at some point in mid-'90s
asciilifeform: clhs wants to live on the thin, glossy paper that bibles were once printed on.
asciilifeform: phf: 1300 pg of laser paper takes up too much width
phf: why is that? i've never tried
asciilifeform: phf: yes. it's the one nobody can print into 1 b00k
phf: asciilifeform: i'm sure you've seen this version http://glyf.org/tmp/CL-ANSI.pdf
asciilifeform: mircea_popescu: i think boolcrap and i knew that chick in meatspace.
asciilifeform: i want the thing printed on archival paper and bound in motherfucking human leather.
phf: i have an ansi copy, it's a piece of shit though
asciilifeform: some dude got to naggum's death library copy just before i did...
asciilifeform: phf: tried many times to have it printed in oncie copy, but NOBODY will promise it in ONE COVER
asciilifeform: phf: hilariously, i've been trying to buy the dead tree for eons
phf: fwiw dpans has a copy of tex source, first or second to last draft before the standard was sent over to ansi. there's two projects that cleanup that tex, one is dpans2texi which produces texinfo formatted (this is what i use from emacs) and another by some russians that produces a clean pdf with hyperlinks
asciilifeform: in all fairness, the lisp world was hit by the equivalent of thermonuke war, and buncha stuff was lost / mixed up / dusted over.
mircea_popescu: the ~only possible interpretation is that people simply suck at knowing what they got.
asciilifeform: given as the thing PREDATES HTML
asciilifeform: considering that we can be quite certain that this is what it began life as.
a111: Logged on 2016-09-08 21:44 gabriel_laddel: I'm translating the spec from HTML to CLIM.
asciilifeform: http://btcbase.org/log/2016-09-08#1537758 << i never understood why the thing was published as html and not as a sexpr thing that gets htmlificated when you like ☝︎
a111: Logged on 2016-09-08 21:58 mircea_popescu: wait, the clim authoritative spec exists as a html ?!
asciilifeform: http://btcbase.org/log/2016-09-08#1537772 << before clim, common lisp's authoritative spec exists as... dead tree. while the version that ~everyone actually works from is html. because no, it is unusable without hyperlinking. >> http://www.lispworks.com/documentation/HyperSpec/Front ☝︎
mod6: thanks Gentlemen. I'll keep polishing here. more to come, I'm sure.
asciilifeform: it is more like flying than driving, if you will.
asciilifeform: mod6: dunno that vtronic builds will ever be a 'most people' affair. ☟︎☟︎
mod6: for you and I, this is routine.
mod6: the setup of v; verify v; place everything where it needs to go, etc.
asciilifeform: was there a major point ?
mod6: not a problem then.
mod6: ok. that's a minor consideration.
asciilifeform: mod6: see my original rotor makefile. it chdirs to the dep dirs, makes there, etc.
mod6: that's the easy part imho. the hard part is getting v, et. al., set up.
mod6: <+mircea_popescu> " then navigate to pressed dir, " << i mean. << ah, ok yeah one ~might~ be able to do that. not an expert on make tho.
mod6: which, seems to be the "pistols" way
mircea_popescu: " then navigate to pressed dir, " << i mean.
mod6: well, one would still need to press everything out, which includes the source and makefiles in this case.
boolcrap: same hair same body type
mod6: does this seem acceptable?
asciilifeform: boolcrap: because i think she does.
asciilifeform: boolcrap: looks like anyone you remember from the old dayz ?
mod6: i'll see where i can get to with this new idea. however, a person would have to do this to build the entire thing: get v; verify v; put vpatches, seals, and keys in place; press; then navigate to pressed dir, then `make`. ☟︎
mod6: i could go on here.. but let's just say i didn't think it all the way through.
mod6: yeah, after discussing with myself in here mainly, i'm not in love with that solution either.
asciilifeform: mircea_popescu: at one point we had 3 (then 2) quite separate types of builder.
mod6: (working on that / testing it now)
mircea_popescu: ok let's approach it this way : what is so special about the make files that they get their own tree ? why not say, all .c files get their own, all .h files get their separate own ? inasmuch as you can't use the makefiles without the sources, they belong in the same tree as the sources.
asciilifeform: ^ the right thing.
mod6: anyway, i might be able to get it so that you press the tree, including the makefiles, and then just build.
mod6: well it's earliest implementation, was a clone of our build scripts, which, used V to pull and press the source.
scriba: Logged on 2016-09-08: [23:52:51] <mod6> this is because the makefile process does it's own V press.
mod6: <+mircea_popescu> http://log.mkj.lt/trilema/20160908/#1160 << so basically we got v running twice ? << yeah, which wont do. after my monologue earlier, I have a hunch that I might be able to get it down to a single press.
asciilifeform: invocations of v are meant to be ~conscious~
asciilifeform: i admit that the thing where v is invoked from inside a make always bothered me.
scriba: Logged on 2016-09-08: [23:59:57] <mod6> Altogether, since we have V, I like the idea of keeping things like makefiles and buildscripts out of the main source tree. One can get V, press the makefile project. Run a `make`, which will in-turn, press everything via V and then build with buildroot.
mircea_popescu: http://log.mkj.lt/trilema/20160908/#1163 << so then every time you try to get software running you have to two two presses, one of makefiles, one of sourcefiles ?
scriba: Logged on 2016-09-08: [23:52:51] <mod6> this is because the makefile process does it's own V press.
mircea_popescu: http://log.mkj.lt/trilema/20160908/#1160 << so basically we got v running twice ?
mircea_popescu: Framedragger can we stop having this "multiline" bs unless specifically togled on somehow ? ☟︎
asciilifeform: try in the classic log ?
scriba: Logged on 2016-09-08: [19:19:02] <mircea_popescu> yeah but the wrong one.
mircea_popescu: http://log.mkj.lt/trilema/20160908/from:998/to:1163#998 << so then every time you try to get software running you have to two two presses, one of makefiles, one of sourcefiles ?
scriba: Logged on 2016-09-08: [19:19:02] <mircea_popescu> yeah but the wrong one.
mircea_popescu: http://log.mkj.lt/trilema/20160908/from:998/to:1160#998 << so basically we got v running twice ?
mod6: asciilifeform: verified, thanks!
asciilifeform: boolcrap: how's the gulag ?
mod6: i've got: 72.83.9.184 listed on the foundation site. is it back up now with different ip alf?
asciilifeform: pre-dates the ~last~ fiber cut, heh.
mod6: Then let the makefiles do all the rest of the heavy lifting.
mod6: The simplest solution would be to perhaps have a deedified tarball of makefiles, much like trinque already has, get the deed, verifiy it, decode it, extract it, navigate to the make script and `make`.
mod6: This hypothetical solution, even if it does work, wouldn't make it a one-button-push solution. Why? A person would need to get V + V.sig, verify, create a .wot dir, sync the patches + seals, manually or automatically, press the tree, then navigate into the pressed tree, and then `make`.
mod6: Will try to work that out.
mod6: Well, if it could be done without having to move anything that is already in place (as far as bitcoin is concerned), then it might work out alright.
mod6: It would probably need some tweaks in there ... hmm.
mod6: Actually, I forgot, trinque already did that part. Makefile.rotor exists.
mod6: good to hear :]
shinohai: Seriously mod6 it's coming along great, today's makefile build was so easy these Eulora noobs could probably do it.
mod6: so, im not 100% off the top of my head, but getting rid of the two rotor build script and integrating that portion into our makefiles ~might~ resolve at least part of the source redundancy issue.
mod6: oh yeah, we're not reall messing with any of that stuff. stator hasn't been a part of the 'orchastra' for some time now.
a111: Logged on 2016-09-09 00:28 mod6: oh, if you're saying that the rotor is no longer needed, i can agree with that. to an extent anyway.
mod6: the two .sh scripts you have in there can be integrated into the makefiles directly i think. however, we'll still need the openssl-004-musl-termios.patch and rotor_buildroot_dot_config
mod6: oh, if you're saying that the rotor is no longer needed, i can agree with that. to an extent anyway. ☟︎
mod6: When you come back, maybe we can discuss what you're trying to exactly say.
mod6: I can't make heads or tails out out of what you're saying there asciilifeform.
a111: Logged on 2016-09-08 23:46 mod6: It was my first hunch, during a pre-emptive go around with this to not place the makefiles in with current source base -- as pressed out via V.
asciilifeform: http://btcbase.org/log/2016-09-08#1537788 << at one time rotor was not the only builder, and so i cut'em off ☝︎
mod6: Altogether, since we have V, I like the idea of keeping things like makefiles and buildscripts out of the main source tree. One can get V, press the makefile project. Run a `make`, which will in-turn, press everything via V and then build with buildroot.
mod6: I think that keeping the Makefiles as a separate V tree would keep things a lot more clean.
mod6: Anyway, I bring this up because if we're determined to stick with this plan; the Makefiles included with the bitcoin source, then there is another ball of wax that needs to be discussed.
mod6: this is because the makefile process does it's own V press.
mod6: The problem is here, that when you run `make`, it'll build everthing under: build/rotor/TEST2/bitcoin/src despite the fact that the source is really under src/
mod6: When creating a makefile vpatch (for examplehttp://wotpaste.cascadianhacker.com/pastes/ebmj3/?raw=true), and then you press this out, you end up with something like this:
mod6: We discussed this, it was Mr. P.'s hunch that these should belong together (I hope I'm remembering this correctly), and I saw the wisdom in that.
mod6: It was my first hunch, during a pre-emptive go around with this to not place the makefiles in with current source base -- as pressed out via V. ☟︎
shinohai takes down the samovar to make tea for this discussion.
mod6: And in doing so, it kinda brings us back to a previous discussion.
mod6: We are pretty much at the point of starting V-ify these makefiles.
deedbot: http://trilema.com/2016/welcome-to-baluba-island/ << Trilema - Welcome to Baluba Island.