phf: i had a similar reaction to dark shadows, what a disappointment.

phf: my concern is post-trump their going to churn out stuff like dunkirk, men films about men bonding for men that are menly

phf: % ./vdiff -u a b

phf: --- a 595beeb13e543b7107c182bb23ac7766bd62ab3122e250e90573eb540f89520122150e5f69bb99140bc47ab526c9b7b2a3b2402b09b44b7391501a149876765c

phf: +++ b b91e1bccfe010e28e544ebc1cdc2e6030f88a899a13c78d521b46d19801a065aec3c81bae6a0238fd6a01d22c314888a0c483c9c74384aee1e4f288e76ff57b8

phf: @@ -1,2 +1,2 @@

phf: hello

phf: -world

phf: +doom

phf: an all-c vdiff

phf: correct

phf: well, now comes the hard part :)

phf: i wonder if we're just witnessing some sort of criticality point, where everything is starting to rapidly deteriorate under the weight of maggot work

phf: asciilifeform: one property of the new vdiff already is that ---- situations work, e.g. http://p.bvulpes.com/pastes/0Niwr/?raw=true

phf: asciilifeform: tbl is venerable, part of grey beard tools, that nobody uses, like eqn or pic

phf: LFS

phf: http://www.linuxfromscratch.org/, i'm being a smartass

phf: hah. well then

phf: to be fair soelim is part of troff suite going back to bill joy's original. i suspect zsoelim is some recent groff addition.

phf: asciilifeform: mostly because it's not a phototypesetter driver

phf: mircea_popescu: i'm not sure what to say, we're like multiple cycles deep into rhetoric here, inception style. i will say though, all this misdirected hate can't be good for digestion, i fear for asciilifeform, he might crack under questioning

phf: mircea_popescu: but the factuals are not necessarily under dispute (~soelim~ is not a phototypesetter is what i said). original hate was against "never seen this wrecker thing before", then when new fangled thing turned out to be an old tool, hate shifted to 1960 phototypesetters. methinks the point is a traveling one..

phf: what trinque said, there's no "straight latex", it's 1.6gb of liquishit

phf: i've actually recently went through an exercise of trying to spin up knuth's plain tex as a standalone typesetting system. it's nearly impossible

phf: mircea_popescu: well, ~i've~ found something to hate, so i joined the fray

phf: mircea_popescu: well they've believed in the whole "single tool for a job" thing back then, and yes it was a hack (not in a sense that soelim is such a tool, but in a sense that tools that soelim is a glue between poorly designed tools)

phf: plan9 rewrote their soelim as a shell script..

phf: right, traditional games, like throwing discus and wrestling

phf: that's evening program

phf: that explains why when i tried building it it was chasing compatibility bugs down a rabbit hole

phf: "then told me that I had to abandon my cleanup work and start over on his tree, and explain everything to his satisfaction (as a Windows guy) before it could go in"

phf: But these days, my complaint is that I have no confidence whatsoever in

phf: tinycc's maintainership. It has the tinycc.org domain, and Fabrice

phf: handed over the project, so it is the official final resting place of

phf: tcc. But it's still stagnant, because Fabrice put a Windows developer

phf: in charge of the project, one who apparently does not understand open

phf: source development in the slightest. He's putting out a windows-only

phf: version of tcc as far as I can tell, one which will never build an

phf: unmodified Linux kernel (has made zero progress on this front in the

phf: past _THREE_YEARS_), thus it cannot ever act as (even an infereior) gcc

phf: replacement.

phf: i think he doesn't like that they keep poaching his changes back into official cvs

phf: selectively

phf: looking at their commit history there's a lot of "utf8 support in ..." and "nls ..." which is probably the proverbial fleas bringing dog

phf: asciilifeform: you know after staring at a lot of bad c code and last two days worth of conversations, i don't think there's much wrong with sbcl, but then i haven't looked at sbcl code in about a year at this point. i think i was mostly objecting to overall trajectory of lisp ecosystem

phf: but specifically it was small annoyances with "modern" enforcements. like style warnings, or my personal pet peeve, the fact that you can't shadow locked packages without having to unlock

phf: the awfully pedantic defconstant behavior (which sbcl specific, and which requires packages like alexandria to have asinine define-constant, which for all practical purposes is what defconstant is supposed to be)

phf: the fact that character type is not dynamic (which to be fair is the property of all free lisps for some reason), so if you're dealing with text, you're forced into 32byte per character nonsense ☟︎

phf: the fact that macro evaluator inlines both compiled and non-compiled macros, which means that you have to manual track macro dependency and tediously reevaluate even when working in an interactive environment ☟︎

phf: i think naggum has a rant about the last one

phf: none of these are properties of sbcl past certain vintage though though, sbcl is already a modernization of common lisp

phf: cmucl fwiw was designed like a lisp machine (though it had damage done to it by modernizers already), where the evaluator + vops was you primary interaction mode, and compilation was a way to evaluate a piece of code to a vop like status

phf: and vops in turn were an equivalent of a lisp machine microcode

phf: you're not going to even approach a performance of a compiled sbcl. at best you would do is non-jited lua

phf: obviously a typo

phf: you kind of have to these days yes

phf: that's a very broad statement

phf: trinque: ccl used to do it that way actually, interested parties might want to poach that code. they call the concept Rune and it's basically a way to support orcograms in a 8bit lisp

phf: maybe he's talking about pascal strings, it's hard to say at this point

phf: https://blog.cryptographyengineering.com/2017/12/19/the-strange-story-of-extended-random/ in unrelated

phf: naggum has a rant about it specifically

phf: heh, while looking for that rant "I have designed and implemented one for my own needs, but I find the number of disgusting losers who would benefit from it if I published my code to be too high."

phf: source reading can be a preprocessor stage (which is a lot saner to do in common lisp than elsewhere), this is also how traditional tex handles orclangs, before xetex and luatex and such. special ascii sequences to represent local lang glyphs and if you don't want to write those by hand, you use (or write) a tool that takes a local encoded document and translates it into ascii

phf: fwiw bulk of these tools have been written through the 90s and what was worthwhile from orcland was published that way then. until silent takeover by latex & 1.8gb tex installations happened and none of these tricks work anymore (because the necessary hooks are so deep within the layer of cruft it's near impossible to get to them, and one way they did it is through standard file system lay out, that requires a chain of compilation stages to move files

phf: from location A to location B to location C where they are expected)

phf: there's not a single article on the subject of compilation, but here's the complete thread https://www.xach.com/naggum/articles/search?q=recompile+reeval&sort=of specific test that he proposes is in https://www.xach.com/naggum/articles/3243682048034314@naggum.no.html and the "dun work" reaction https://www.xach.com/naggum/articles/3243737082731156@naggum.no.html

phf: there's two sbcl apologists further in thread, one of them saying "I don't mind provocative /per se/, but what you were saying gives the a| impression that SBCL is willfully bad, as opposed to in development. But lumping it in with willfully noncompliant systems for this reason, | given it's version number, is inappropriate." and the other one is a core dev saying that they might add it. of course the expected behavior is still not there

phf: even though i hear that sbcl now has a evaluator added back?

phf: etc etc etc

phf: yeah, bill newman had to rip it out when he was doing the original bootstrapping work

phf: i mean guy says so himself "Specifically, you never really get absolute proof. There’s always some innocent or coincidental explanation that could sort of fit the evidence — maybe it was all a stupid mistake."

phf: oblig. ptacek link

phf: mp is quite dangerous with them bash scripts

phf: i think a proper cuntoo doesn't even need x11, bash scripts, lynx and framebuffer to render images if need be

phf: eh it's all irrelevant to substance anyway, since bulk of computer activity amongst programmers is, to badly quote logs from memory, getting everything ready to meet girls by doing some misplacing activity somewhere where girls will never be

phf: that is to say, that mp machine can really be anything, and it'll be used directly, rest of us will spend next year fucking around with dlls