#trilema | 2019-05-29

04:38

good morning, #t! following up on the http://btcbase.org/log/2019-05-22#1915244 thread: I've started reviewing hunchentoot, the first step would be to figure out which code base to use as a starting point from the genesis.

04:38

a111

Logged on 2019-05-22 21:36 lobbes: http://www.thetarpit.org/posts/y05/090-tmsr-work-ii.html#selection-197.31-205.258 << I wager there's a good chance you'll publish a genesis of tbnl/hunchentoot before I eat through mod_lisp, but I agree: as pieces emerge, we can sync up, regrind as needed, etc.

04:38

spyked

*for the genesis

04:41

spyked

so far I've been looking at the project changelog and the patch history and the patches seem like a mixed bunch, with (perhaps) some useful things and breakage a la sslism. so before going further, I'd like to get some idea of what version of hunchentoot the lordship and the L2 are using

04:42

spyked

in particular, I saw trinque, phf and ben_vulpes have been using it in the past, so I'd appreciate any input you have on the matter

↖

04:47

spyked

and I guess I'd prefer starting from code already battle-tested by L1 (in the form of a tarball + ksum?) rather than shithub, and turning _that_ into a genesis. although I suspect I'll have to dig deeper into the heathenpits of git commits anyway.

04:51

spyked

http://btcbase.org/log/2019-05-22#1915245 <-- in other c coad, /me spent his last 2-3 weeks looking at the tcp stack implementation in linus' kernel. it is truly a fungus, macguyvered with duct tape and rubber bands, such that changing one line almost anywhere breaks shit all over the place.

↖ ↖

04:51

a111

Logged on 2019-05-22 21:36 lobbes: As it stands I have two full pages of hand-written notes with various c and apache-stack likbez, and that was just so I could understand up to line 152 of https://github.com/mbattyani/mod_lisp/blob/master/mod_lisp2.c (only 900 or so lines left to eat). I most likely will publish these notes as a blog post once all is said and done

04:54

spyked

http://btcbase.org/log/2019-05-22#1915226 <-- indeedly so.

04:54

a111

Logged on 2019-05-22 19:44 mp_en_viaje: spyked, some very sad http://trilema.com/2017/the-world-has-changed/ shit right there.

05:00

diana_coman

by now I suspect most code out there is the fungus sort as that's how things "naturally grow"

05:06

spyked

the weird part is that the linux tcp stack ~works~ for the most part. I imagine the maintainer of that particular subsystem must be a neckbeard with 20+y experience in tcp (because sure, it's not only the implementation, the protocol itself is a mountain of complexity)

↖

05:12

spyked

and then looking at e.g. https://elixir.bootlin.com/linux/v3.10/source/Documentation/CodingStyle , one can readily notice that half or maybe more of the tcp code doesn't meet that minimum set of criteria. and yet... it's there. so, I guess re. http://btcbase.org/log/2019-03-15#1902966 he was prolly dumb from the very beginning

05:12

a111

Logged on 2019-03-15 21:44 mircea_popescu: made slightly mroe interesting by it having been written before rather than after linus went dumb.

05:22

*

spyked bbl, cotton-bit-pickin

~ 4 hours 49 minutes ~

10:11

04:38	spyked	good morning, #t! following up on the http://btcbase.org/log/2019-05-22#1915244 thread: I've started reviewing hunchentoot, the first step would be to figure out which code base to use as a starting point from the genesis.
04:38	a111	Logged on 2019-05-22 21:36 lobbes: http://www.thetarpit.org/posts/y05/090-tmsr-work-ii.html#selection-197.31-205.258 << I wager there's a good chance you'll publish a genesis of tbnl/hunchentoot before I eat through mod_lisp, but I agree: as pieces emerge, we can sync up, regrind as needed, etc.
04:38	spyked	*for the genesis
04:41	spyked	so far I've been looking at the project changelog and the patch history and the patches seem like a mixed bunch, with (perhaps) some useful things and breakage a la sslism. so before going further, I'd like to get some idea of what version of hunchentoot the lordship and the L2 are using
04:42	spyked	in particular, I saw trinque, phf and ben_vulpes have been using it in the past, so I'd appreciate any input you have on the matter
		↖
04:47	spyked	and I guess I'd prefer starting from code already battle-tested by L1 (in the form of a tarball + ksum?) rather than shithub, and turning _that_ into a genesis. although I suspect I'll have to dig deeper into the heathenpits of git commits anyway.
04:51	spyked	http://btcbase.org/log/2019-05-22#1915245 <-- in other c coad, /me spent his last 2-3 weeks looking at the tcp stack implementation in linus' kernel. it is truly a fungus, macguyvered with duct tape and rubber bands, such that changing one line almost anywhere breaks shit all over the place.
		↖ ↖
04:51	a111	Logged on 2019-05-22 21:36 lobbes: As it stands I have two full pages of hand-written notes with various c and apache-stack likbez, and that was just so I could understand up to line 152 of https://github.com/mbattyani/mod_lisp/blob/master/mod_lisp2.c (only 900 or so lines left to eat). I most likely will publish these notes as a blog post once all is said and done
04:54	spyked	http://btcbase.org/log/2019-05-22#1915226 <-- indeedly so.
04:54	a111	Logged on 2019-05-22 19:44 mp_en_viaje: spyked, some very sad http://trilema.com/2017/the-world-has-changed/ shit right there.
05:00	diana_coman	by now I suspect most code out there is the fungus sort as that's how things "naturally grow"
05:06	spyked	the weird part is that the linux tcp stack ~works~ for the most part. I imagine the maintainer of that particular subsystem must be a neckbeard with 20+y experience in tcp (because sure, it's not only the implementation, the protocol itself is a mountain of complexity)
		↖
05:12	spyked	and then looking at e.g. https://elixir.bootlin.com/linux/v3.10/source/Documentation/CodingStyle , one can readily notice that half or maybe more of the tcp code doesn't meet that minimum set of criteria. and yet... it's there. so, I guess re. http://btcbase.org/log/2019-03-15#1902966 he was prolly dumb from the very beginning
05:12	a111	Logged on 2019-03-15 21:44 mircea_popescu: made slightly mroe interesting by it having been written before rather than after linus went dumb.
05:22	*	spyked bbl, cotton-bit-pickin
		~ 4 hours 49 minutes ~
10:11	asciilifeform	http://btcbase.org/log/2019-05-29#1916143 << for some values of 'works' , lol.
		↖
10:11	a111	Logged on 2019-05-29 09:06 spyked: the weird part is that the linux tcp stack ~works~ for the most part. I imagine the maintainer of that particular subsystem must be a neckbeard with 20+y experience in tcp (because sure, it's not only the implementation, the protocol itself is a mountain of complexity)
10:11	asciilifeform	the usual megatonne-of-c-liquishit style of 'works'.
		~ 41 minutes ~
10:53	spyked	asciilifeform, sure. it's surprising that it works "for some values" even, imho. to illustrate: http://archive.is/9Ssvh#selection-71637.0-71641.16 <-- saucerful of barf
		~ 20 minutes ~
11:13	asciilifeform	spyked: not only is the implementation what it is, but tcp per se is massive pile o'shit, where it aint even possible to implement it w/out 9000 tonnes of state machine gnarl
11:14	asciilifeform	i vaguely suspect that it may have been the original usg-'standards committee'-powered explicitly-organized nobus generator
11:18	asciilifeform	the protocol is 'fractally retarded' -- i.e. broken on absolutely erry possible level. starting from where it takes exactly 1 trivially forged packet to close someone's connection, to where 'allcomers' get a substantial chunk of memory allocated , and make ddos trivial , to where it forces 9000x moar complicated design of routing gear, to... could continue but why.
		↖
11:20	asciilifeform	see e.g. old thrd re subj .
11:20	a111	Logged on 2016-08-26 13:34 asciilifeform: tcp is evil, fundamentally because it violates the 'NEVER something-for-nothing-to-all-comers-FUCKOFFRANDOS' principle.
11:23	asciilifeform	... and here ; and elsewhere.
11:23	a111	Logged on 2017-03-14 14:31 asciilifeform: Framedragger: the problem with tcp isn't simply that enemy can insert an RST packet and make you blame your peer. (and whitelists do 0 against this.) but that it is very expensive , computationally, long before you have any idea who you're talking to.
11:27	asciilifeform	tcp shows erry possible sign of having been designed, from the start, to extend the ease of snoopage from traditional circuit-switched telco grid, to the packet world. consider e.g. the 'helpfully' plaintext sequence numbers.
		↖ ↖
11:28	asciilifeform	or how routers on erry hop of a connection are forced to maintain a 'circuit' state in memory, as if they were telco switches
11:37	BingoBoingo	And routers on network borders need the entire BGP table in RAM
		~ 47 minutes ~
12:24	asciilifeform	meanwhile, in alert readers, http://www.loper-os.org/?p=2118&cpage=1#comment-19800