#asciilifeform | 2021-03-09

03:14

*

adlai has finally begun reading about [rateless] erasure codes, thanks to thread

03:14

Logged on 2021-02-21 10:32:03 trinque: verisimilitude: I'd like to see an implementation of this in Ada or CL, and haven't the time currently https://pdos.csail.mit.edu/~petar/papers/maymounkov-online.pdf

03:17

adlai

http://logs.nosuchlabs.com/log/asciilifeform/2021-02-14#1031588 << quite simply because I"m a bit sick of thinking about how to improve my HTTP site's structure, given that it's planned to host both 'blog', and mirror content, and possibly some dynamic content -- although by 'dynamic' I mean generated by the server at request-time, as opposed to client-side javascript.

03:17

snsabot

Logged on 2021-02-14 22:01:06 verisimilitude: That still doesn't answer why Gopher was mentioned, adlai.

03:19

adlai

I'd begun reading about Gopher after finding it mentioned in your site, although so far, I have not explored 'gopherspace' beyond verisimilitudes.net; thus, you serve as a de-facto ambassador of that protocol.

03:20

*

adlai understood from your response that this is similar to asking "tell me, Michael, what is it like to be black?"

03:22

adlai

as for the input methods thread: my 'training and expertise' with that other keyboard makes me not mind chording in the least, although I do prefer modal systems when they are possible

03:23

*

adlai doesn't see why a sufficiently configurable chording device couldn't have keys 'stick' for chords, effectively producing short-term time-dependent modal chording

03:24

adlai

what asciilifeform implemented in his javascript demo of the microwriter is cool; especially, the fact that synchronisation is only required at the release; however, ideally you'd never have to press more than one key simultaneously

↖

~ 5 hours 19 minutes ~

08:43

trinque

howdy adlai

08:44

trinque

$ticker btc usd

08:44

trinque

still dead. looks like 54.5k-ish today

08:44

trinque

so stimulated!

~ 25 minutes ~

09:10

03:14	*	adlai has finally begun reading about [rateless] erasure codes, thanks to thread
03:14	snsabot	Logged on 2021-02-21 10:32:03 trinque: verisimilitude: I'd like to see an implementation of this in Ada or CL, and haven't the time currently https://pdos.csail.mit.edu/~petar/papers/maymounkov-online.pdf
03:17	adlai	http://logs.nosuchlabs.com/log/asciilifeform/2021-02-14#1031588 << quite simply because I"m a bit sick of thinking about how to improve my HTTP site's structure, given that it's planned to host both 'blog', and mirror content, and possibly some dynamic content -- although by 'dynamic' I mean generated by the server at request-time, as opposed to client-side javascript.
03:17	snsabot	Logged on 2021-02-14 22:01:06 verisimilitude: That still doesn't answer why Gopher was mentioned, adlai.
03:19	adlai	I'd begun reading about Gopher after finding it mentioned in your site, although so far, I have not explored 'gopherspace' beyond verisimilitudes.net; thus, you serve as a de-facto ambassador of that protocol.
03:20	*	adlai understood from your response that this is similar to asking "tell me, Michael, what is it like to be black?"
03:22	adlai	as for the input methods thread: my 'training and expertise' with that other keyboard makes me not mind chording in the least, although I do prefer modal systems when they are possible
03:23	*	adlai doesn't see why a sufficiently configurable chording device couldn't have keys 'stick' for chords, effectively producing short-term time-dependent modal chording
03:24	adlai	what asciilifeform implemented in his javascript demo of the microwriter is cool; especially, the fact that synchronisation is only required at the release; however, ideally you'd never have to press more than one key simultaneously
		↖
		~ 5 hours 19 minutes ~
08:43	trinque	howdy adlai
08:44	trinque	$ticker btc usd
08:44	trinque	still dead. looks like 54.5k-ish today
08:44	trinque	so stimulated!
		~ 25 minutes ~
09:10	asciilifeform	wb adlai
09:11	asciilifeform	http://logs.nosuchlabs.com/log/asciilifeform/2021-03-09#1033004 << ftr i did not invent this -- it was a behaviour of the original machine being emulated
09:11	snsabot	Logged on 2021-03-09 03:24:42 adlai: what asciilifeform implemented in his javascript demo of the microwriter is cool; especially, the fact that synchronisation is only required at the release; however, ideally you'd never have to press more than one key simultaneously
		~ 42 minutes ~
09:54	adlai	right, you still copied the precise behavior, rather than some pantsuit "this is what I came to expect"
09:55	adlai	/away hola, alf et mic
09:57	*	adlai must confess: while sober - wastes time at solving trivial textbook problems of group theory, iteratively alpha-beta-ing mobile games, sieving reddit for interesting conversations [e.g. "what if you owned an acre, but no municipal garbage collection?"]
09:57	*	adlai must confess,,, drunk? irc!
09:58	adlai	my hesitation for writing the, honestly, rather trivial erasure coding, is that there is no specified prng
		↖
09:58	adlai	easy cop-out #1 - sha[2], of broadcaster address; vulnerable to "you got relayed, but we forgot to relay your original address"
		↖ ↖
09:59	*	adlai thought of that one after asking self, "why would anyone need this anyway? receiving bitcoin data?"
10:00	adlai	so, my thinking is that a sane erasure code assumes that prng params are part of the broadcaster address.
10:01	adlai	i.e., if you connect to silkroadvb5piz3r.onion with ketsak key /livebeef/[base64], yet site is broadcasting on /deadbeef/[base58], tough shit.
10:01	trinque	I think it just wants a uniform distribution of block assignments.
10:01	adlai	mno
10:01	trinque	don't know that any kind of real randomness matters that much
10:01	adlai	well, mya, that is the trivial version
10:01	trinque	the elusive mya!
10:01	adlai	if there is mutual understanding, you can save a factor
10:01	adlai	perhaps -- avoid idiocy, broadcast explicit inclusion set w/each block
10:02	*	adlai -> piss, then reread
10:04	trinque	and ftr, yes, trivial, and yet somehow still on IRC