tmsr - Search: i

34500+ entries in 0.267s

a111: Logged on 2018-05-25 12:31 spyked: but I knew how the rss bot design should look (I wrote up a draft before starting the implementation), the problem was rather fitting it up with all the "imported" elements.

mircea_popescu: no. write good code from the get go. there's no room here for "oh, this is the code i wrote, but it's not really who i am". it fucking is who you are, irrespective of what the monkey may be saying in your ear. THAT is what you are.

mircea_popescu: i expect 80% of the cases of "oh, i am being asked too much, this republic sucks" come from exactly this inept "oh, i won't do a good job, and wait for time to resolve it". yes, time will resolve it, but odds are the solution will hurt the butt.

mircea_popescu: ave1, i'm not advanced enough to make much of a teacher ; but commercial speech is a discipline unto itself. i suppose one approach is to keep track of what seduces you and why ; and others. eventually you end up with some kind of mental image.

trinque: if Mocky were here I'd ask him if he sees the problem yet ☟︎

asciilifeform: lobbes: lol! last i knew, pseudo-analogue clocks were 'retro-chic' and 'hip'... i guess no moar

a111: Logged on 2014-11-06 20:34 asciilifeform: thestringpuller: in practice - i am what i am.

asciilifeform: trinque: d00d seems to suffer from some bizarre compulsion, to leave the 'am what i am' clue

KernelKhaos: lol, podcasts, but no blog, and I avoid publishing most things.

a111: Logged on 2018-05-25 17:19 KernelKhaos: hmmm. I am not what I am. Probably nobody, till I'm properly keyed.

KernelKhaos: I wotpasted my public key.

KernelKhaos: hmmm. I am not what I am. Probably nobody, till I'm properly keyed. ☟︎

ave1: http://btcbase.org/log/2018-05-25#1818722, I'm in the last rounds (building gnat-musl using gnat-musl on aarch64). The last part is the slowest (I have one core for it). Also, for some reason the download sites are becoming unstable, I had to change one and it just failed on another. ☝︎

mircea_popescu: and i shall be back soonish!

mircea_popescu: http://btcbase.org/log/2018-05-25#1818596 << meanwhile i figured out what you meant. here's the problem : http://btcbase.org/log/2018-05-08#1811449 / http://www.thedrinkingrecord.com/2018/05/04/collected-gripes/ ($2 -- speaking of which hey bb, selectotron when?) or more in the general http://trilema.com/2009/omul-inutil-sau-pledoarie-pentru-protectie-sociala/ ☝︎☝︎

asciilifeform: ( possibly entirely harmless, or quickly lethal, depending on your orccraft, i suppose )

a111: Logged on 2018-05-23 18:46 phf: http://btcbase.org/log/2018-05-23#1817710 << yes, i seem to make these kind of homonym errors a lot actually, and i reread the sentence every time, but don't see it until the issue is pointed out.

asciilifeform: do i need to go on, or does this make sense.

asciilifeform: 1) https://archive.li/nLxLX << this is a chinese 100W sw piratetron. it eats ~150W of mains current, and is the size of a brick, and needs forced air etc. costs about 100 $ , just by itself ( i.e. without whatever gossiptron logic and signal synth you'd have. )

asciilifeform: i'ma illustrate with picture, the difference in equipment. :

asciilifeform: i'ma make a long thread , shorter, and summarize, it makes considerably moar sense to have smaller, cheaper, unattended 1-10W checks-rsa-and-relays boxen, then massive 'pirate' mast.

a111: Logged on 2018-05-25 01:07 mod6: baby, i've got a fever.

asciilifeform: BingoBoingo: i am considering here the tmsr use cases, rather than the old farts chatting about grandchildren use case (the latter already is quite well-documented, folx who are interested in it can simply rtfm)

asciilifeform: i suppose i oughta detail in the logs, in concrete example, for fyootoor ref.

phf: trinque: i like that idea, i'll see if it's easy to implement though. it doesn't map quite cleanly to current search architecture though, which is single pass and stateless ☟︎

asciilifeform: BingoBoingo: funnily enuff i was listening to web-sdr boxen in UY not long ago. there's several public ones.

phf: asciilifeform: i'm establishing scope, rather then actually doing anything. firefox on my x60 has an uper limit on tabs i have open, because it runs out of memory, so i started thinking that this ain't no way to live :> ☟︎

asciilifeform: phf: werentcha in the middle of a vtron or do i misremember tho

asciilifeform: trinque: i've wanted this for a long time; might be cpu-expensive tho.

asciilifeform: phf: i also considered looking for the key check thing in the stock bios and simply NOPing it out

asciilifeform: also interestingly, they published the src for their uboot, it is theoretically possible to scrub the crapola out and rebuild. however i have not succeeded in building it.

a111: Logged on 2018-05-25 01:07 mod6: baby, i've got a fever.

asciilifeform: i'd like to properly degooglize the box.

phf: i'm just repeating things for own understanding

asciilifeform: right, i thought i already said.

phf: and i guess you're trying to replace google's uboot with a built-from-src one, that doesn't also have rsa checks etc. ☟︎

asciilifeform: i'd like a normal, built-from-src rockchip uboot for this thing. and i worked over the pcb with magnifying glass and found a buncha debug contacts, some of which almost certainly gotta be spi rom. but -- currently no time to chase this.

phf: ah i see, yeah i figured it's a drop in replacement, but the uboot bit complicates the matter. uboot i take it is a boot loader, or the kernel also goes in there?

asciilifeform: i booted a heathen linux on the thing from the sd card, via the heathen ('press ctrl-d in 10 sec on boot or i format yer disk') nonsense, it ran.

asciilifeform: and yes i also will be replacing my x60 with it, if/when i get the time to finish this to a decent state

asciilifeform: phf: i did find the (quite gnarly) method to replace the built-in rsa keys with mine, and how to sign arbitrary kernels. what i did not do , is to build a proper non-googlistic uboot for it, and flash it in in place of old ( 'sapper errs once' , there doesn't seem to be any spi rom contacts brought out on the mainboard for a 2nd try if fails )

phf: asciilifeform: hey, have you had further progress with the chromebook? i'm thinking of replacing my x60 with it, or rather i want to try a portable cuntoo install, without nuking existing setup

asciilifeform: i'd really like a working arm64 gnat...

a111: Logged on 2018-05-16 13:43 ave1: BTW the aarch64 will build static but not by default, so that part I will check.

phf: yeah, privmsgs are silent, so you're always making assumptions about your name. i think i have to handle NICK messages from server to know when the bot is force renamed..

ave1: I was wondering if something was wrong with my client.

phf: back in the canonical log days, i'd patch the name in the logs, but since we're past that i think it can be left as is, unless there are objections as far as log authenticity. ☟︎

Mocky: i mean seems like deedbot + voice is already keeping randos from speaking here

spyked: the rationale is that there are a few O(n) operations involved (e.g. walking through the list of rss feeds) and I don't want every rando landing on Freenode to be able to abuse this (in particular for advertising sent via pm). let 'em get their own rss bot if they wanna.

a111: Logged on 2018-05-25 08:59 spyked: re b. I have a rfc in the works, should be ready for publishing at the beginning of next week. this is also why I haven't disclosed the bot's name and the channel where it operates at the moment, I want to limit access to relevant bot commands to L1 and L2 only.

spyked: ave1, yeah, querying deedbot is kinda what I had in mind (not sure if it can be avoided). something like

spyked: as far as I can see, this first writing pass exposes incorrect assumptions about e.g. data structures and how they're used. if this keeps happening, then it might be that the problem is larger than the mind can chew on in one take. ☟︎

ave1: BTW how would you do this "I want to limit access to relevant bot commands to L1 and L2 only" ? Check if the nick has voice or download the WOT or can deedbot be queried?

ave1: Also, yes, once you write it you may see how it should have been written. But I wonder if this is the first step in an infinite cycle.

ave1: I sometimes have it all tought out and then it completely fails on writing the first couple of lines.

spyked: so eventually I expect some of the rss bot components (e.g. command handling, self-voicing) to evolve into standalone scaffolding for #trilema bots that implements a decent subset of http://trilema.com/2016/trilema-bot-spec/ ☟︎

spyked: but I knew how the rss bot design should look (I wrote up a draft before starting the implementation), the problem was rather fitting it up with all the "imported" elements. ☟︎

spyked: ave1, the approach in http://btcbase.org/log/2017-12-30#1761284 seems to work here, in many cases I learned how the proper version of some piece of code needs to look after writing it. ☝︎

ave1: spyked, I've seen this happening many times (the code is done, but ugly, so I cannot publish yet). I geuss there is a general process behind this. At time (a) you need some code, at time (b) you have it but the code is "ugly". At time (c) the code still works and you have little incentive to clean it up. I know only two ways to get less "ugly" code, in the time between (a) and (b) write non "ugly" code. Or at time (c) publish the code and get shamed into s

spyked: now, the reason why this has been taking me so long is that I hoped I would publish the pieces as I went along. but this is harder than it looked, had to write what is now unreadable (other than by myself) prototype, then (point c. on my list) I'll have to rewrite/refactor and then publish. all this despite the fact that this is "known item", not FFA.

spyked: re b. I have a rfc in the works, should be ready for publishing at the beginning of next week. this is also why I haven't disclosed the bot's name and the channel where it operates at the moment, I want to limit access to relevant bot commands to L1 and L2 only. ☟︎

spyked: re a. it's occured to me a few days ago that the bot must maintain +v/-v state so that it doesn't needlessly attempt to talk in-chan before self-voicing. I intend this part to be separate from the rss bot itself and reusable.

a111: Logged on 2018-05-24 15:38 trinque: I'd be happy to add it, but spyked, how's your rss bot coming along?

spyked: http://btcbase.org/log/2018-05-24#1818005 <-- it's a few (2-3?) weeks away from deployability. full report: basic functionality is pretty stable now (cl-feed-parser library was throwing random errors and wasn't fittable-in-head; so I had to replace and rewrite a significant portion of the rss and atom parsing bits). I still need to implement a. self-voicing and b. an access control mechanism. ☝︎

ave1: http://btcbase.org/log/2018-05-24#1817983, yes something I need to learn. My current approach is too try something and see if that's it, but I wonder if suggestions or help can be beneficial in learning this? or should all of this be self effident like baking pastries http://btcbase.org/log/2016-06-23#1488189 ? I have no idea how to talk to businessmen or come even close to a make an effictive sales pitch. ☝︎☝︎

Mocky: my lisp/ai teacher's first and last names were anagrams, and i *never* believed that was his real name. super weird, long white hair. craig graci

trinque: I couldn't tell you 5% of my teacher's names

Mocky: can't think of name, but i can picture him, scruffy luddite

asciilifeform: i can't picture how else

mircea_popescu: Mocky, you know, i remember my college by teacher. how else ?

mircea_popescu: asciilifeform, to ask you what you asked me, "It is not clear why it needs cement poured. What, will escape ?" and to answer me what i answered you,

Mocky: when i took data structure class, mind was blown. "you can do all this stuff with a comp and you've been teaching me for loops?"

trinque: so I'm not bemoaning anything

trinque: I left after one semester.

trinque: I'd have loved to have gone to an actual university.

mircea_popescu: oyu know, i've yet to meet the zek who has somewthing ~other~ to say about his secondary education time than "i was an idiot"

Mocky: i was an idiot, got endebted elsewhere, took all the cs classes, and then quit to get a job programming

mircea_popescu: so i jsut re-read the set ; indeed it is grand. i love reading myself.

mircea_popescu: i had nfy u of m rejects anyone

Mocky: i'm going to bed

mircea_popescu: i was a very happy 13yo.

mircea_popescu: i sadly dribbled it out in pieces, which i now regret. anyway, http://trilema.com/the-next-generation http://trilema.com/2016/the-darkening/ http://trilema.com/2016/the-dog/ http://trilema.com/2016/the-chosen/

trinque: I find myself daydreaming that mircea_popescu world quite often

Mocky: trinque, i don't think so

trinque: I guess their fetish is more or less jars of eyes.

Mocky: i mean why be absolutist about it >> http://btcbase.org/log/2018-05-25#1818515 ☝︎

mircea_popescu: previous discussion re leverage with alf aside ; it's a wonder if they can make one man out of 1k of the neets tied together. i have my doubts. so it's not just "need leverage" on this side -- the empire graciously provides anti-leverage out of its own meat.

trinque: hmmm, I can see it. in the personal anecdote, I wasn't overflowing with loyalty for anyone either.

mircea_popescu: (and no, i don't mean "talk to kids ~anbout bedwetting~, what a stupid fucking subject. i mean take them out for a sundae and do something interesting or other, whatever the spur of the moment provides.)

mircea_popescu: but i'm not persuaded by the lying-beating coupling.

mircea_popescu: "african-americans" however are universally "curious about what'd it be like", and for the obvious reason : they're about as african as i am.

Mocky: no, i'm not familiar with clinical sadism

mircea_popescu: a lot less numerous* i mean.

mircea_popescu: Mocky, the problem with scotsman fallacy is that there's an unresolved duality. i don't propose that one's a) a scotsman for being born of scottish woman and AT THE SAME TIME b) a scotsman because not an irredeemable asshole. i'm just going with b, and that's logically sound.

mircea_popescu: and if you had, i'd just say "he never really was"

Mocky: i see that my theory does propose that, yet I've never seen a socialist cured

Mocky: i'm not seeing the partial seems like trying full integration

lobbes: In other "the logs had it first", I just investigated a barf I received on an emerge and exclaimed "who are these orcs@gentoo.org that are masking gcc < 5.4??". I must've missed this thread entirely >> http://btcbase.org/log/2017-12-30#1761183 ☝︎

Mocky: I've noticed lately that google has some results that it damn well doesn't want to give me... all the results say "missing search term" for what ever I searched for, even when I put a + infront. but it *has* my results and if i repeat the search term 3 or for times in the query eventually relents and shows me

Mocky: ah I see it