#asciilifeform | 2020-02-11

Reading Ch. 6 now asciilifeform ... nifty gnupg patch there.

~ 1 hours 50 minutes ~

03:15

shinohai: arguably that kludge is obsoleted by the pgpturd-eater in ch20

03:18

the ch6 gpg mutilation has no everyday use, it was to simply get at the #s for didactic example.

03:19

at the time i did not feel like writing a rfc4880-eater.

03:19

( and imho to include one in early ch would have been an epic distraction from the material )

03:21

shinohai: unrelatedly : sad to hear that you set bozo bit on qntra. imho qntra is 1 of the few good things going atm.

03:22

*

asciilifeform would write more often to qntra, but very rarely finds truly good material.

~ 8 hours 32 minutes ~

11:55

http://logs.nosuchlabs.com/log/asciilifeform/2020-02-11#1006189 <<< of course, I thought it clever hack so as not to write entire rfc4880-eater.

11:55

Logged on 2020-02-11 03:15:57 asciilifeform: shinohai: arguably that kludge is obsoleted by the pgpturd-eater in ch20

11:56

http://logs.nosuchlabs.com/log/asciilifeform/2020-02-11#1006193 <<< I agree Qntra is good thing, I think it better said why I just move on after reading BingoBoingo 's blog.

11:56

Logged on 2020-02-11 03:21:26 asciilifeform: shinohai: unrelatedly : sad to hear that you set bozo bit on qntra. imho qntra is 1 of the few good things going atm.

~ 28 minutes ~

12:25

./ticker btc usd

12:25

Current BTC price in USD: 9860.35

~ 36 minutes ~

13:02

*

adlai holds off bloviating further about the ring of nonnegatives before reviewing terminology

13:10

~the~ "ring of nonnegatives" is the multiplicative structure of the whole numbers, including zero; each prime number (2, 3, etc) can be multiplied arbitrarily to yield other nonnegative rings, each entirely contained within the outermost one.

13:13

my impression is that a serious attempt to generate broken moduli that seem to be a product of two similarly-sized primes would be performed using a bunch of similarly-sized primes, bunched around the cuberoot of the target modulus bitlength

13:14

the best way of choosing distinct triples from such a set of ~consecutive~ primes, and why they have to be consecutive, is left as an exercise for the wreckers

~ 15 minutes ~

13:30

zeder

Perhaps of passing interest: https://www.reddit.com/r/lisp/comments/f1r0yh/schematics_for_a_3600_or_ivory_cpu/

~ 1 hours 19 minutes ~

14:50

adlai, if you did that you lose the plausible deniability

14:51

you have to keep the prime distribution of the p's and q's close to that of a random number of such bitlength

14:56

Apocalyptic: ok, so hypothetical 2nd person singular wants plausible deniability of quite an expensive attack? much easier to optimize attack efficiency by punting plausible deniability to the legal dept., and doing both mathematics and arithmetic efficiently.

14:57

well my method is quite efficient too, you lose only a few seconds per key at most

14:58

the method being one using only a single high-entropy ring, or alternating between a small set of high-entropy rings from likely-nonconsecutive units?

14:59

(using only a single ring will quickly cause the phuctor approach to have a noisy day)

15:01

*

adlai sees no tangible value in any one of the three hypothetical weakenings; much effort expended over ... what exactly? if you want people using broken cryptosystems, why not have the system itself be a broken one, instead of trying to worm weakening keys into a stronger system?

~ 1 hours 24 minutes ~

16:26

adlai, Apocalyptic : currently aint clear to me that it is possible to produce a signature that'll verify over such 'soup' modulus.

16:27

*

adlai will need to review some of the literature before considering practical ramifications of such idiocies

16:27

http://logs.nosuchlabs.com/log/asciilifeform/2020-02-11#1006205 << sadly asciilifeform aint actually 'one of the most knowledgeable folks out there on Symbolics machines' and folx who show up looking for one will be very disappointed.

16:27

Logged on 2020-02-11 13:30:26 zeder: Perhaps of passing interest: https://www.reddit.com/r/lisp/comments/f1r0yh/schematics_for_a_3600_or_ivory_cpu/

16:28

probably there are still folx alive somewhere with a clue. problem is, they're 'silent like partizans'.

16:29

that being said, the 'cadr' and '3600' series machines are likely the ~last~ thing a would-be designer of 'lisp cpu' oughta be looking at;

16:30

'cadr' was designed to be buildable with the off-the-shelf discrete semiconductors ('bitslice') of its day, and as a consequence extremely baroque. 3600 inherited this gnarl.

↖ ↖

16:32

'ivory' series is imho the only interesting , even from archaeological pov, bolix product. but almost nothing is known about its internal design.

16:33

how did the ivory design sidestep that gnarl?

16:33

adlai: was an ab-initio design (incompatible w/ 36xx)

16:36

http://logs.nosuchlabs.com/log/asciilifeform/2020-02-11#1006214 << see e.g. this schoolbook review .

16:36

Logged on 2020-02-11 16:27:30 adlai: will need to review some of the literature before considering practical ramifications of such idiocies

16:40

in other lulz, usg Officially admits that it owns 'crypto ag' corp.

16:40

(trilema) 2016-07-07 asciilifeform: re earlier, i never quite grasped why every 'classy' diddled crypto peddler has to move to geneva. just because nsa-operated crypto AG corp. is there ?

16:44

lol read that earlier, was quite lulzy

16:44

looks like CADR and 3600 were as much products of 'junkyard wars' as the 'cardano' could've been, had the accelerator been properly glued to the floorboards

16:44

* coucou_bot (~coucou_bo@94.239.63.9) has joined #asciilifeform << which 1 of you was this ?

16:44

Not I

16:46

adlai: bolix could not have built 'ivory' in 1980, the necessary vlsiism did not exist. so inevitably went through junkyard wars stage.

16:47

( as did e.g. texas instruments, the ~third~ mit licensee that scarcely anyone talks about, given as their design was a straight CADR clone. but observe, ti corp had semiconductor fab. and still did not produce '1 chip machine'. because the necessary 'micrometrage' simply not yet existed. )

~ 56 minutes ~

17:44

feedbot

http://qntra.net/2020/02/usg-ownership-of-crypto-ag-and-use-of-the-firms-products-to-spy-on-allies-and-customers-released-to-the-public/ << Qntra -- USG Ownership Of "Crypto AG" And Use Of The Firm's Products To Spy On "Allies" And Customers Released To The Public

~ 34 minutes ~

18:18

"The Post was able to read all of the documents, but the source of the material insisted that only excerpts be published." hopefully someone leaks the whole archive somewhere

18:19

Apocalyptic: for that matter, e.g. the snowden thing still not leaked afaik.

18:20

yeah, for reasons quite not obvious to me

18:21

BingoBoingo

May be an intentional leak. Thing doesn't seem to be doing anything for them anymore. Might as well advertise the existence of the seekrit blackmail files before everyone in them dies of old age.

18:22

Apocalyptic: imho quite obvious. d00d leaked to 'carefully selected people' (i.e. never burned the usg outpost ~in own head~) instead of to warez. so naturally only 'selected' pieces made public.

18:23

asciilifeform, no I meant why none of those 'selected' people thought it might be a good idea to upload the whole bunch after a while

18:23

because they're loyal usg stooges ?

18:24

in that case, *snore*

18:24

cryptome.org only links to this site regarding Crypto AG as well

18:26

Apocalyptic: today's 'expose' is instance of what's known as 'controlled burn'. i.e. 'let the idjits think that crypto ag is dissolved, and the project is in the past'.

18:26

I see

18:27

whereas in fact massive and continuing success, simply not requires a swiss shill corp in any way. why, when you have 'openssl'ism etc

18:27

aha

18:27

if you don't need to sell physical products to select official dudes anymore, why bother

18:28

in point of fact the physical-products people 100% controlled to this day.

18:31

all of this of course distraction from the greatest success of usg : where people with a clue have been drive entirely out of the crypto field on ~both supply and demand sides~.

18:31

*driven

18:32

witness e.g. the total lack of otp products on commercial market. for more than half century.

18:32

"Two companies purchased most of Crypto’s assets. The first, CyOne Security, was created as part of a management buyout and now sells security systems exclusively to the Swiss government."

18:32

" CyOne has more substantial links to the now-dissolved Crypto, including that the the new company’s chief executive held the same position at Crypto for nearly two decades of CIA ownership."

18:32

they didn't even bother changing key people

18:35

Apocalyptic: why to change'em ?

18:35

crypto iron (and softs) is archaetypical 'lemon market'.

18:38

100% of the buying power belongs a) usg-controlled entities ('international finance', pseudosovereign vassal states, etc) and b) konsoomer, who can't tell own head from arse and will buy whatever is pictured in glossy magazine

18:40

(a) will knowingly buy strictly usg-compliant items. (b) will buy whatever is told to buy.

18:40

(trilema) 2016-12-15 asciilifeform: iirc we had a 'crypto-lordosis' thread

18:42

( for bonus lulz, (b) typically will 'buy' only what is given 'for free' and so generally has 0 impact on market offerings at all )

18:43

someone , somewhere, might buy the 'there is massive untapped market for honest crypto!111' notion, but asciilifeform -- after 4y of selling FG -- does not buy.

18:46

the ~only money in crypto market is the kind that bought 'crypto ag'.

18:47

asciilifeform, were FG sales so poor ?

18:48

Apocalyptic: 200 were made, and all eventually sold. but iirc all but 1 were bought by folx who were tuned into #t .

18:48

not quite 'massive built-up demand' picture.

18:49

indeed

18:49

the (a) folx continue to buy crypto-ag & co's $20,000 'quantum' black box (welded shut.) (b) continues to happily lap up e.g. intel's on-die rng.

18:52

notably absent is any sign of (c) -- buyers who give half a fuck re auditability / 'fits in head' and buy what actually worx, per own grasp.

18:53

even the #t folx, near as i can tell, bought fg 'on faith', there was 0, afaik, attempts to replicate, or meaningful commentary re the published src

18:56

fwiw this picture was not a surprise to asciilifeform .

18:58

'low-information konsoomer' always eventually gets the iron/softs he deserves. crypto aint any sort of special case.

19:01

try sell honest rng ? konsoomer: 'intel included RDRAND on die' . try sell proper inline disk cryptor? konsoomer: 'why should i buy 1000 $ box , samsung includes password feature'

19:02

try sell onetimepad phone ? konsoomer : 'why? i can buy 'redphone'' etc

19:03

nsa doesn't even need to send gasenwagen to pick up 'terrorist' designer.

19:03

'the market' does a+++ job of ensuring that nuffin but nsaware is available.

19:05

this is not even to mention the fact that 'honest' product has to be financed by crackpots, whereas 'professional'(tm)(r) is generously funded by unlimited printing press.

19:09

it's what it's there for anyway

19:09

which?

19:10

the printing press

19:10

a. naturally.

19:13

presently i suspect that you could drop e.g. fg outta airplanes, like propaganda brochure, and ~100% would end up as keychains.

19:22

http://logs.nosuchlabs.com/log/asciilifeform/2020-02-11#1006197 << i dungetit, shinohai , if good thing -- wainot write for it ?

19:22

Logged on 2020-02-11 11:56:55 shinohai: http://logs.nosuchlabs.com/log/asciilifeform/2020-02-11#1006193 <<< I agree Qntra is good thing, I think it better said why I just move on after reading BingoBoingo 's blog.

19:23

asciilifeform, to me that comment is quite clear. He fears it would damage BingoBoingo's standing in the "republic"

19:23

^

19:25

loox like BingoBoingo has ~2wks left in the mother church, as things stand

19:25

(trilema) 2020-01-25 BingoBoingo: Give me one month to try to shed the alfisms, and make a hard outreach push to break the 'Qntra looks a whole like like BingoBoingo's other blog' appearance. I'll include a day by day in the weekly recaps, and if material improvement isn't showing in a month... I'll pledge to #ossasepia or #trilema-hanbot if they'll have me in their castles.

19:27

after which a) shtrafbat b) out into the frost , his option

19:28

The Stockholm syndrome too strong there. He'll pledge to ask diana_coman for permission to write before long.

19:28

shinohai: would rather have the fella speak for self.

19:29

"I'll pledge to #ossasepia or #trilema-hanbot if they'll have me in their castles."

19:32

./google nature will castigate those who don't masticate

19:32

1. Horace Fletcher - Wikipedia

19:32

https://en.wikipedia.org/wiki/Horace_Fletcher

19:32

Horace Fletcher (August 10, 1849 – 1919) was an American food faddist who earned the nickname "The Great Masticator", by arguing that food should be chewed thoroughly until liquefied before swallowing: "Nature will castigate those who don't

19:32

masticate.

19:32

19:32

2. Nature Will Castigate Those Who Don't Masticate - Puppy Out ...

19:32

http://puppyoutofbreath.blogspot.com/2011/03/nature-will-castigate-those-who-dont.html

19:32

Mar 26, 2011,

19:32

Nature Will Castigate Those Who Don't Masticate. We called him Uncle Harry, but he wasn't really our uncle. Mom and Dad had us kids

19:32

use ...

19:32

19:32

oh hey it worx shinohai .

19:32

ddg worx too

19:32

./ddg nature will castigate those who don't masticate

19:32

1. Horace Fletcher - Wikipedia

19:32

https://en.wikipedia.org/wiki/Horace_Fletcher

19:32

Horace Fletcher (August 10, 1849 - 1919) was an American food faddist who earned the nickname "The Great Masticator", by arguing that food should be chewed thoroughly until liquefied before swallowing: "Nature will castigate those who don't masticate."He made elaborate justifications for his claim.

19:32

19:32

2. Chew Digest - Nature will castigate those who do not ...

19:32

https://chewdigest.com/

19:32

My goal is to help you better understand one of the simplest things that can be done to improve your overall health. Simply chew more. Even if you don't change what it is you are putting in your mouth, change what you do once it's in there and time will take care of the rest.

19:32

19:32

3. Chew on This: More Mastication Cuts Calorie Intake by 12 ...

19:32

https://blogs.scientificamerican.com/observations/chew-on-this-more-mastication-cuts-calorie-intake-by-12-percent/

19:32

Chew on This: More Mastication Cuts Calorie Intake by 12 Percent. ... "nature shall castigate those who don't masticate." ... 12 hours ago — Davide Castelvecchi and Nature magazine.

19:32

19:33

could be moar compact (wai blank lines?) but otherwise edible

19:35

shinohai: oughta put in ea. ln. the max fitting segment of the string

19:36