#ossasepia Logs for 09 Sep 2019



April 20th, 2020 by Diana Coman
diana_coman: lobbes: mind getting auctionbot in here too? [03:11]
diana_coman: shrysr: iirc my map of Canada, Alberta&Calgary are closer to British Columbia/Vancouver than to Toronto on the other side; is this correct? Have you considered BC/Vancouver too since you are looking countrywide anyway? [03:17]
diana_coman: and at any rate, ideally you'd want some remote work in fact. [03:18]
diana_coman: I have to admit that reading those job descriptions (the actual ones) triggers my alergies instantly; it seems to me though that you should focus on "junior analyst/programmer" + target the application to each place you send, at least (though yes, I'd discard some just for being too toxic environments really) [03:28]
diana_coman: re cv help, perhaps nicoleci would be kind enough to give you some feedback as she worked as head of HR in the USA so she knows this sort of shit way better than me. [03:29]
diana_coman: shrysr: btw, there's this Python job offer right here, as an idea re "alternatives": http://logs.ossasepia.com/log/trilema/2019-09-07#1934605 [04:02]
ossabot: Logged on 2019-09-07 05:13:01 mircea_popescu: so : who would like to do a spot of work for hire for me ? the ideea is to write and publish as a vpatch a stan logbot extension which a) processes search, through talking to a [presumably present] mysql server, and spitting out the results (formatted as in http://trilema.com/2019/ [04:02]
diana_coman: it's not clear to me that you are ready for this particular offer but you should at least think it through. [04:03]
lobbes: !Xping [11:02]
auctionbot: pong [11:02]
lobbes: diana_coman: it is in, but I do not have it set to list auctions [11:02]
lobbes: you okay if I do? I'll just need to add it to the cron job [11:03]
lobbes: will bblx01 [11:05]
shrysr: diana_coman: yes, BC/Vancouver are closer than toronto on the other side. 12 hr bus ride from calgary to vancouver. ~3 hr flight. vancouver generally equivalent to toronto overall, i guess (+ loose internet feedback). good number of IT/tech companies from job postings (less than toronto iirc, but still likely higher than calgary). THe plan was to look at those in detail after calgary (based on proximity), [11:13]
shrysr: but yes, I've actually been thinking its worth a quick foray to expand the list to consider vancouver. fwiw: in my first jobless year - after a point I made abs no distinction in geography, if there was a vacancy - i applied. [11:13]
diana_coman: http://ossasepia.com/2020/04/20/ossasepia-logs-for-09-Sep-2019#1001675 - yes, please do. [11:15]
ossabot: Logged on 2019-09-09 11:03:02 lobbes: you okay if I do? I'll just need to add it to the cron job [11:15]
diana_coman: shrysr: what sort of jobs did you apply for in that first year anyway? [11:16]
shrysr: http://logs.nosuchlabs.com/log/ossasepia/2019-09-09#1001672 this sounds quite interesting. afaik : sql and db fluency is quite important for data analyst/scientist whateva, and it sounds like text analysis is involved here, plus the html-like output, which are alll good. My general sense is that I was making progress in V, needed to brush up python at a point (i did do that well) - and i think that if I [11:22]
snsabot: Logged on 2019-09-09 03:47:30 diana_coman: it's not clear to me that you are ready for this particular offer but you should at least think it through. [11:22]
shrysr: focus only on that (and python) for a bit - i can overcome my learning epsilon-threshold on the subject - and shd be able to breathe and take in faster. It just so happened that I found Matt dancho and his R/Data scientist related courses are the best I've seen - plus for general data cleaning, i believe (from very small exposure) and also from feedback from guys (who know their stuff, atleast a lot more [11:22]
shrysr: than me), R is just easier than python (pandas).... since anyway most of the job is cleaning, and I just 'got R'... i plunged into R. fwiw : after gaining proficiency in R, it is actually not so bad using python. Its just that i need to do it for a while and consolidate the pieces in my head. [11:22]
shrysr: http://logs.nosuchlabs.com/log/ossasepia/2019-09-09#1001682 towards the end about 90\% mech engg stuff (10\% data hag :P). design engg, application engineer (like a sales engg or whateva). Even the dreaded project mgmt roles (the roles which have WAY too much P, and very superficial knowledge gain), eveyrthing in between. In reality - the distinction is that - there are different kinds of design. i.e there [11:33]
snsabot: Logged on 2019-09-09 11:01:00 diana_coman: shrysr: what sort of jobs did you apply for in that first year anyway? [11:33]
shrysr: are guys who just make technical drawings or 3D models. Say - making a drawing of a pump, but not needing to knowing the physics inside, but more abt manufacturing constraints.... and doing things as told. I fall into the 'advanced analysis' where I can do a bit of technical drawings (i learnt that bit only to fulfill my 'complete exposure' thing) - but i'm doing design calculations, and telling these guys [11:33]
shrysr: how it shd be or whateva. The distinction is somewhat complex in india - you call such design engg 'drafters' in india, and they dont get paid terribly much... here, predominantly - the design engg roles are drafter roles, and overall - it appears very few companies (perhaps only 2 handsful) do any great analysis. Its of course quite different in the USA. [11:33]
shrysr: in a way - the distinction is kindda clear out here in many ways.... (still there are pros and cons) For ex, application engg (technical sales) roles - which is not bad, but still potentially lot of (P) are few in number, and much much more sales oriented than technical... I guess that is a mistake I did, but by definition - theres supposed to be a distinction between application engg and a fucking sales [11:50]
shrysr: guy! probably needless to say - i can do sales, and probably understand the needs of a customer and relate it well to cost/our product - but its a lot of.... lying. Then its like convincing yourself aabt degrees of lying, or half truths or whateva. Anyway - while all this is in my head - i figured I cd get in and then find my way around, as usual. Before migrating - while i did apply to many jobs from [11:50]
shrysr: india itself - i guess I still had the enthusiasm of 'finding' my way around inside companies. After all the so called thinking... i came to the conclusion - it was stupid fundamentally. I can't keep switching companies or departments in search of better exposure, because in general they dont seem to like it or see the value in a wide exposure/learning capability anyway! i cant learn the way I want to and [11:50]
shrysr: fight the industry...(or anything outside myself perhaps) If knowledge/skill is atleast one cornerstone of power - then i need a field where I *can* learn and *can* come up with work reflecting that... and YES- remote work. Data analysis/even ML (yes, ppl offer ML lol on fiver iirc) is reasonably conducive to this. its another reason. Overall - my thinking is that in most cases ppl view my [11:50]
shrysr: experience/communication/results as good n nice n sometimes even impressive - and then find 5X more experienced ppl as safer choices. [11:50]
diana_coman: http://ossasepia.com/2020/04/20/ossasepia-logs-for-09-Sep-2019#1001684 - myeah, but so far focus doesn't seem to stay all that long in one place with you. [11:52]
ossabot: Logged on 2019-09-09 11:22:30 shrysr: focus only on that (and python) for a bit - i can overcome my learning epsilon-threshold on the subject - and shd be able to breathe and take in faster. It just so happened that I found Matt dancho and his R/Data scientist related courses are the best I've seen - plus for general data cleaning, i believe (from very small exposure) and also from feedback from guys (who know their stuff, atleast a lot more [11:52]
diana_coman: http://ossasepia.com/2020/04/20/ossasepia-logs-for-09-Sep-2019#1001694 - are you sure it's the 5X experience solely? [12:03]
ossabot: Logged on 2019-09-09 11:50:47 shrysr: experience/communication/results as good n nice n sometimes even impressive - and then find 5X more experienced ppl as safer choices. [12:03]
diana_coman: I have no idea re Canada but at a quickest of looks around here, it certainly seems that there are more jobs than people willing to take them on; still, I can't see any non-tmsr-job as something helping you much esp when actually doing the calculation re whether they pay you or it's in fact the opposite [12:05]
diana_coman: http://ossasepia.com/2020/04/20/ossasepia-logs-for-09-Sep-2019#1001691 - fwiw this is ~true, yes. [12:05]
ossabot: Logged on 2019-09-09 11:50:41 shrysr: guy! probably needless to say - i can do sales, and probably understand the needs of a customer and relate it well to cost/our product - but its a lot of.... lying. Then its like convincing yourself aabt degrees of lying, or half truths or whateva. Anyway - while all this is in my head - i figured I cd get in and then find my way around, as usual. Before migrating - while i did apply to many jobs from [12:05]
diana_coman: ^ meant re usual "sales" of shit is lying; by necessity even. [12:09]
shrysr: I see why you say that re diluted or sproadic focus....it has been true of late, particularly in the last year. For abt 5.5 years - i lived and breathed only CFD. but in this case, or atleast in the phase I've been in chan - it is really that I'm being pulled in too many directions. use R for erp... learn R for ML...so called portfolio project... so called 'transformation' from mech engg to 'the analyst [12:11]
shrysr: salt-mines want/need!' (shd be a commercial eh) the tension and borderline panic of getting out of hell hole... increasing with Every interaction with orcs here. You know- i did mention, there was a phase when I was trying to just .... accept - its fine here, i am all wrong abt everything. the reality is that i work in a place where i've literally seen ppl fired on the spot! for seriously crazy reasons! [12:11]
shrysr: whims, having a bad day - looking around for target and meandered in path like unsuspecting deer!! despite being very good at their jobs... all the more worse - no notice period, or even a notion of 'transfering knowledge/current work'. where less than 10\% employees are salaried (i am)... and apparently viewed Purely as liability/cost, while called 'family'! LOL. nowadays i can laugh at it. [12:12]
diana_coman: myeah, utterly poor deal you got yourself there; but the mind-boggling thing is that you seem bent on getting in another similar arrangement on dubious hopes that "they can't be as bad" and somehow they will be more committed to you. [12:17]
diana_coman: shrysr: for that matter - don't you have any holidays you can take or what? [12:25]
shrysr: lol. c'mon the above description is definitely not how it is in a lot of places! Even at ownership type companies (esp at lower levels).... i had 2 conversations last week. One with a human, and one with black orc, after ppt. the human happens to be a design engg, drafter whatever - older, tasked now to support me... and help review n stuff as required. The human, a local canadian for practical [12:25]
shrysr: considerations immediately told me of her own accord (no prompting from me) - yes... its..uhhh... quite different here....it is run uhhhh... quite different, and very hard to get used to. She comes here 3 days a week, on a contract. One Must wonder - why wd one come here, if they had any choice? I intend to ask her if i get a chance. [12:26]
diana_coman: http://ossasepia.com/2020/04/20/ossasepia-logs-for-09-Sep-2019#1001708 - superficially I'm sure it's not, indeed; and you probably get there to see the "naked" non-pretense thing; in the rest of places, for as long as there are still more resources around, the pretense is better kept and nicely packaged; and many - otherwise well intended - buy it in, too; but the core remains - the employer (esp western style) has no real incentive to any commitment, [12:32]
ossabot: Logged on 2019-09-09 12:25:58 shrysr: lol. c'mon the above description is definitely not how it is in a lot of places! Even at ownership type companies (esp at lower levels).... i had 2 conversations last week. One with a human, and one with black orc, after ppt. the human happens to be a design engg, drafter whatever - older, tasked now to support me... and help review n stuff as required. The human, a local canadian for practical [12:32]
diana_coman: esp if you are to still have any energy left at the end of the day. [12:32]
shrysr: i'm inclined to agree abt western style. This is not prevalent in india at all.. (yes there are other factors going into allowing that to happen... foreign masters, gen lower pay (esp v/s inflation (to the minor extent i know that topic)... overall culture, and a Very clear segregation of what 'educated' work entails v/s non-educated labor) ok i completely agree at the core it Is a pretense everywhere in [12:46]
shrysr: varying degrees..... in one way you may say atleast hez not hiding his so called feelings and fires ppl instead of fucking them up slowwwwllyyyy, (maybe he cant afford to do so). ha! but unfortunately - for subsistence - some mine is required... till i find my way in tmsr, and esp improve my coding skills. a better pretense could also mean - a place i can put my head down and work, without being affected [12:46]
shrysr: by black orcs. In fact, thats why i thought, ok - why not do that here...and failed. [12:46]
shrysr: fwiw: failed twice i guess. once before tmsr and once after. [12:48]
diana_coman: hm. http://ossasepia.com/2020/04/20/ossasepia-logs-for-09-Sep-2019#1001707 [12:57]
ossabot: Logged on 2019-09-09 12:25:24 diana_coman: shrysr: for that matter - don't you have any holidays you can take or what? [12:57]
shrysr: brb 10 min. [13:00]
diana_coman: re http://ossasepia.com/2020/04/20/ossasepia-logs-for-09-Sep-2019#1001682 lobbes' structured account of it is a good example of attempting to figure stuff out so you can ask well-formed questions http://blog.lobbesblog.com/2019/09/the-mp-wp-bot-job-offer/ [13:03]
ossabot: Logged on 2019-09-09 11:22:28 shrysr: http://logs.nosuchlabs.com/log/ossasepia/2019-09-09#1001672 this sounds quite interesting. afaik : sql and db fluency is quite important for data analyst/scientist whateva, and it sounds like text analysis is involved here, plus the html-like output, which are alll good. My general sense is that I was making progress in V, needed to brush up python at a point (i did do that well) - and i think that if I [13:03]
snsabot: Logged on 2019-09-09 03:47:30 diana_coman: it's not clear to me that you are ready for this particular offer but you should at least think it through. [13:03]
shrysr: in fact - i cant say i was ignorant abt the western style or did not think abt that before migrating. there were several occassions I'd be working with a d00d in the USA, and next day hes gone! poof! not a hint. I took it as a fact of life...that wd not be insurmountable. early on after migrating in fact - during jobless phase - thats why i did atleast some research on remote gigs... the idea was to [13:19]
shrysr: continue that even after getting work and atleast make pocket money to begin with and slowly only contract jobs .... earn in $, gain good exp - go back home to leverage. For that matter - that still had flaws.... business approach quite different n etc. [13:19]
shrysr: re: holidays i get 2 weeks of paid leave in a year i believe.... took none the first year. my parents are gonna be in new jersey over winter and i was to visit for 2 weeks.... supposed to start looking into tickets now. Get to meet my new newphew who is due in some weeks. The Great Hope was that I'd be outta here longggg before... i.e its still in a fluid state atm, in the sense - getting out is first [13:26]
shrysr: priority. [13:26]
diana_coman: I see; and according to you, it might anyway be that you get fired the moment you take any leave, be it paid or unpaid, lolz. [13:29]
diana_coman: shrysr: don't you actually need a visa for the US? [13:30]
diana_coman: newphew is great word btw [13:31]
shrysr: lol nephew* TBH typo. I have a B1/B2 visa thanks to my prev salt mine and visit to oklahoma... still valid for some more years. technically i can visit on business. [13:32]
shrysr: re fired. possibly yes.. lol. I will admit as i said - they are growing suspicions... fwiw: the volatility also means that may not happen.. atleast for me -- it was kindda liberating to just accept that its significantly more likely than unlikely (unless business starts roaring... fwiw: thats supposed to happen from sep... must own crystal ball i havent heard of)... liberating because I cd let go and [13:41]
shrysr: really plunge into getting out. also liberating because of you i guess... theres suddenly hope that if i dont give up - and dont allow my self to be distracted.. or go wayward due to stupid confusions - i can find a stable place and construct a realistic exit from slavery in time... and not ... hide anymore. [13:41]
diana_coman: shrysr: so come and visit Reading, have a talk face to face at least, then. [13:43]
shrysr: hmm.. UK > iirc need visa? or was that only for work... need to check if i made notes on that. iirc citizen of canada needs no visa to visit, idk if same for PR like me. [13:45]
diana_coman: shrysr: that's for you to check indeed, it's your papers anyway. [13:47]
diana_coman: shrysr: can you apply for canadian citizenship? [13:48]
shrysr: http://logs.nosuchlabs.com/log/ossasepia/2019-09-07#1001518 may i ask why? as in why you barely spoke a few words/day for months? [13:48]
snsabot: Logged on 2019-09-07 14:18:54 diana_coman: re talking, if it's any help, I probably get it more than you think (I still remember full months when I barely spoke a few words per day); but for this very reason I can tell you now that you should still talk. [13:48]
shrysr: http://logs.nosuchlabs.com/log/ossasepia/2019-09-09#1001669 how shd I get in touch? private ping / email ? [13:50]
snsabot: Logged on 2019-09-09 03:14:03 diana_coman: re cv help, perhaps nicoleci would be kind enough to give you some feedback as she worked as head of HR in the USA so she knows this sort of shit way better than me. [13:50]
shrysr: http://logs.nosuchlabs.com/log/ossasepia/2019-09-09#1001737 yes. Need to have been here for i think 2-3 years... forget the exact number of days... theres a test (canadian history culture etc). Wd have to give up citizenship of india.... but i have nt gone into details abt this. [13:56]
snsabot: Logged on 2019-09-09 13:32:55 diana_coman: shrysr: can you apply for canadian citizenship? [13:56]
diana_coman: shrysr: at the time I didn't have a clear "why"; what I can find looking back is mainly http://ossasepia.com/2019/08/25/the-new-old-vilnius-of-2019/?b=Perhaps&e=#select (though there's ofc nothing online from *that* time). [13:59]
diana_coman: you can politely ask nicoleci in #trilema if she'd help you [13:59]
diana_coman: shrysr: ugh, that sounds lousy; why give up indian citizenship? [14:00]
diana_coman: will bblx01 [14:00]
shrysr: well idk - my cousinz in toronto... hez the reference i meant abt the 100k ... he is a citizen and said it was necessary as india do not allow dual citizenship or something. Wd ofc need to evaluate for myself to kno exact reason. [14:02]
asciilifeform: http://logs.nosuchlabs.com/log/ossasepia/2019-09-09#1001683 << shrysr , i'ma not ask again, but must ask, what are you stuck on in v.py ? it's a 400ln proggy, and i recall no case where anyone needed 2months+ of sweat to understand it [14:08]
snsabot: Logged on 2019-09-09 11:06:52 shrysr: http://logs.nosuchlabs.com/log/ossasepia/2019-09-09#1001672 this sounds quite interesting. afaik : sql and db fluency is quite important for data analyst/scientist whateva, and it sounds like text analysis is involved here, plus the html-like output, which are alll good. My general sense is that I was making progress in V, needed to brush up python at a point (i did do that well) - and i think that if I [14:08]
asciilifeform: shrysr: i wrote it, and am willing to answer whatever q re how it worx [14:09]
asciilifeform: shrysr: ftr it took barely 2days to write [14:09]
asciilifeform: shrysr: did you read ben's walkthrough ? it did not help ? [14:11]
shrysr: asciilifeform: yes, thank you and I do appreciate you asking again... and I apologise if I have not given you an explanation. Its really not that I am not taking you up on it -i wanted to understand hashing and get atleast of the major components behind security, and then move forward. i am in fact following ben's intro, and was at the point of vdiff, wherein i wanted to understand what SHA512 did to a [14:19]
shrysr: better level than intuition, and signatures, which is prolly the point where you shared the handbook. fwiw: the discussions in the past week are about exactly the fact that - I am torn in multiple directions and actually havent spent anymore close to 2 months sweating V alone, and I am trying to resolve other things so that I *can* focus much better on this. [14:19]
asciilifeform: shrysr: if you expect to 'understand hashing' the way you understand e.g. ohm's law, will be waiting a long time. hashing is not a physical law concept , but cryptological -- i.e. authors of hash algoes ~believe~ that the operation where input turns into a corresponding garbage is difficult to reverse, but there is no proof aside from 'no one publicly showed how to reverse in N years'. [14:21]
asciilifeform: shrysr: hash algos are built with 2 objectives -- 'difficult to find x when given H(x)' and 'difficult to find y, y != x, where H(y) == H(x)' . [14:23]
asciilifeform: shrysr: all traditional hash algos try to achieve this with a series of arithmetic operations, arranged in such a way that a small change in x (e.g. 1 bit) will produce large change in H(x) (i.e. affecting all or most of the bits of H(x).) [14:26]
asciilifeform: shrysr: my point is that if you are looking for a proof that 'is in fact difficult', will probably spend the rest of your life looking. there aint any such proof, for any of'em. [14:26]
asciilifeform: shrysr: similarly, in e.g. rsa, ~it is thought~ that to get primes p, q out of a large N == p * q , is expensive. there is, similarly, no known proof . [14:28]
asciilifeform: ( in fact sometimes it is very cheap, if the operator who made N was careless . ) [14:29]
asciilifeform: shrysr: in order to grasp 'v', it is necessary to first postulate that hash and rsa work . (~then~, later, will be very obvious what happens if they do not) [14:30]
shrysr: http://logs.nosuchlabs.com/log/ossasepia/2019-09-09#1001756 publicly and believe and difficult > does that mean the reverse exists/known by some? [14:31]
snsabot: Logged on 2019-09-09 14:06:15 asciilifeform: shrysr: if you expect to 'understand hashing' the way you understand e.g. ohm's law, will be waiting a long time. hashing is not a physical law concept , but cryptological -- i.e. authors of hash algoes ~believe~ that the operation where input turns into a corresponding garbage is difficult to reverse, but there is no proof aside from 'no one publicly showed how to reverse in N years'. [14:31]
asciilifeform: shrysr: in the case of e.g. 'md5' hash, was found to be inexpensive to violate both conditions. [14:32]
asciilifeform: shrysr: authors of a hash algo make the promise that it is necessary to 'brute force', i.e. walk all possible inputs, from 0000.....0 to 111111....1 , in order to find y where H(y) == H(x) for given x. but historically often turns out that is not necessary to walk whole space. [14:34]
asciilifeform: shrysr: when this happens, the algo is declared 'broken' and someone proclaims a new one. [14:35]
shrysr: asciilifeform: what abt the impact of the broken algo? what happens after that? for eg i use SHA512 to encrypt a whole buncha things -- if say I save a hashed message on archive.is > the broken algo means it can be decrypted correct? for eg this chan log, or some several-year communication was secret and encrypted as a daily log, and meant to remain so, but now it is not any longer? so I change to the new [14:45]
shrysr: algo henceforth, but what abt all the old stuff? [14:45]
asciilifeform: shrysr: hash is used in signing, rather than encrypting [14:46]
asciilifeform: when a hash algo is broken, it becomes possible for someone to bring that 'y' where H(y) == H(x) and say that you signed y as well as x . [14:47]
asciilifeform: shrysr: to see how hashes are used in gpg , see this passage in ffa ch.6 . [14:47]
asciilifeform: shrysr: in 'v', hashes are also used to indicate the predecessor state of a text . i.e. in this example, the machine is told that the previous MANIFEST.TXT must be such that hashed to 65b06852de..... and the new one being formed, via the given transform, will hash to 3f84fd0800..... . [14:53]
asciilifeform: a successor patch to this one, will expect , supposing that it alters MANIFEST.TXT, that the given one will be such that hashed to 3f84fd0800..... [14:54]
asciilifeform: shrysr: the gpg signatures are used to verify that particular people (i.e. holders of particular gpg private keys) had in fact signed a particular patch . [14:55]
asciilifeform: shrysr: initially the original author of a patch, publishes it along with signature. afterwards, people who believe that they understand how it works, publish also signatures of their own. [14:56]
asciilifeform: shrysr: observe how in e.g. logotron vtree, there are signatures from several people for each patch. [14:56]
asciilifeform: shrysr: look at the patches, and observe what happens to the hashes. the successors ( which appear after '+++ ...' ) become the predecessors ( i.e. appear after '--- ...' ) in the next patch in the series each time. [14:58]
asciilifeform: shrysr: the current vtron uses diana_coman's implementation of 'keccak' hash algo, rather than sha512, for this. [15:01]
asciilifeform: ( while gpg uses sha512 in the signatures ) [15:01]
shrysr: ok. ima go through these links and revert. fwiw: the signing was also a part i was stuck... and never reached in the handbook. so the private key used in a sign says it is You, verified finally by your public key..to be from you.. ok....but i cant connect how is the File is 'signed' to be from you and verified? is it like a random part of the file is extracted and mashed with your private key? why do you [15:02]
shrysr: share a patch and separate signed file and why separate.. even if latter is out of convenience or i was told 'transport logistics' or so in #gnupg, but still - why not a single signed file? [15:02]
asciilifeform: shrysr: 'verify' is the operation illustrated in the ch6 link [15:03]
asciilifeform: via rsa, you find out that 'this hash was signed by the holder of the private key that goes with this-here public key' [15:04]
asciilifeform: shrysr: there is no 'mash', the operation is the modular exponentiation that is the 'rsa operation' . [15:05]
asciilifeform: shrysr: the signatures are separate files, rather than appended ('clearsigned' as sometimes is used for a single-author document) so as to enable multiple sigs of a particular patch . [15:06]
asciilifeform: (i.e. by different people) [15:08]
asciilifeform: shrysr: if it isn't clear -- 'verification' is something ~you~ do when you possess a 1) signed document 2) its signature 3) the public key of the purported signer. [15:09]
asciilifeform: it is a mathematical operation that produces a 'true' or 'false' output. [15:09]
asciilifeform: in the case of 'v', the vtron does it, when it produces the working set (i.e. list of patches and their authors) when you run it. [15:10]
asciilifeform: shrysr: this is why it takes three directories -- 'wot' (public keys) ; 'seals' (signatures of patches, by the owners of the privkeys that correspond to the pubkeys in 'wot') ; 'patches' (the patches themselves.) . [15:13]
asciilifeform: shrysr: when you do operations in 'v', the working set (of patches) is a subset of the contents of 'patches' for which there are valid signatures (in 'seals') by the owners (i.e. privkey holders) of the pubkeys in 'wot' . [15:14]
asciilifeform: shrysr: a '--- false' -> '+++ hash....' transition is called a 'root'. all patch segments in a 'genesis' (the 1st patch in a project, that creates files from 'thin air') are 'roots'. [15:24]
asciilifeform: shrysr: when you 'press to a given patch', vtron starts with the roots in genesis, and tries to find a path , walking through the ---/+++ pairs, to the hashes found on the +++ side of the patch being 'pressed to' . [15:25]
asciilifeform: if succeeds in finding this walk (with all the stones stepped on , being validly-signed patches) , you have a successful press. [15:25]
shrysr: i think i'm beginning to see a little bit. ....> so when diana also signs a patch you have released - it means she believes she has understood it, and endorses that the patch works as you claim? her sign is independent of your sign and separate. And since i trust you and diana (WoT) - i can be 'more' reassured to the extent of trust - that it does indeed work as claimed by whoever released patch? versus [15:27]
shrysr: say a single signature from you, which would still work for all who had you in their WoT? so for eg: if me/donkey-on-road signed a patch without reading a line of it? [15:27]
asciilifeform: shrysr: correct. if there are only signatures from 1 person, and then he goes mad and starts to sign garbage, or is killed and enemy takes his private key somehow -- then the work is lost. [15:28]
asciilifeform: (i.e. it becomes indistinguishable from the background hum of the universe, coad found on shithub, etc. someone could read it still and sign it, but few will find any desire to bother ) [15:29]
asciilifeform: a signature carries weight to the extent readers believe in the honesty and competence of the author , AND in his ability to keep the privkey secret from third parties [15:30]
asciilifeform: if one were to sign random garbage w/out reading or thinking, his signature will not be very interesting to anyone. [15:30]
shrysr: have to bbl >>> this is getting more interesting by min though.x01 [15:31]
asciilifeform: shrysr: plz read ben's overview. [15:31]
asciilifeform: it covers all of this. [15:31]
lobbes: http://logs.ericbenevides.com/log/ossasepia/2019-09-09#1001680 << done. It is currently set to announce auctions in here once every 24 hours. Let me know if you'd like a different frequency [20:39]
ericbot: Logged on 2019-09-09 14:16:30 diana_coman: http://ossasepia.com/2020/04/20/ossasepia-logs-for-09-Sep-2019#1001675 - yes, please do. [20:39]
ossabot: Logged on 2019-09-09 11:03:02 lobbes: you okay if I do? I'll just need to add it to the cron job [20:39]
lobbes: fyi, the "!Xlist" command will allow you to 'manually' list all active auctions at any time (including in PMs) [20:40]

Comments feed: RSS 2.0

Leave a Reply