Results 1 ... 250 found in all logged channels for 'f:stjohn_piano_2' | →
(trilema) stjohn_piano_2: New comment policy added for TMSR comments on Edgecase: http://edgecase.net/pages/how_to_comment_on_edgecase_datafeed
(trilema) stjohn_piano_2: http://btcbase.org/log/2019-05-22#1914979 << i am not self-supporting. live with parents, focus on recovery and retraining self.
(trilema) stjohn_piano_2: http://btcbase.org/log/2019-05-22#1914978 << aha. well, currently, if intelligent heathen really wants to tell me his thoughts, he can create an account.
(trilema) stjohn_piano_2: asciilifeform: aha. make sense.
(trilema) stjohn_piano_2: asciilifeform: this is not deliberately a spam trap. i had a choice while working: to build an open-access-point to the wider internet, or not to do so. i chose - not.
(trilema) stjohn_piano_2: ah, wait. so far, none, because i never permitted it.
(trilema) stjohn_piano_2: asciilifeform: i did consider both. ultimately, i decided to stay with identity-based commenting rather than take a sift-the-noise approach.
(trilema) stjohn_piano_2: 3) there may be an approach to signed comments that does not involve the primary key: examples: a) sign with a subordinate key b) validate your current IP for a time period (e.g. a month / year) by decrypting an OTP, like deedbot.
(trilema) stjohn_piano_2: in future, this will become easier for established identities, as i work out how to automate parts of the process.
(trilema) stjohn_piano_2: 2) this first-pass at a comments system makes it impossible for people without an identity to speak. established identities can speak, although it is hard, yes.
(trilema) stjohn_piano_2: 1) until recently, no comment system existed. it is still experimental and manually driven. i am unsure about its final structure.
(trilema) stjohn_piano_2: there's several threads here. i'll go through them.
(trilema) stjohn_piano_2: your comment has been published. i'm pleased that the comment system worked. yours is the first external comment.
(trilema) stjohn_piano_2: i understand your point. i've thought about this quite a bit.
(trilema) stjohn_piano_2: it's certainly public-facing, but i have little interest in interaction with speech that does not come from an established identity of some sort.
(trilema) stjohn_piano_2: i add comments manually to the data at present. in future, i may construct an automated system that can accept signed comments.
(trilema) stjohn_piano_2: http://btcbase.org/log/2019-05-22#1914934 << aha. yes, for the moment, this is how i can handle comments from people in the forum.
(trilema) stjohn_piano_2: http://btcbase.org/log/2019-05-22#1914933 << this is the comment policy for the wider internet.
(trilema) stjohn_piano_2: i logged and summarised my experience of joining the web of trust (probably of interest mostly to noobs): http://edgecase.net/articles/joining_the_web_of_trust
(trilema) stjohn_piano_2: http://btcbase.org/log/2019-05-17#1914458 << yes. also, easier to scoop up dice again for next throw.
(trilema) stjohn_piano_2: i'm now trying to restructure myself into a dev in a field in which remote work can become a viable option.
(trilema) stjohn_piano_2: http://btcbase.org/log/2019-05-17#1914408 << yes, exactly. sadly, small business was not the right path. i've observed firsthand the final stages of the integration of small businesses into the gov. the regulatory overhead is enormous and ever-growing.
(trilema) stjohn_piano_2: http://btcbase.org/log/2019-05-17#1914400 << yes, tied to the UK. I'm currently looking for python dev work in Cambridge/Suffolk area.
(trilema) stjohn_piano_2: http://btcbase.org/log/2019-05-17#1914395 << from logs + mocky's "why ada", i understand some of the technical merits of ada. however, my perception of the current job market is that trying to make a living at ada leads directly to a hardware sharashka.
(trilema) stjohn_piano_2: a small village near newmarket, in suffolk.
(trilema) stjohn_piano_2: re how/why: due to rsi (repetitive strain injury), i currently run at 40-50% text production capacity, if i type slowly and steadily, with breaks. this is up from 0.1% capacity two years ago. trying to respond in real-time, as i did yesterday, is too high a rate for me to keep up.
(trilema) stjohn_piano_2: http://btcbase.org/log/2019-05-16#1914325 << a distinction of sorts, i guess. "one body at a time, minefield is cleared".
(trilema) stjohn_piano_2: http://btcbase.org/log/2019-05-16#1914167 << i have no doubt that there are interesting problems in eulora, else you and mircea popescu would not spend so much time on it. unfortunately, i have to prioritise looking for paid work / finding a job.
(trilema) stjohn_piano_2: http://btcbase.org/log/2019-05-16#1914162 << i did not look at them in detail. i filed them under "investigate further when i buy one".
(trilema) stjohn_piano_2: http://btcbase.org/log/2019-05-16#1913871 << have been pondering this question, and particularly the additional question "what answer would i like to be able to give in future?".
(trilema) stjohn_piano_2: i contemplated the possibility of 20-50 authors on edgecase. not people here now, who already have their own blogs, workers, etc, but new people, who perhaps don't want to run an entire blog, but would like to occasionally sign and publish their work.
(trilema) stjohn_piano_2: this led to: multi-author, every article is signed (optionally by the author as well), articles are in a linear order (hence "103"), unalterability means that additional authors can trust that the owner can't sabotage their work / code.
(trilema) stjohn_piano_2: perhaps it's worth mentioning that, at every point when building edgecase, i asked myself "what should a reputable publishing system in the age of cryptography look like?".
(trilema) stjohn_piano_2: re: broken hands. yes, i cannot sell sheer volume of typing, only reputable attention/focus/correction.
(trilema) stjohn_piano_2: i studied some russian once upon a time, i could almost certainly correct it, with a little practice.
(trilema) stjohn_piano_2: http://btcbase.org/log/2019-05-16#1913861 << i was not contemplating doing the bulk of the typing myself. my preferred approach would be to find a russian speaker, get them to make a GPG key, do the bulk work, and send it to me for correction.
(trilema) stjohn_piano_2: example: 1) numbered image files containing schematics 2) text files containing transcribed text 3) notes files containing the text on the schematics, with a reference to the corresponding image file.
(trilema) stjohn_piano_2: asciilifeform: what would the desired result look like?
(trilema) stjohn_piano_2: woops. wrong window.
(trilema) stjohn_piano_2: yum install qpdfview
(trilema) stjohn_piano_2: http://btcbase.org/log/2019-05-16#1913779 << this should have been "decentralisation vs parsimony". the article in question is http://trilema.com/2015/that-scary-thing/
(trilema) stjohn_piano_2: http://btcbase.org/log/2019-05-16#1913740 << an error, which i will have to go and fix. checkpoint 8 has been held back from "published" status, because there is a little bitcoin in the address derived from the checkpoint.
(trilema) stjohn_piano_2: http://btcbase.org/log/2019-05-16#1913698 << i also completely forgot that GPG encrypts to a subkey, not to the primary, so of course the key ID would be different.
(trilema) stjohn_piano_2: i'll go through some dropped references from yesterday:
(trilema) stjohn_piano_2: turns out that for me fast real-time chat is still rather unaffordable at present. going forward, my replies will have significant latency.
(trilema) stjohn_piano_2: diana_coman: gtg for now. will reply at a later date.
(trilema) stjohn_piano_2: diana_coman: i had a look.
(trilema) stjohn_piano_2: ^ trinque. hope you see the above about OTP before doing any investigation.
(trilema) stjohn_piano_2: diana_coman: yes. makes sense. annoying mistake on my part.
(trilema) stjohn_piano_2: in my notes, i had this sequence: 1) join forum, 2) deedbot will present OTP, 3) decrypt OTP, 4) use !!v OTP.
(trilema) stjohn_piano_2: by mistake, i was attempting to decrypt the OTP sent to asciilifeform
(trilema) stjohn_piano_2: just back to say: solved (non-)mystery of OTP
(trilema) stjohn_piano_2: asciilifeform, mp_en_viaje: thanks again for the patient help.
(trilema) stjohn_piano_2: i'm going to get off the screen for today.
(trilema) stjohn_piano_2: mp_en_viaje: yes.
(trilema) stjohn_piano_2: to quote myself "To attack a hidden-key address, an adversary would need to discover weaknesses in: ECDSA, SHA256, RIPEMD-160. These weaknesses would also have to be compatible. To attack a known-key address, an adversary would only need to discover a weakness in ECDSA."
(trilema) stjohn_piano_2: section is called "Does the hash in a Bitcoin address provide any protection?" (if you search the page for that string, you'll get to it)
(trilema) stjohn_piano_2: i wrote this up myself, to figure it out
(trilema) stjohn_piano_2: http://btcbase.org/log/2019-05-16#1914109 << yes. i didn't even trust the implementation of the hash function though. i started out with the assumption "it's all terrible" and the conclusion was "the only true test is to get a transaction from this address into the blockchain".
(trilema) stjohn_piano_2: asciilifeform: thanks. i have read it already though (and much of trilema).
(trilema) stjohn_piano_2: re: testing: excellent. i will eventually buy one. need to get a job first.
(trilema) stjohn_piano_2: people reeeeally try though. "machine learning" (rather than "statistical sampler with paremeter adjustment").
(trilema) stjohn_piano_2: yes. people try, but final responsibility can never be outsourced.
(trilema) stjohn_piano_2: although, i am surprised. i had thought "driver" was one category of subordinate where you could trust that the subordinate would do the work carefully, for his own sake.
(trilema) stjohn_piano_2: asciilifeform: ah.
(trilema) stjohn_piano_2: the question in my mind is: does this strategy lose out to write-bitcoin-scripts-in-scripting-language to run offline on whatever cheap hardware china makes?
(trilema) stjohn_piano_2: question becomes: spend lots of time becoming expert in old tech, run homebrew hardware for running a wallet, but then: how to monetise? the expenditure in time alone is enormous.
(trilema) stjohn_piano_2: i'm in the uk, for my sins.
(trilema) stjohn_piano_2: so digging through junkyards is actually workable for my lifetime?
(trilema) stjohn_piano_2: huh. now that is interesting.
(trilema) stjohn_piano_2: asciilifeform: yes. i noticed. i have a 10yo macbook that functions still, while newer ones.... well, not so good.
(trilema) stjohn_piano_2: well, when (roughly) can ~all the 70s, 80s stuff be expected to be dead, purely from entropy.
(trilema) stjohn_piano_2: ah interesting. did not know that. had assumed old stuff would break in some way over time.
(trilema) stjohn_piano_2: yes, very nearly. but some 90s-era cpus still run, no?
(trilema) stjohn_piano_2: i contemplated a future (my middle age?) in which all of the old stuff no longer works.
(trilema) stjohn_piano_2: asciilifeform: yes. the approach, for now, is fine.
(trilema) stjohn_piano_2: e.g. dig dig dig "ah, this Y is a wrapper around a half-melted X"
(trilema) stjohn_piano_2: having spent significant time in their guts
(trilema) stjohn_piano_2: ah. when i started on quest, i had filed all computers under "untrustworthy".
(trilema) stjohn_piano_2: asciilifeform: all this is true. is it still a bad idea if you run the raspi offline forever?
(trilema) stjohn_piano_2: http://btcbase.org/log/2019-05-16#1914041 << well, i don't yet know how to do ecdsa, ripemd160, sha256 on paper.
(trilema) stjohn_piano_2: well, in my initial half-understanding of bitcoin, was nervous about moving bitcoin into offline address without knowing (for certain) that i could retrieve it.
(trilema) stjohn_piano_2: variant in this case is: use raspberry pi to generate offline transaction to confirm address validity.
(trilema) stjohn_piano_2: interest from my friends and acquaintances in "store bitcoin on paper securely" using edgecase stuff has been: ~0.
(trilema) stjohn_piano_2: asciilifeform: yes, unfortunately.
(trilema) stjohn_piano_2: asciilifeform: i have indeed.
(trilema) stjohn_piano_2: mp_en_viaje: makes sense. i don't think i'm a good choice for advertising work.
(trilema) stjohn_piano_2: in this case, mini-mill for consumer.
(trilema) stjohn_piano_2: asciilifeform: i learned early on that he who has the money gets to decide what to spend it on.
(trilema) stjohn_piano_2: mp_en_viaje: electronic engineering
(trilema) stjohn_piano_2: asciilifeform: i agree. goal was to make "cnc mill for consumers" i.e. cheap.
(trilema) stjohn_piano_2: although for speech rec, it was small company (my father), so I also did bookkeeping, reading about taxes, setting up computers, etc.
(trilema) stjohn_piano_2: mp_en_viaje: yes, exactly
(trilema) stjohn_piano_2: speech rec company: learned to build / tune /test speech rec systems (nuance, grxml). ran transcription projects for the test data. wrote statistical sampler scripts for the output.
(trilema) stjohn_piano_2: to make mini-mill thing for wood.
(trilema) stjohn_piano_2: mp_en_viaje: furniture factory: i studied all the components i was given and tried to make them work together: so, camera, computer, motion controller, servomotors, wiring etc.
(trilema) stjohn_piano_2: asciilifeform: so i discovered, yes.
(trilema) stjohn_piano_2: asciilifeform: ^ those, for pay.
(trilema) stjohn_piano_2: speec rec company wanted transcription projects to test the speech systems.
(trilema) stjohn_piano_2: furniture factory wanted to see if they could make and sell a cnc mini-mill.
(trilema) stjohn_piano_2: ah. actually two businesses. one was a furniture factory, one did speech rec system tuning.
(trilema) stjohn_piano_2: that was just experimentation with edgecase.
(trilema) stjohn_piano_2: no, not as paid work.
(trilema) stjohn_piano_2: as a quick snapshot.
(trilema) stjohn_piano_2: I've done: writing python scripts to sort large amounts of transcription data, running transcription projects, setting up raspberry-pi-powered cameras, figuring out what to do about gdpr, bookkeeping.
(trilema) stjohn_piano_2: hm. let's do a quick run-through:
(trilema) stjohn_piano_2: not so far, but i can certainly see if i can do it.
(trilema) stjohn_piano_2: gpg: decryption failed: secret key not available
(trilema) stjohn_piano_2: gpg: encrypted with RSA key, ID C8EFFF13
(trilema) stjohn_piano_2: strange. doing the operation again with the original OTP still produces the error.
(trilema) stjohn_piano_2: asciilifeform: ah. thanks.
(trilema) stjohn_piano_2: asciiliform, mp_en_viaje: thanks for the help
(trilema) stjohn_piano_2: possibly browser introduced some stuff.
(trilema) stjohn_piano_2: ah. first time -> opened link in browser, copied text into file. did not use curl.
(trilema) stjohn_piano_2: !!v 224B5AF74B3FDF7674566BBBC2C7AF0E46D68F6586E03C255C2631FD34BE60DD
(trilema) stjohn_piano_2: 224B5AF74B3FDF7674566BBBC2C7AF0E46D68F6586E03C255C2631FD34BE60DD
(trilema) stjohn_piano_2: UP stjohn_piano_2
(trilema) stjohn_piano_2: gpg: encrypted with 4096-bit RSA key, ID 625FF273, created 2019-04-15 "stjohn_piano_2"
(trilema) stjohn_piano_2: mp_en_viaje: result:
(trilema) stjohn_piano_2: should i use the first or second, or either?
(trilema) stjohn_piano_2: !!up stjohn_piano_2
(trilema) stjohn_piano_2: mp_en_viaje: ok. round 2 i.e. new OTP?
(trilema) stjohn_piano_2: gpg: decryption failed: secret key not available
(trilema) stjohn_piano_2: $ gpg --decrypt encrypted_otp.asc
(trilema) stjohn_piano_2: gpg: encrypted with RSA key, ID C8EFFF13
(trilema) stjohn_piano_2: so my working assumption is that the priv is in fact the right one.
(trilema) stjohn_piano_2: asciilifeform: have exported pub-derived-from-priv and diffed it with pub-downloaded-from-deedbot. no difference other than "Version: GnuPG v1.4.10 (GNU/Linux)".
(trilema) stjohn_piano_2: asciilifeform: i have imported the corresponding priv. will now export pub and confirm that it's the same as the one in deedbot.
(trilema) stjohn_piano_2: i've abandoned the stateless commands and am using the basic stuff.
(trilema) stjohn_piano_2: my pgptron is gpg 1.4.10, compiled from mp's original source in deedbot.
(trilema) stjohn_piano_2: http://btcbase.org/log/2019-05-16#1913894 << really wanted more-or-less self-contained unchangeable publishing. thought i might as well make it look nice.
(trilema) stjohn_piano_2: mp_en_viaje: re: maturity: good point.
(trilema) stjohn_piano_2: (correct key ID of the public key) *
(trilema) stjohn_piano_2: puzzler: why the key ID in the otp message is C8EFFF13, while my key's fingerprint is 599152AC. importing the private key into gpg produces (in gpg --list-keys) the correct key ID 599152AC. diffing the key returned by deedbot with my public key shows no difference.
(trilema) stjohn_piano_2: i know, hadn't expected this part to produce trouble.
(trilema) stjohn_piano_2: asciilifeform: no. still tracing back over the steps.
(trilema) stjohn_piano_2: i guess i could say that i'm very good at reading a lot of material and picking out the bit that is necessary for a problem.
(trilema) stjohn_piano_2: i have unfortunately worked mostly in small businesses, making me something of a "learn this thing quickly well enough to get it to do X function".
(trilema) stjohn_piano_2: mp_en_viaje: thanks for feedback, by the way.
(trilema) stjohn_piano_2: asciilifeform: ah interesting.
(trilema) stjohn_piano_2: http://btcbase.org/log/2019-05-16#1913862 << yes, reading logs has thickened skin substantially.
(trilema) stjohn_piano_2: BingoBoingo: I use relatives.
(trilema) stjohn_piano_2: asciilifeform: this computer doesn't have a djvu reader, but i have no doubt that it's painful.
(trilema) stjohn_piano_2: well, i'd use human-powered ocr then (given residual rsi). i've hired transcribers before. i'd then correct the result myself.
(trilema) stjohn_piano_2: ah. i'll look at the original.
(trilema) stjohn_piano_2: well, we'd choose some hourly rate agreeable to both of us. i'd use an OCR site and correct the result. eventually, i might invest time in my own ocr fork.
(trilema) stjohn_piano_2: mp_en_viaje: i highly doubt that edgecase can compete with slave labour.
(trilema) stjohn_piano_2: asciilifeform: yes. i have imported the corresponding priv key to this instance of gpg.
(trilema) stjohn_piano_2: mp_en_viaje: i agree. no problem, i'm used to that reaction.
(trilema) stjohn_piano_2 is reading http://edgecase.net/articles/basic_gpg_commands to try to use GPG the normal way to decrypt the otp.
(trilema) stjohn_piano_2: mp_en_viaje: yes
(trilema) stjohn_piano_2: i have not implemented a javascript select thing yet.
(trilema) stjohn_piano_2: mp_en_viaje: my name is actually StJohn Piano. my brother's name is Nicholas Piano. the company is just me.
(trilema) stjohn_piano_2: mp_en_viaje: the name of the company
(trilema) stjohn_piano_2: asciilifeform: diffing the output of pgpdump -i [key] and pgpdump -i [deedbotkey] shows no difference.
(trilema) stjohn_piano_2: transactions for timestamping is a bonus.
(trilema) stjohn_piano_2: http://btcbase.org/log/2019-05-16#1913785 << well, i had time to think. wanted to understand transactions, and by doing so achieve airgapped bitcoin storage.
(trilema) stjohn_piano_2: mp_en_viaje: certainly. i'm not certain that i knew about xxd at the time.
(trilema) stjohn_piano_2: asciilifeform: exactly
(trilema) stjohn_piano_2: uploaded to one of the various broadcast places.
(trilema) stjohn_piano_2: hand-crafted transactions.
(trilema) stjohn_piano_2: asciilifeform: i did not. thanks.
(trilema) stjohn_piano_2: you wrote an article once about parsimony vs efficiency, i think.
(trilema) stjohn_piano_2: mp_en_viaje: yes, I've learned a lot. also distribution of risk.
(trilema) stjohn_piano_2: i cannot alter previously published articles.
(trilema) stjohn_piano_2: checkpoint articles are hashed and made into bitcoin addresses. some bitcoin is transferred to this address, much like deedbot.
(trilema) stjohn_piano_2: i have not spent the time to go through the pieces manually.
(trilema) stjohn_piano_2: i mean, i understand the idea.
(trilema) stjohn_piano_2: asciilifeform: yes, although i have not been able to understand it.
(trilema) stjohn_piano_2: thanks for voice.
(trilema) stjohn_piano_2: mp_en_viaje: first thing i bumped into that solved my problem.
(trilema) stjohn_piano_2: repetitive strain injury. too much typing, causes muscle buildup.
(trilema) stjohn_piano_2: asciilifeform: will eventually switch over to v, yes.
(trilema) stjohn_piano_2: until recently, chat has been an unaffordable liability.
(trilema) stjohn_piano_2: but i've been recovering from severe rsi
(trilema) stjohn_piano_2: asciilifeform: a long time.
(trilema) stjohn_piano_2: manage_ip_addresses is a small interface for the paywall ip addresses. it's a thrown-together paywall, not fantastic, but functional.
(trilema) stjohn_piano_2: mp_en_viaje: no, just haven't got to it yet.
(trilema) stjohn_piano_2 is learning pgpdump in the background, btw
(trilema) stjohn_piano_2: except that the signature is wrapped and embedded at the end of the text article.
(trilema) stjohn_piano_2: asciilifeform has it.
(trilema) stjohn_piano_2: the original items can be downloaded via the "Download this article" link.
(trilema) stjohn_piano_2: it's an html rendering / presentation of a series of text articles.
(trilema) stjohn_piano_2: asciilifeform: python
(trilema) stjohn_piano_2: but i did consider it.
(trilema) stjohn_piano_2: mp_en_viaje: i severely dislike PHP.
(trilema) stjohn_piano_2: asciilifeform: will do.
(trilema) stjohn_piano_2: i studied trilema's layout.
(trilema) stjohn_piano_2: mp_en_viaje: yes, very much so.
(trilema) stjohn_piano_2: asciilifeform: gpg --list-packets shows that the key ID is C8EFFF13 in the OTP file. The key ID of my public key is 5991 52AC. Is this expected? (the fingerprint shown by deedbot in the earlier line is correct, though).
(trilema) stjohn_piano_2: asciilifeform: diff shows "Version: GnuPG v1.4.10 (GNU/Linux)" as the only difference.
(trilema) stjohn_piano_2: asciilifeform: when i run your command there, I get: "gpg: encrypted with RSA key, ID C8EFFF13, gpg: decryption failed: secret key not available"
| →
Download hourly DB snapshot | Get Source Code