mircea_popescu: simple "offline" qt : have two talk to each other, inspect the interface, extract tx, dump it to printer.
mod6: i'll also update the graph while im at it.
mircea_popescu: what distinction do you do there ? so you sign the txn on one system, and dump it as a rawtx on another. this is a big deal somehow ?
mod6: yah, shinohai tested these for me today. seemed to work for him.
adlai: Carli-: not really, i don't think this is the same kind of procedure as googling three-ingredient brownies
Carli-: adlai: is there a guide, that you recommend for that?
adlai: Carli-: you build the tx on the online node, transfer it across the airgap, sign on offline wallet, transfer back.
Carli-: adlai: How do you make a transaction while keeping the wallet offline?
adlai: Carli-: well, you can keep your wallet(s) on an offline computer, and only send the addresses around on the internet - using any program.
mod6: asciilifeform: any objections to me replacing those vpatches on the mirror now, or should I wait for you to give them a try first?
adlai: and it didn't scream "these people hate javascrypto" at you?
Carli-: asciilifeform: cool, is there a guide that you recommend to do " pre-generated tx" ?
adlai: Carli-: out of curiosity, how did you find this place? people usually have some inkling of its philosophy when they wander in...
asciilifeform: Carli-: the state-of-the-art among thinking folk is that pre-generated tx are stored on paper and fed into a hot node when necessary
adlai: Carli-: again, the point here is to simplify bitcoin to the minimal required functionality. feature creep, even if they're useful features, is to be avoided.
asciilifeform: fact is, there ~is~ no non-braindamaged 'wallet', aside from what each of us has quietly hacked together.
adlai: asciilifeform: would it still be correct to say that there is no intent that people use trb as a wallet? or at least, use the same trb as a network node and a wallet node?
Carli-: asciilifeform: How do you do Deterministic or Offline Wallets, without weakening the crypto?
asciilifeform: because it is a motherfucking REFERENCE implementation
adlai: Carli-: note that http://therealbitcoin.org has no wallet.
asciilifeform: Carli-: it is possible to do with without weakening the crypto.
adlai: this is an operational concern. the bitcoin node software doesn't even need to know what a private key is!
Carli-: asciilifeform: well, the point is to keep your private keys on a Air-gapped computer offline forever
adlai: Carli-: primarily you should understand that the main goals of bitcoin development (in this channel, at least) are to minimize the amount of moving parts
asciilifeform: 'this pistol fires the bullet backwards into my face. are there any ~other~ problems with it?'
Carli-: asciilifeform: yes, i think one of the problems: if someone has 2 of your Private keys AND your Master Public Key, they can find ALL of your private keys. IIRC
adlai: there are some use cases for it, sure. but there's a reason that they weren't automatically merged in despite being invented by the same people
Carli-: asciilifeform: is there something wrong with BIP32?
adlai: Carli-: lol, you should probably read the bip if you're that excited about it
adlai: this is the whole point of bip32 - backups which only need to be stolen once
asciilifeform: Carli-: do you understand what this means mathematically ?
Carli-: asciilifeform: so you can back up your wallet to a piece of paper once, only 18 words long is the "seed"
asciilifeform: why the FUCK would i
asciilifeform: Carli-: therealbitcoin.org
Carli-: asciilifeform: which wallet do you use? or are there any other wallets, that can do: Deterministic, Offline transactions, Multiple Wallets ?
Carli-: asciilifeform: And you can run Multiple wallets at the same time. Is there a way to do that with bitcoin-qt? I am trying to figure that out
BingoBoingo: Because at one point loading the whole blockchain into RAM was an idea that didn't give most people pause.
asciilifeform: Carli-: i also don't much care what the best stove for burning pieces of my sofa to heat my house would be, either
BingoBoingo: asciilifeform: The point of Armory was making people with too much RAM and CPU for their own good feel speshul
asciilifeform: http://log.bitcoin-assets.com/?date=04-02-2016#1395969 << see also the fake 'uber' cabs, and similar
assbot: Logged on 04-02-2016 00:11:10; BingoBoingo: Case Study: Watch a shitgnome in the wild paper over its enshittening https://www.reddit.com/r/Bitcoin/comments/4420gt/the_end_of_the_road_for_armory/czmu3k5
asciilifeform: http://log.bitcoin-assets.com/?date=04-02-2016#1395952 << i for one never understood the point of 'armory' to begin with
assbot: Logged on 04-02-2016 00:08:24; mircea_popescu: http://log.bitcoin-assets.com/?date=03-02-2016#1395693 << the way i expect this to work at least for a while will be shiva-p which will be a pretty limited thing and shiva-l which'll be an actual debugger and all that.
BingoBoingo: guruvan: rope the fugly meerkat?
mircea_popescu: and in "views from the dirigible" news, http://40.media.tumblr.com/cb1e90b06df5ded1f71394833fb37595/tumblr_mvm35nynvq1s62w2ao1_1280.jpg
BingoBoingo: Just enough rope for a person to shoot themself in the foot when they get surprised that rope does that
felipelalli: but yes, it's not the safest thing anyway!
mircea_popescu: but basically the same thing except as a java ?
felipelalli: mircea_popescu, thank you mircea to understand my motivation. Yes, I tried to help. The "magic" to make the thing easy was to derivate a bitcoin keypair from the username+password using scrypt. I understand this can be unsafe if the user uses a bad password, but like you said, it is just to let the lazy people to taste the power of wot.
mircea_popescu: but in other news, http://41.media.tumblr.com/8ef2f0ce131bb0bde4bc4efdbfc8e247/tumblr_nw2l90w5Du1uu92gho1_1280.jpg
mircea_popescu: felipelalli the reason you're getting so much grief is that it's not clear how your premises work. i mean i get it, you want to help people, and teach them about otc and so on. this is one thing. but i mean what did you do, i don't follow, reimplement gpg as a java thing ?
BingoBoingo: felipelalli: Maybe change the license on the software to: "You do not have license to use this code unless you translate the Java Bytecode to Fortran 95"
felipelalli: I'll put a link to source code and in Github I'll put the jar in "releases" section. I think it is better in this way at least while the app isn't so mature yet.
felipelalli: BingoBoingo, ahahh yes. But there is no special code to WIN32 actually in Java. The "exe" was just a wrapper. But if the guy can't execute a jar in Windows he just won't be able to use WoT anyway.
BingoBoingo: felipelalli: I mean you can leave source for the old win32 stuff so the lazy but desperate can still try building it. Think of it like putting the icecream shop at the mall on the top floor so fat people gotta choose icecream or using stairs.
felipelalli: danielpbarron, BingoBoingo etc. you guys convinced me to remove the win32shit stuff. I'll remove all files, actually.
felipelalli: http://log.bitcoin-assets.com/?date=03-02-2016#1395886 << thank you mircea_popescu - the article I am writing starts with a quote made by your PR translated to Portuguese: "Se você não tem uma conta WoT você não faz parte do ecossistema Bitcoin. Esse é o critério, não importa o que você possa achar. É onde todos olham, não importa o que alguma rede social tem dito pra você. Se você não está no WoT você não está no
pete_dushenski: off to pickup new specs. if i'm not back this evening, i'll be around under 'pete_d_out' over the coming days. cheers!
pete_dushenski: "Mr. Alverado said those fake buildings were necessary because getting to the first page in Google results now took ingenuity and cunning. “You have no idea,” he said, sounding a little weary when asked about competition. Israelis were his toughest rivals, he said, and they had instilled a kind of awe in him. “I can tell you point-blank, they are freaking smart,” he said. “I really admire them.”"
pete_dushenski: "Many of the locksmith lead gens are run by Israelis, and Avi learned their modus operandi by working for them. When he landed at La Guardia Airport in 2008, he wanted to work with computers, the field he had trained for in Israel. But it was the height of the recession, and he did not have many options."
BingoBoingo: pete_dushenski: Derps derping their way out is bullish for a 2013 replay. All we need now is RealSolid returning with yet another altcoin scam.
pete_dushenski: "Along the way, we accumulated a mess of legal and corporate complexity that has made it difficult to do anything constructive with Armory's intellectual property." <--> "You think you want to take dollar investment, you think you want to make a non-Bitcoin company ? Think again. Think long, think hard, and start with this : what makes you think a fiat company is even possible anymore ?"
mircea_popescu: or make anything worth the mention whatsoever.
mircea_popescu: "I could probably write a book about the depth and complexity of events of the past year, the lessons I've learned and the personalities I've dealth with. However, at this point it's all history, and I've always been more interested in planning the future than dwelling too much on the past." << no, he couldn';t write a book. not anymore than boshiman could build a city on all the land he walks. moreover, orc imagines h
BingoBoingo: Case Study: Watch a shitgnome in the wild paper over its enshittening https://www.reddit.com/r/Bitcoin/comments/4420gt/the_end_of_the_road_for_armory/czmu3k5
assbot: Logged on 03-02-2016 20:57:51; thestringpuller: aha. so pretty much ~only~ for labratory use as stated.
mircea_popescu: http://log.bitcoin-assets.com/?date=03-02-2016#1395693 << the way i expect this to work at least for a while will be shiva-p which will be a pretty limited thing and shiva-l which'll be an actual debugger and all that.
pete_dushenski: mircea_popescu: that's probably still another 1.5 years off. you're usually 3 years ahead of the curve, neh ?
ben_vulpes: of particularly amusing note is that their payment-accepting system is broken.
mircea_popescu: pete_dushenski and missing the "as mp said back in january 2015" bit, of course.
assbot: Logged on 03-02-2016 18:51:25; shinohai: "This also means that Gavins assumptions/estimates regarding the usage for 20 MB blocks unfortunately were very wrong:" https://redd.it/440cq6
pete_dushenski: "Blockstream, a tech startup that employs several core developers of the bitcoin protocol, has raised $55 million in venture capital to develop its sidechain technology and expand its global operations." [...] "Horizons Ventures, AXA Strategic Ventures and Digital Garage led the Series A funding round, which brings the startup's total venture funding to $76 million to date." << all this paper (and cardamon smell) and
mircea_popescu: phf and i don't need to see anything re genesis etc. just about the final destination.
mircea_popescu: meaning that all those contributed and the press depends on either a set made out of alf or a set made out of mod6
phf: and what's the work?
mircea_popescu: phf well it's ok-ish in the sense that i have all the data to try and do the work by my own eyes.
BingoBoingo: mircea_popescu: I mean for the operators
assbot: Logged on 03-02-2016 20:44:23; BingoBoingo: Of all the #b-a projects, deedbot's probably up there for frustration induced
phf: signers for the whole press, vs. signers for individual vpatches
assbot: Logged on 03-02-2016 20:40:10; danielpbarron: thestringpuller, i got into it with a fairly popular girly about "nice guys" and sexual dominance or something; turns out i think we were agreeing with eachother without realizing it
assbot: Logged on 03-02-2016 20:30:30; felipelalli: I am writing an article about the importance of the Bitcoin OTC WoT. This image is part of it: http://i.imgur.com/tHVbbva.png
pete_dushenski: so an idiot is me and i feel sorry for anyone who tries to do bitcoin casually because there's no goddam way.
pete_dushenski: the alternative (atm) is a ~fixed~ set of peers determined at random on start-up, as seen in 'multibit', which i can tell you from first hand experience this week has not been broadcasting select transactions from select keys. at all. and this is in no way fixable.
assbot: Successfully updated the rating for cazalla from 3 to 2 with note: friend of the Republic, former qntra guy
assbot: You rated user cazalla on 18-Aug-2014, with a rating of 3, and supplied these additional notes: Dedicated reporter..
assbot: Logged on 03-02-2016 17:01:32; ascii_butugychag: these things spend a good chunk of their time not node-in !1111
pete_dushenski: http://log.bitcoin-assets.com/?date=03-02-2016#1395266 << ~0.5x time is 'connecting' and 'disconnecting'. seems inefficient, but maybe that's the price of a distributed network ?
assbot: Logged on 03-02-2016 17:01:22; ascii_butugychag: does nobody but me actually watch their trb logs ?
pete_dushenski: http://log.bitcoin-assets.com/?date=03-02-2016#1395247 << i've had to restart as well every 2-5 days in the last 3 weeks. same 'network' issues causing lag of 10-100 blocks. fixed instantly by restart, new peer discovery.
assbot: You rated user PeterL on 14-Oct-2014, with a rating of -2, and supplied these additional notes: He made fetchbot, the bot that can't stay up..
assbot: Logged on 03-02-2016 13:29:59; asciilifeform: in other news, zoolag was blackholed for a record-breaking 8 hours.
pete_dushenski: http://log.bitcoin-assets.com/?date=03-02-2016#1395112 << tevye was about the same. also yesterday.
mircea_popescu: and for the record, the name for the opposite attitude used here is "power ranger". because totally.
topynate: ah, i see. the project in toto
mircea_popescu: no i mean, #bitcoin-assets etc. trying to understand how it works, fixing the code etc.
assbot: Successfully added a rating of 1 for topynate with note: I showed him the log, and marvel of marvels, he joined!
mircea_popescu: an argument persuasive enough that the whole foundation and the most serene republic is hard at work doing just that.
mircea_popescu: similarly to how my argument is that rather than "improving" bitcoin, that time'd be better spent fixing the codebase as it is.
topynate: nice koan. so istm your point is that the effort on fiddling with the parameters would be better spent on understanding why it works in the first place?
mircea_popescu: we so far fired a single bullet out of a gun we don't even know who the fuck made. seems a lot of unwarranted cockyness all around for this little knowledge.
mircea_popescu: this IS of course true, provided the child manages to alter the powder charge too, so the bullet's speed and its angle are always in the right ratio.
mircea_popescu: he then thinks to himself : "i could shoot as high or as low as i want, and the bullet wouldstill hit the apple, because i'm that special, and because that's what man made muskets for and how bullets work etc".
mircea_popescu: now this child sees an apple in the tree, shoots at it and hits.
mircea_popescu: topynate quite. consider, just for the sake of conversation, the situation of a child who rfound his father's musket.
topynate: if it really is a universal constant, then there are a lot of wasted words on the block size limit, including my own - anyone trying to abrogate it will just fall flat on his face
mircea_popescu: moreover, whether it does or might isn't the point. the point is, can you prove it doesn't have such nature ? because unless you can, speaking of it as a quota is iffy.
mircea_popescu: topynate well, consider that we can alter the value of h by altering the units of measurement. suppose all alteration of the 1mb block limit reduces to similarly a mere change in the units of measurement.
topynate: that would be remarkable; i can see the 1mb size being such a strong convention that no-one would have any expectation of successfully altering it, but unsure how it could have a universal nature like that of h
mircea_popescu: http://log.bitcoin-assets.com/?date=03-02-2016#1395578 << hey, he lives in a country where the poor buy tvs instead of fridges. what choice does he have, to move to wash dc ?
mircea_popescu: in this sense the argument can be brought that while the early block limit as implemented by miners (using 60% of the 1mb blocks) WAS a quota, the actual 1mb block size is not a quota at all,
topynate: ah. no, i don't think you can stretch it that far. that would be like saying the total quantity of oil in the ground was a quota on oil production.
mircea_popescu: besides, the same selective memory thing that was deliberately put into deedbot works here as well. some dead chains will be forgotten before others.
punkman: also could be valuable to see how B started out and then fixed, without having to diff the various versions of B patch
mircea_popescu: punkman the first is a->b'
topynate: i wrote that bit on the block size at https://topynate.wordpress.com/2016/01/26/note-on-the-block-size-limit/ that you called "not terrible". i'm a math-literate developer.
assbot: Logged on 03-02-2016 18:52:30; phf: so there's two sig's that are not sha512. genesis.vpatch.trinque.sig is sha256 and polarbeard_add_getpeerinfo_rpc.vpatch.polarbeard.sig is sha1
mircea_popescu: topynate hey there. who're you again ?
mircea_popescu: regrind, to be perfectly clear, is the act of making a tree out of patches instead of keeping them linear. this is a known efficiency-of-search trope, and i shouldn't have to explain how distributing the patches out saves work when i'm pressing 3 years later and want to read the whole string of patches from genesis.
assbot: Logged on 03-02-2016 18:05:59; ascii_butugychag: they smack of pious fraud, 'this was always perfect, born of the gods'
mircea_popescu: http://log.bitcoin-assets.com/?date=03-02-2016#1395483 << no dude, it's just "i have enough respect for the reader to not go A^HB^HA^HB^HC
mod6: put this in the cryptonomicon ^
mircea_popescu: you don't like doing the work, don't fucking write.
mircea_popescu: it's really up to the maintainers.
mircea_popescu: http://log.bitcoin-assets.com/?date=03-02-2016#1395474 << in your case you should end up with a->c'. if there's a much longer chain then the -b method makes more sense.
mircea_popescu: "Also, I heard that people are more willing to help animals than other people. Thus the donations to a pet rescue fund can be seen as the upper estimate of how donation to the webpage capture project would perform."
assbot: Logged on 03-02-2016 17:56:08; polarbeard: oh, now I see the complete picture
mircea_popescu: http://log.bitcoin-assets.com/?date=03-02-2016#1395440 << how about the logical thing, unique names PER TREE.
punkman: yeah but these are patches not scripts
assbot: Logged on 03-02-2016 17:55:51; ascii_butugychag: they get orphaned.
mircea_popescu: http://log.bitcoin-assets.com/?date=03-02-2016#1395437 << so if i make a script to fuck whores and i call it fuckwhores.pl and then it's mistaken and abandoned now i'm stuck using sexingsluts.pl instead and if that also gets burned i'll end up doing intromissiveman1pulationofwillingwenches.pl ?
mircea_popescu: http://log.bitcoin-assets.com/?date=03-02-2016#1395431 << i dunno that i renamed anything when signing it. i thought thatg's how we do it, i take the original v.pl and turn it into v.pl.me.sig
trinque: mod6: the tool you used was pywallet ?
mod6: trinque: hmm. yah, i had to like use that tool, do a rescan and a restart, and then it was in there. but yea, that back port, if sane, would be best.
trinque: mod6: I did a rescan but don't see my key as part of the default "account". I'll fiddle with the wallet.dat a bit more, then take a look at backporting that importprivkey
assbot: Logged on 03-02-2016 17:30:57; polarbeard: otherwise just check the sig date using pgp, last patch wins
assbot: Logged on 03-02-2016 17:28:22; jurov: lxr may be vtronic *if and when* someone takes my questions serious. so, asking again, what to do with reground or otherwise resubmitted patches?
BingoBoingo: Tis the natural order of things
mircea_popescu: i bought things so small you could carry them inside your nostril.
mircea_popescu: it's inconceivable without seeing it. like 20 sq feet affair that was maybe five foot tall, you had to go in two fucking steps. and then the actual shop, down a steep flight of stairs, another 30 sqft or so i guess.
BingoBoingo: mircea_popescu: What sort of wears were in the tiny shop?
mircea_popescu: so i've just been shopping in the tiniest shop i have ever seen.
phf: in other news takes 13.78 seconds to match all patches to all sigs with elimination on my laptop
thestringpuller: cowboy coding and debugging on production is for the truly brave.
ascii_butugychag: with no motherfucking 'idiomatic c'/
ascii_butugychag: but even then, a debugger that listens on tcp is a dangerous thing to have around
ascii_butugychag: thestringpuller: for starters, to find all of the 10,001 buffer overflows in tinyscheme and fix
thestringpuller: ascii_butugychag: re: http://therealbitcoin.org/ml/btc-dev/2016-January/000198.html << what needs to happen for this not need to be handled with radiation suit?
ascii_butugychag: which dies when the latter dies.
phf: doesn't need persistent for slime type use either. can just connect and mainloop, so the state will be around for duration of session?
ascii_butugychag: i am merely stating this for the benefit of noobz
phf: if i have a gpgme key instance, is there some way to patch its trust level?
trinque: ben_vulpes: first I'm getting deed txns going by using the trb wallet; I don't want to have to wait for an implementation of eatrawtx to resume deeds
gribble: The operation succeeded.
ben_vulpes: ;;later tell funkenstein_ yo are you going to submit that privkey import backport to the ml?
felipelalli: BingoBoingo, not all of them are my "friends". Actually I want to identify the bad actors in the community. I could implement an automatic -1 from me for who install the exe version!! lol
ben_vulpes: hey you're patching an eatrawtx into shiva then?
trinque: and now... there's shiva !
ben_vulpes: he probably destroyed what's left of the trinquetop in rage
BingoBoingo: felipelalli: Anyways it's a better use of time doing things that can't be done on windows and making your friends want to leave the windows favella than just doing everything for them.
felipelalli: I am going to my swimming lesson now. Thanks for the obvious comments that I was expecting. I tried to hide from you, but you got me! outch!
felipelalli: BingoBoingo, say about the Brazilian Bitcoin market is a good thing? For example?
felipelalli: and I am ashamed to trying to help the lazy people! But it is being fun.
BingoBoingo: danielpbarron: Still, you gotta learn the slurs and work on a more natural delivery I recommend just sticking reading the text posts here https://voat.co/v/fatpeoplehate/
BingoBoingo: related danielpbarron If you want to keep triggering fatties on twitter you gotta study and learn to deliver the lines yourself.
felipelalli: the exe I generated with a tool inside a VM
felipelalli: danielpbarron, they are lazy?
thestringpuller: "[We should see Bitcoin classic use ramp up] after binaries are released. Most people don't want to compile". << These people really shouldn't be involved in Bitcoin. Like going to war without learning to clean gun.
felipelalli: I am writing an article about the importance of the Bitcoin OTC WoT. This image is part of it: http://i.imgur.com/tHVbbva.png
phf: today lazy, tomorrow dead. poor technology choices kill. this message was brought to you by the tmsr agitprop dpt
BingoBoingo: felipelalli: What you should do for great lolz is depreciate the windows version.
felipelalli: ascii_butugychag, yes. my first idea was make this 100% web, but the IRC let me crazy. So I had just two dawns to implement it, I choose the cheapest for my good will and the win version is because they demanded it
felipelalli: jurov, they are learning the importance of the thing, at least, I hope. They are learning the importance to make solid contracts and verify the people reputation.
BingoBoingo: <ascii_butugychag> when do we get 'reactor controlz for the lazy' << In Japan. Turns out Fukushima was running off of one of those dollar store Sharp organizers.
PeterL: Do they not have real computers in Brazil?
felipelalli: jurov, I understand. But it is being very useful so far in Brazilian community. Actually the app can sign messages (not verify it yet)
ascii_butugychag: when do we get 'reactor controlz for the lazy'
trinque: jurov: nah, deedbot is still jammed until I get trb to send the next bundle
trinque: speaking of which, thanks to the helping hand of the other republic nodes, deedbot.org is a fully synced trb public node
assbot: Logged on 03-02-2016 17:01:22; ascii_butugychag: does nobody but me actually watch their trb logs ?
ascii_butugychag: aha this was the first major 'pgp is broken beyond repair' thread, iirc.
jurov: phf this was tried several times but since the hash in the signature is done with a date and whatnot,
phf: i was trying to see if could look up the corresponding vpatch by decoding the sig's hash, and then assoc into a precomputed map of vpatch hashes. unfortunately openpgp concats the sig header to the payload, so can't precompute :/
mod6: ascii_butugychag: here's another look at the difference between both via sha512 hash diff: http://dpaste.com/0SAENPF.txt
phf: so there's two sig's that are not sha512. genesis.vpatch.trinque.sig is sha256 and polarbeard_add_getpeerinfo_rpc.vpatch.polarbeard.sig is sha1
shinohai: "This also means that Gavins assumptions/estimates regarding the usage for 20 MB blocks unfortunately were very wrong:" https://redd.it/440cq6
PeterL: ascii_butugychag> as that would require the hash of (-b) and (c') to be equal << is it looking at the hash of the pressed value, which should be the same, or the hash of the patches themselves, which I understand are quite different?
assbot: Understanding Darcs/Patch theory - Wikibooks, open books for an open world ... ( http://bit.ly/1Kqo5u2 )
punkman: ascii_butugychag: this looks useful when regrinding things https://en.wikibooks.org/wiki/Understanding_Darcs/Patch_theory#The_complex_undo_revisited
assbot: Logged on 03-02-2016 18:10:43; punkman: https://en.wikibooks.org/wiki/Understanding_Darcs/Patch_theory#Commutation
phf: (peace be upon them
ascii_butugychag: because they aren't tmsr and do not have our axioms.
ascii_butugychag: 'To understand commutation, you should understand why we cannot keep our original patches, but are forced to rely on evil step sisters instead.' << seems like they solved it in the PRECISELY most-anti-vtronic way.
assbot: Understanding Darcs/Patch theory - Wikibooks, open books for an open world ... ( http://bit.ly/1UKKpPF )
PeterL: wouldn't they be equal?
ascii_butugychag: as that would require the hash of (-b) and (c') to be equal
punkman: don't we have to be careful with the antipatches though, so as not to introduce cycles in the graph?
mod6: I'll probably just send the patch to the ML to recruit people to help me test over the next two weeks.
mod6: in other news, ive got a patch for V that's not only the fix for the post-press hash checking, but also for the problem when running the graphing tool with a node that has no decendants. and a couple of small cleanup things.
ascii_butugychag is still not sure why it was necessary to discard the history and compress the old patch + its fix into a new one
ascii_butugychag: looks like they are
mod6: are they not identical?
ascii_butugychag: what i am expecting to see in a correct press is for the trees to be IDENTICAL
ascii_butugychag: phf: not so much unpressable, but that ~you~ lost the ancestors
mod6: i guess that makes sense you wanted to see it that way - i thought the first way just because that way there ~is~ a diff to look at. but maybe this is better.
ascii_butugychag: your local one consists of all of the ones you downloaded for which you have the signatures you accept as valid
ascii_butugychag: but the global, imaginary tree contains all vpatches ever authored
ascii_butugychag: every patch ever written ~is in the tree~ as it appears to archaeologists
polarbeard: oh, now I see the complete picture
ascii_butugychag: and the whole point of it.