416700+ entries in 0.275s
trinque: or barring
that seems like one must simply be chosen and
that's
that
trinque: I mean in
the case of
the hash, implicitly so
trinque: and
then yes, perhaps
the sig validates
that information
trinque: you must at least know what hash
to
try, yeah?
ascii_field: what if i don't want
to hear so much as one byte if sig isn't from l1 ?
trinque: with maybe
the hash used and byte length preceding, actually
trinque: just put
the sig immediately following
trinque: what I mean by header is perhaps no more
than: -----BEGIN PGP SIGNED MESSAGE-----
punkman: you --sign --armor
the detached sig :P
jurov: punkman and how does
that help with verifying
the sig as soon as posible?
trinque: ascii_field: yeah, why not? you send me a GET, I first
tell ya how many bytes
to expect,
then signature,
then body?
ascii_field: punkman: didja miss all
the 'clearsigning is retardedly broken'
threads ?
☟︎ trinque: sig comes over as part of header,
then body
trinque: I mean were www sane it would work precisely
that way
ascii_field: punkman: whereas 'sign
the sig' again means
that you can't
take a single file at a
time.
ascii_field: 'take any sig and
then see' creates a ddos vector
punkman: ascii_field: well you can sign
the detached sig if you please
ascii_field: say, a public www where
these can be dropped in a hopper
ascii_field: trinque: sometimes one wishes
to eat
the sig prior
to
the payload
trinque: ascii_field: maybe in
the case where
the payload was munged in regards
to whitespace or
the like ?
trinque: danielpbarron: heh me
too re: ren & stimpy
ascii_field: (i could explain why one may wish
to do
this, but will leave it as an exercise)
pete_dushenski: lest i
try
to piledrive my younger brother into
the concrete sidewalk or something
pete_dushenski: danielpbarron
the only show i wasn't allowed
to watch was wwf wrestling
ascii_field: so it is quite impossible
to verify
that it is valid per se, without having
the payload
ascii_field: btw here's another
thing
to add
to
the 'gpg is mega-retarded' list: gpg sig does not contain a sig of self !!
danielpbarron: and I "wasn't allowed"
to watch ren and stimpy (had
to sneak it)
☟︎ assbot: Logged on 13-08-2015 05:04:30; pete_dushenski: mircea_popescu: i'm
taking notes here on cartoonz for kidz, so keep 'em comin.
ascii_field notices
that he has been using 'vdiff' and 'vpatch' interchangeably.
this is not really an error.
danielpbarron: PinkPosixPXE would be
the exception
to
that l1 only
theory
ascii_field: punkman: perhaps it does? but iirc
the only contributors
thus far live in l1
ascii_field: and, for
the first
time ever in history of versionatrons, you can
take a
tree of unknown constitution and determine what patch set it corresponds
to (if any)
ascii_field: both vdiffs and sigs (the former by virtue of having one or more of
the latter matching) are immediately attributable
to an l1 pubkey
danielpbarron: i noticed
today
that you don't need
to rename
the files in order for
them
to verify; you can sha1sum and gpg --verify
the downloaded attachments as
they come, and go straight
to extracting
the
tar afterwards
ascii_field: mircea_popescu:
the importance of
the
thing i intended, roughly, is
that ~everything is either a vdiff or a sig~
jurov: if your email client is known not
to mutilate
text, put clearsigned
text into
the body instead of first attachemnt
mircea_popescu: jurov cool, ima
try
this and bitch more if i can find a way
to.
jurov: mircea_popescu: only clearsigned
text is mandatory,
there can be none or any number of further attachments(not necessarily only patches), each with detached attachment.name.ext.sig
mod6: hanbot is
testing out my i686 changes for buildroot/rotor.sh - even got a script for it. so
that's p-cool.
mod6: wth is wrong with me
this afternoon.
ascii_field: mircea_popescu:
third, if
the second exists - a detached .sig for it
mod6: <+pete_dushenski> really, if
this wasn't obvious
to mod6, ppl like hanbot and i have no chance <<
this probably should have been obvious
to me. but sometimes, hard
to see
the forest
through all of
the
trees. i've been looking at
the compare diff a lot lately, and forgot
that /all/ is contained in
the genesis.diff.
ascii_field: mircea_popescu:
the actual patch (optional, if
the message concerns a patch) - 2nd attach
ascii_field: mircea_popescu:
to use jurov's
thing, you need 3
things: a clearsigned readme (1st attachment)
jurov: you mean
to btc-dev?
mircea_popescu: speaking of which, who wants
to hold my retarded hand while i do
that again ? i forget, what do i do, send an email with
the plaintext as body and
the detached sigfile as attachment ?
mircea_popescu: ima
try
to a signing statement once i put it
to
the ml
mircea_popescu: i kept meaning
to do a statement explaining
the major importance of
that
thing but i didn't really find my words with
the deed.
mod6: just ran
the `patch -p1 < genesis.vpatch` in an empty dir, and
then volla, bitcoin source exists
pete_dushenski: really, if
this wasn't obvious
to mod6, ppl like hanbot and i have no chance
mod6: err, so, use my newly created genesis.vdiff, apply
to pulled original v0.5.3 ?
mircea_popescu: thin priviledge is hanging out with friends, pointing out
that i'd be more sensible
to not give citizenship
to children of fat people
than
to children of illegal immingrants, and not offending anyone.
ascii_field: (which won't match because it has
timestamps, but ~order~ will match)
mod6: yup, ok. and when I have my own genesis.vdiff, should compare with `diff -u`
to
the one you posted in your original email? or compare with vdiff?
mod6: and
try
this without LC_ALL=C ?
mod6: I can say
that your change
to vdiff.sh `diff -uNr $1 $2 | awk 'm = /^(---|\+\+\+)/{s="sha512sum \"" $2 "\" 2>/dev/null " | getline x; if (s) { split(x, a, " "); o = a[1]; } else {o = "false";} print $1 " " $2 " " o} !m { print $0 }'` certainly removes
the
timestamps from
the output hashes.
assbot: Logged on 21-08-2015 20:57:40; ascii_field: mod6, ben_vulpes, mircea_popescu, et al: here is an experiment: change ' print $0 " " o ' in vdiff.sh
to ' print $1 " " $2 " " o '.
this results in
timestamp amputation.
the patches work fine, except
that gnupatch is no longer aware of preexisting files and will happily create
the same file
twice, if so run, glomming on
top of itself.
assbot: Logged on 21-08-2015 01:22:45; asciilifeform: mircea_popescu:
try
this on for size:
assbot: Logged on 20-08-2015 16:56:49; ascii_field: and is our latest discovery
that 'diff' is UNFATHOMABLY retarded ?!
assbot: Logged on 21-08-2015 03:25:47; asciilifeform: iirc hanbot also had
the out-of-order
thing with genesis
thestringpuller: so as soon as I figure out
the wp
thing, i'll have a dedicated blog
trinque: which is why
today's cock-tease of a market slump does me no good. come on baby, lets have
the full rip
ascii_field: trinque: it is in
the initial stage, of cementing
the precedent, via jailing an already unpopular fella
pete_dushenski: ;;later
tell ben_vulpes and how is
the lovely lady v
these days ? staying active i
trust ?
trinque: ascii_field: it is abundantly clear
to anyone
thinking
that
this is a runaway bureaucratic process with no clear intentions/goals/etc, just *must feed*
trinque: pete_dushenski: and
to you as well
assbot: Logged on 26-07-2014 18:48:15; asciilifeform:
this is why i simply don't get people who continue
to
think of
their relationship with
the u.s. court system in legal, rather
than military,
terms.
hdbuck: question why would usg need our browzer history via our computer since
they already have it all via google
ascii_field: poor schmuck failed
to destroy
the evidence of destroying
the evidence..
ascii_field: the one law
that nobody ever cancelled is 'not caught, not
thief'
trinque: so we go from
the 5th amendment directly
to "you must investigate yourself on behalf of
the all powerful state"
trinque: presumably any piece of information can be used in any imaginary investigation of
the future