256100+ entries in 0.174s
BingoBoingo: <mircea_popescu> are
there any stations even left
that do 100kw in soviet us ? << 50kw is
the cap and requires "clear channel" Daytime range of such power not
too impressive
PeterL: could relay messages from one gossipd node
to another
jhvh1: asciilifeform:
The operation succeeded.
shinohai: neat asciilifeform ... I haven't sourced one of
the others yet :/
Framedragger: PeterL: ah, good point. and regarding first packet, i wonder if some very light lighthouse version would make sense, a pre-shared sequence of whitelisted nonces between
two given nodes, whatnot. in
the beginning, interchangeable over internet. and later, updateable over same radio channel ("here's my payload, and also please add
to your whitenoncelist
this encrypted sequence of nonces for future channel initiations.")
jhvh1: asciilifeform:
The operation succeeded.
shinohai: Let's hope not, unless you like scams
too.
shinohai: Exactly who I
thought of PeterL lol
PeterL: hi burnsd03, who are you? name makes me
think of burnside, wasn't
that
the btctc guy?
burnsd03: mircea_popescu could you please make me an account? Just following
the quick-start guide :)
PeterL: not sure if
that makes sense with what asciilifeform was
talking about?
a111: Logged on 2016-09-22 14:46 asciilifeform: you have a nonce in
the packet, and
the packet only registers as valid when said nonce is equal
to hash of some packet
that is in
the receiver's current buffer.
Framedragger: asciilifeform: you mean an initial packet after a period of silence? hm yeah; one way would be
to have a scheme in place which i
think you may have suggested anyway:
to have stations constantly stream packets at some rate, 24/7.
Framedragger: maybe i'm not getting
the picture how quickly entries in
the buffer can expire,
though
Framedragger: same exact packet - he would assume
the nonce in packet refers
to a packet
that is still in
the receiver's buffer
Framedragger: asciilifeform: ah, right, i forgot everything's signed, lol. nice! i guess enemy could replay for a (choosably) small window of
time, some degree of flooding possible? but
this could be minimized and
that's
the point, right?
shinohai: burnsd03:
this is
the chan where you register your key
Framedragger: asciilifeform: i wonder, could
this not be predicted by
the enemy? (value of nonce - say
the enemy is aware of
the (likely) packets received by
the receiving end.) i'm not sure, just
trying out all possibilities in
the head, because otherwise sounds nice
Framedragger: indeed, indeed. and yet incorporating a lighthouse lifts
the scheme from pure elegance
to additional complexity. but maybe
that's not
to be avoided, and such is life
Framedragger: nice. what of DoS via replay, asciilifeform ? or should
this get ignored? i.e. enemy just floods with identical packets (so both checksum and sig ok), and
those get redistributed
shinohai: YAgi's are cheap and easy
to make would
that work ?
mircea_popescu: somewhat less conclusive, but more expensive. still, it may
turn out very interesting
mircea_popescu: either because lazydumb or because afraid someone might microscope/probe
the
things and find it.
mircea_popescu: asciilifeform
tru, but dollars
to donuts
they didn't change
the magic for each chip.
mircea_popescu: thestringpuller ideally you want
to do it all in ram and dump
to disk 2gb at a
time
thestringpuller: "The main cost in
the initial synchronization is disk I/O when seeking for
transactions in
the blockchain. Default cache size (dbcache) is 100MB, and it is
too low. I always set it manually depending on
the amount of RAM in
the system (5GB for dbcache on a 8GB system)." << re: discussions of using SSDs on initial sync.
mircea_popescu: the companion
to Framedragger 's exploration
to be done at home is evidently -
there's only 3 chips, right ? get a realtek, an intel and a broad and fuzz
them
mircea_popescu: word ; i was
thinking of proposing exactly
this but
then i got distracted.
a111: Logged on 2016-09-22 12:20 mircea_popescu: you have a reasonably fresh list of respondent boxes, feed
them 0x32 at offset 0x47f
then poll in 5 and 500 seconds see if
they're up.
a111: Logged on 2016-09-22 12:28 mircea_popescu: by now it's a 5 year old
thing, if
they were gonna do something about it
they've had done by now.
Framedragger: need
to minimize energy expenditure at $job. "go home
tired" sort of case, it's sad. i'd like
to, e.g., among other matters (incl mkj log etc of course), write a very initial very
to-be-criticized gossipd for udp/ip, using (kill me now) python
twisted. i guess many live like
this, savouring free
time, with personal projects falling behind. [i rant
today, because exhaustion + coffee infusion => wee bit psychedelic]
Framedragger: mmm. i2c interface, etc. yeah.
this is a very neat useful
thing.
mircea_popescu: ballas' celebrated "so you feel strongly about something, strong enough
to not do anything" is just as much a statement.
there's no lack of observation, it's just
that piggy likes his
truffle.
mircea_popescu: it's very much a destructive psychiatric condition ; harmful for self, harmful for others. smoke and don't do
this, rather
than don't smoke and do it.
mircea_popescu: where
they
take
turns
to play
the obese and
the enabler for each other.
mircea_popescu: the fundamental point of reddit & co, and
the main reason
to avoid
them, is
that
they format people into
this specific mode,
mircea_popescu: the only sane way
to look at it is over BingoBoingo 's mayo-enabler pair.
Framedragger: mircea_popescu:
the sad part is
that by "battling out" ideas in
this wordspace, one gets a bit of serotonin rush,
to feel *as if*
they have actually done something material. i know i get
this, at least. evolutionary side-effect,
this whole "curiosity -> completion" gap used
to be much larger, not it short-circuits easily. it sucks, need
to make conscious effort
to resist sinking into nihilistic bliss, i
think. </rant>