263200+ entries in 0.172s
phf: trinque: i
think
that
the
timing is just overlapping, i'll have
to grep for you wget's specifically, but
the day is basically an endless stream of /log/2016-05-22&docid=nDn56WVkvV-86M&tbnid=wQqrCY5i08s2qM:&w=245&h=244&hl=en-US&source=sh/x/im and /log/?date=01-05-2014%27%20or%20(1,2)=(select*from(select%20name_const(CHAR(111,108,111,108,111,115,104,101,114),1),name_const(CHAR(111,108,111,108,111,115,104,101,114),1))a)%20--%20%27x%27=%27x
shinohai waves
to phf, hopes he enjoyed vacation.
trinque: that actually ended up bringing
the bot back,
then was bouncing on/off, so I slowed
the crawl
trinque: after it went away for a while I did a wget -R on
the www as a backup, hoping
the www wouldn't also disappear.
phf: i'm sorry for failing, whoever put
this nailgun in my bed, congratulations.
there were no similar spikes in
the past, and
the attack happened precisely on
the day i left for vacation, which was publicly known. given
the swift and strong reaction i'm not quiet sure i'm even welcome on
the channel anymore, which sucks.
phf: attempted
to connect for
the duration of attack, which in
turn
timed out for
the rest of
the day. i'm not sure how i could've mitigated
this attack, short of running bot and web on separate machines, with bot's ip hidden.
phf: hmm, just returned into civilization. so before leaving for my vacation i replaced
the bot with
trinque's reconnecting version, i haven't
touched
the server since, so obviously reconnecting worked. for 29th specifically i see a flooding spike of bogus
http request, which presumably saturated
the digitalocean connection, because ping/pongs were still going
through, until only pings were being sent, until bot would
timed out.
then it
mircea_popescu: incredible example of "smarts don't help
the castrated"
mircea_popescu: speaking of which : one of
their latest episodes was sheer genius.
they completely deconstructed
the sjw + advertising bullshit. and
their conclusion ? you know what it was ? WE WILL HAVE
TO DO A BETTER SJW!!!!
trinque: the claim
that china cannot do anything involving what's currently understood as "computing" is not nearly
the same one as "can create asciilifeform's personally designed diamondputer"
mircea_popescu: i can't
think of one historical example where
they did.
trinque: asciilifeform: I do not claim
to know, but hesitate
to say
the major industrial powerhouse of earth lacks understanding of what it does.
mod6: ah, mircea_popescu, maybe i'm wrong
then; in
that sense of "models"
mircea_popescu: whether
they find a real root in between
these or not is up in
the air.
mircea_popescu: anyway. it's evident
that
the cray which
the us used (ie, max centralization) doesn't work for
this. arguably it didn't even work for
the us, in 1950. it's vaguely clear
that per-rocket chips as currently used (ie, max decentralization) also doesn't work - for obvious computational reasons.
trinque: that
thread
too was "already had"
shinohai: " Sorry, SMF was unable
to connect
to
the database.
This may be caused by
the server being busy. Please
try again later. " <<< perhaps
theymos should reconsider
taking you ad revenue in order
to buy functioning servers.
mircea_popescu: trinque i discussed all sorts of
things ; including "buy a boat" about a month before
the world was "shocked" by a certain bankruptcy. not
that anyone notices,
to my eternal amusement.
trinque: mircea_popescu: iirc you discussed
this area denial strategy a year or
two ago
mircea_popescu: this is a project roughly of
the style
that created
the us bomber command after ww2 victory.
mircea_popescu: that
they want
to pilot
ten billion rockets in flight over
their geosphere.
mircea_popescu: trinque
the big state secret here, and
the problem
the chinese are spedning about five
trillion over
twenty years
to solve is
mod6: <+mircea_popescu> anyway. point being
the situation's windy rather
than calm. << agreed.
mircea_popescu: anyway. point being
the situation's windy rather
than calm.
trinque: loongson is
there if usg.intel fails
mircea_popescu: it's just
that little meaningful can be said about any of
them. and im sure not a complete lsit.
mircea_popescu: mod6
they actually have multiple discernible alt models. one we even discussed in logs, ifyou recall. alf, much in character, hated it etc.
mod6: i very much doubt
this is
their long
term strategy.
mod6: they just need
to stab "wintel"
mod6: well,
they probably don't need a new model of computation.
mircea_popescu: the macro problems is
that it's not altogether clear which side of what will end up facing where.
mircea_popescu: it's pretty impossible
to make predictions of
the scale required here.
mircea_popescu: asciilifeform
the
thing here, which of course isn't visible from english, because anglopress is made by niggers and retards, and so forth :
mircea_popescu: even if i could spend a million engineer hours a second, brought from another galaxy, i would not commit
them
to making me a sane compiler+kernel by friday.
trinque: asciilifeform: lemme restate
that
the "stuckness by sheer mass, age, etc" of gcc brings
to mind a giant's dead body
that, before it decomposes, maybe serves as shelter or warmth
mircea_popescu: but
the worst part about all
this, and
the
true reason i'm inert on
the
topic, is
this :
the situation is evidently unhinged, and changing much faster
than
the sort of process which is involved in making a software stack, compiler kernel et all.
mod6: This sort of leads us back
to
the original question:
to Ada, or not
to Ada.
mircea_popescu: but in
this sense, merely writing c code is
the sin ; compiling it in
the nazi flavour of socialism or in
the soviet flavour of socialism, llvm vs gcc, is sort-of like, what flowers
to leave on
the raped body, roses or
tulips.
mircea_popescu: the problem here (as well as above, with llvm) is
that without
the strongarm
to kill sipa / force obedience from hardware vendors etc, you will never be able
to make a compiler
that works
a111: Logged on 2016-09-05 20:28 mod6: not
that I'm saying "i'm going
to write a
tmsr compiler
this year."; however, seeing
the need for
this years ago, having one written by us
that we can
trust, is a worthwhile project. i've
taken steps in
this direction by doing some very much required reading on
the subject.
trinque: and in
this, GNUshit is an ally of Microshit
trinque: it may very well have reduced
the market price of building software
to *below* what could sustain actual software companies
trinque: asciilifeform: I could as easily say
this whole "free software"
thing ruined
the industry and start from *there*
mod6: help africa, give
them fb!
trinque: I would go
to church if I wanted
to make decisions based on
that kind of argument.
mircea_popescu: asciilifeform his idea is
that any software is made of discrete software-states, and if he doesn't like something definite later on he just freezes and forks.
a111: Logged on 2016-09-05 20:21
trinque: seems
technically excellent and
the only criticism has been "apple made it" of which, nazis made
this and
that modern medicine, what
a111: Logged on 2016-09-05 20:27
trinque:
to ben_vulpes link, it does not diminish my regard for openbsd
that it is moving
towards llvm. considering solely
the political aspect,
they've previously used
their own old (and maintained) fork of gcc.
trinque: I would not be surprised if
they forked llvm if
they ever needed.
mod6: not
that I'm saying "i'm going
to write a
tmsr compiler
this year."; however, seeing
the need for
this years ago, having one written by us
that we can
trust, is a worthwhile project. i've
taken steps in
this direction by doing some very much required reading on
the subject.
☟︎ trinque: to ben_vulpes link, it does not diminish my regard for openbsd
that it is moving
towards llvm. considering solely
the political aspect,
they've previously used
their own old (and maintained) fork of gcc.
☟︎ mod6: but better
the devil you know...