312300+ entries in 0.198s
phf: mod6: lisp is not always (and not usually) interpreted, i don't
think
there's anything wrong with
targeting jvm, except for
the fact
that jvm itself is not good
mod6: why would anyone do
this?!
mod6: ugh. so wait, instead of a general lisp interpeter its compiled into jvm byte code? what abortion is
this?
mod6: what is
the black magic behind
the scenes? is it not based upon a lambda calc?
ben_vulpes: keep running into
the most inane shit getting it running
tho
ben_vulpes: trains, bikes, conferences, may even have some sort of conversation piece done at
that point
mod6: been
trying
to plow
through sicp and
that compiler book. and ada.
ben_vulpes: going up
to clojure/west
to acquire a few more problems
mod6: im
tinkering around with lexical parsing with scheme for a sec here.
ben_vulpes: i've been wondering where
that patch got off
to
mod6: hey, im just posting
this here so we don't lost it --
this is from funk_ iirc, he never sent it
to
the list -- i'm
told
that it "works" but have never
tested it myself:
http://dpaste.com/2WTCSHV.txt gernika: If asciilifeform ever accidentally ends up at
the same hotel as a rails convention, I expect
to see a lot of shattered
teeth on
the floor.
jurov: asciilifeform: ver well, so if you were
tasked
to do
that "family" function you'd started
to kick out
teeth around?
jurov: "In my experience, software bloat almost always comes from smart, often
the smartest, devs who are
technically
the most competent. Couple
their abilities with a few narrowly interpreted constraints, a well-intentioned effort
to save
the day.."
phf: err, overflows,
though float errors are another one
phf: like
the attempts
to
teach
type system about integer overfloats
phf: haskell's answer was
to pretend like von neumann machine doesn't exist, a goal at which it failed in various interesting ways
phf: lisp's answer is given von neumann what's most flexible model, ada's answer is given von neumann how do we put enough constraints
that
there are runtime reliability guarantees
phf: fixed width natural numbers are implemented at bedrock, so in order
to support anything else you have
to create an abstraction, keep it consistent with
the rest of language features, etc. so you either get overspecialized languages like apl (with
their runtime costs) or monsters like haskell
☟︎ phf: mircea_popescu: i
think it's because whole
thing is various degrees of successful abstraction on
top of von neumann machine
mircea_popescu: even numbers and natural numbers are EXACTLY as anything as
the other one.
mircea_popescu: 2) why
the fuck exactly do you
think you can get 3 as 2+1 in "natural" space but you don't see
that you get 4 as 2+1 in "even natural" space ?
mircea_popescu: and now it
turns out
that + and - are in
the same exact situation.
mircea_popescu: then we discover
that > is nevertheless defined, in spite of EXACT EQUIV OF MOD
mircea_popescu: asciilifeform i
thought "mod is not defined because bignum"
trinque: davout: no way in hell I'd run such a
thing without it involving
the airgap dance.
mircea_popescu: how are even numbers any less "contiguous"
than say natural numbers ffs.
mats: thanks for all
the fish
mircea_popescu: "While range specifications are very useful,
they are also highly limited in
their usefulness. Ada ranges must always be contiguous.
This means
that you cannot specify an integer data
type of only even numbers, for instance. Even numbers are not contiguous."
mats: i initially started with ~15btc in my coinbr piggy a little more
than
two years ago, deposited another 15 a year later.
today, i'm walking away with 60btc in profit and a 30btc stake in s.nsa
mircea_popescu: asciilifeform can you explain
the concept of "contiguity" as seen in
this ada spec discussion ?
mircea_popescu: "o noes, who will we sue when
they
turn out
to be full of piss!"
mircea_popescu: i don't see hospitals using bags of "saline"
that come with a "warning : not suitable for use" on
them.
mircea_popescu: the
thing clearly says "no guarantee not even for fitness
to a particular purpose". if you use such items in your "safety engineering" you're liable.
mircea_popescu: how
the fuck are you going
to introduce payment for, eg, deeds ? and if you do, how is it going
to beat $pay deed.
mircea_popescu: and don't fucking start with
the dc current scandal all over again!
mircea_popescu: asciilifeform yes, i mean so
that
trinque gets experience with castles, sure.
phf: davout: easiest option is
to always put an iso8601
timestamp on
things, 2016-04-06T23:59:00Z or somesuch
davout: asciilifeform: no you don't it, so we can
tip each other!!1
davout: so noon GMT on
thursday it shall be
mircea_popescu: diana_coman maybe by
the
time
that's on
the plate, we actually change in-game chat
to gossipd.
davout: re a better specified auction deadline, i
think it's best
to precise it
to "the end of wednesday
the 6th of april, in whatever
timezone
this event happens last"
mircea_popescu: someone pretending
to do dev work and not being in wot is ipso facto saying "i am a fraudster
trying
to sell my shit as software".
mircea_popescu: for one
thing,
those people are schmucks. but even if
they weren't schmucks, not in wot.
mircea_popescu: go
through
the logs,
the
thing is specced in
the prev discussion with
trinque.
mircea_popescu: diana_coman well in game-stuff i ideally want
to change
the chat
to irc, but hey.
diana_coman: ah, you mean auctions and reporting in chan, I was
thinking in game stuff
diana_coman: bwahaha, is
that any eulora player in particular?
mircea_popescu: im currently
trying
to get an eulora player
to implement proper auctions and reporting, but it's slow going.
davout: apperently, according
to condesk it got shut down after a 15 btc
theft
davout: asciilifeform:
there was, a long
time ago
davout: ;;later
tell pete_dushenski you have a point, i did not specify a
timezone for "the end of wednesday
the 6th of april"
davout: i'm pretty sure we'll look back five years from now and witness pretty much
the same
thing
mircea_popescu: but yeah, any indignation at how retarded people are is readily
tempered by memory of how fucking insane btcland was five years ago.
PeterL: nafario asked me if I wanted
to come
to some conference
to
talk about how
to use
the
thing
davout: nefario asked me if i wanted
to write
the
thing
davout: at
that
time i was like "i have nfi what
this glbse
thing is supposed
to do, maybe i should do drugs instead"
mircea_popescu: but
the way mpoe happened was
that basically i
threw up
the idea in chan and a bunch of
the people
then-active commented and so on.
mircea_popescu: i don't rightly recall what it was, and i'm not even sure i was yet keeping irc logs at
the very early
time all
this happened.
mircea_popescu: anyway, guy got in faq because iirc he found some
typo or something.