►
From YouTube: The Perfect Programming Language - CodeConf 2016
Description
Presented by Michael Bernstein
After years of searching, I have found the perfect programming language. It is capable of anything, and is all things to all people. Unfortunately, I'd have to recommend that you don't use it. In this talk, I'll share the language with you, describe its design and history, and explain why using it is a terrible idea.
About CodeConf
CodeConf improves the software community by providing a forum for thought-provoking talks and forging social connections. The fourth installment of the CodeConf series took place in Hollywood in 2016. This year's event focused on systems engineering projects, practices, and programs in the open source community.
For more information on this year's CodeConf, go to:
https://codeconf.com/
A
So
hi,
my
name
is
Mike
Bernstein,
and
this
is
a
talk
that
I
wrote
for
this
conference
called
the
perfect
programming
language
I'm
very
excited
to
be
here,
giving
the
opening
talk
of
code,
comps
2016
I've
been
a
part
of
the
github
and
open
source
community
for
a
long
time
and
when
github
started.
Putting
these
conferences
together.
I
always
hope
that
I
would
have
an
opportunity
to
speak
at
one
and
Here.
A
I
am
and
I'm
in
fabulous
Los
Angeles,
which
is
a
city
that
I
love
and
even
though,
as
a
born-and-bred
on
New,
York
or
I
was
raised
to
distrust
Los
Angeles,
but
that's
okay,
I
got
over
that
and
yeah
I'm
super
stoked
to
be
here.
Thanks
for
everyone
for
having
me,
and
for
you
know
it's
all
of
you
for
coming
a
little
bit
about
me.
I
work
for
a
company
called
code
climate,
we're
a
hosted,
static
analysis
platform.
That's
dedicated
to
making
it
easier
for
teams
to
write
high
quality
software
together.
A
Ask
me
for
more
information
at
the
conference.
If
you
see
me
and
I've
had
sounds
interesting
to
you,
I'd
love
to
talk
your
ear
off
about
it,
something
that
I
spent
a
lot
of
time.
Thinking
about
like
Chris
mentioned
I
also
recently
finished
co-authoring
a
book
about
making
pizza
at
home
sets
of
printers
right
now
you
should
buy
it
going
to
be
shipping
in
a
couple
weeks
and
it
will
change
your
life
and
I'm.
An
amateur
ceramicist
and
Potter
I
make
stoneware
pottery
at
a
studio
in
Northeast,
Washington
DC
near
where
I
live.
A
It's
one
of
my
favorite
pastimes
couple
pictures
of
that
stuff.
So
all
of
this
is
to
say
I'm,
clearly
eminently
qualified
to
be
addressing
you
about
the
perfect
programming
language,
and
so
so
here's
here's
the
story,
I'm
going
to
start
the
story.
Now
we
don't
really
have
a
lot
of
time
and
I've
really
been
looking
forward
to
telling
this
to
you.
So,
let's
get
started
if
this
story
begins,
as
a
lot
of
my
stories
do
with
me
digging
through
a
pile
of
junk,
because
I
spent
a
lot
of
time
digging
through
piles
of
junk.
A
If
there
was
like
a
fourth
thing
on
this
slide
up
here,
it
would
be
just
like
a
pile
of
junk,
but
that
doesn't
photograph
very
well,
so
I'm
just
going
to
tell
you
about
it.
And
anyway,
one
day
I
was
digging
through
a
pile
of
junk
and
I
found
one
of
my
favorite
things
to
find
when
I'm
digging
through
a
pile
of
junk
and
that's
a
box
of
records.
A
It
was
a
notebook
and
inside
that
notebook
were
things
that
I
really
three
relays
really
really
really
didn't
expect
to
find,
not
to
mention
that
in
this
box
of
Records
was
not
a
normal
box
of
records
by
any
means
when
you're
used
to
digging
through
junk,
like
I
am
a
normal
box
of
records
that
you
find
is
usually
terrible
and
mostly
composed
of
very
old
unplayable
classical
albums.
But
this
particular
box
that
I
found
on
this
fateful
day
included,
in
addition
to
a
lot
of
other,
very
high-quality
jazz
albums.
A
One
of
my
favorite
records
of
all
time
look:
this
looks
fabulous
on
this
big
screen.
This
beauty,
a
1968,
ESP
disc
issue
of
some
rods,
heliocentric
Worlds
of
Sun
raw
Volume
one,
and
on
the
back
cover
there,
you
can
see
a
reproduction
of
the
original
cover
of
the
original
1965
issue
of
the
album,
but
that's
a
story
for
a
different
time,
but
I'm
really
excited
to
be
showing
this
to
you
on
the
giant
screen
here.
So
so
that's
a
little
bit
of
an
aside
back
to
the
notebook
in
a
lot
of
ways.
A
This
notebook
was
an
ordinary
notebook.
One
of
those
like
marble
cover
jobs
like
iYou
sees
in
middle
school
and
but
as
soon
as
I
started
to
look
a
little
bit
more
closely
at
it.
I
knew
it
wasn't
a
typical
notebook
because
I
pulled
it
out,
and
this
is
what
it
said
on
the
cover
it
read
and
neat
handwriting
the
perfect
programming
language
by
jpn.
A
Oh,
so
you
can
imagine
my
astonishment
here
as
someone
who
is
both
into
some
raw
and
programming
languages,
it
seemed
beyond
unbelievable
that
there
was
someone
else
in
the
world
who
was
interested
in
both
and
I.
Felt
I
had
found
a
kindred
spirit
in
this
mysterious
JPL,
so
I
opened
the
notebook
up
and
residents
curious
dedication,
which
was
written
in
a
language.
A
I
didn't
recognize
at
first,
so
I
pasted
it
into
Google
Translate
I,
don't
know
how
to
really
read
this
I
think
it
says
that
it
keeps
out
all
four
his
son,
Raj,
McCarthy
and
so
I
pasted
it
into
Google
Translate,
and
it
came
back
with
this.
It
was
apparently
a
dedication
written
in
Esperanto
to
bore
hands,
son,
Ron
and
McCarthy
three
of
my
favorite
people
from
history,
and
this
is
starting
to
get
like
a
little
weird,
and
so,
since
not
everyone
will
know
who
these
folks
are.
A
A
He
was
unfathomably
prolific
and
original,
uncompromising
and
literally
not
of
this
world.
The
man
who
would
become
to
be
known
as
son
Ron
was
born
as
born
as
Hermann
Buhl
blunt
in
Alabama
in
1914.
This
is
a
photo
of
him
about
fifty
years
after
that
and
one
of
the
greatest
son
Roth
stories
tells
of
his
first
visit
to
Egypt
in
1971,
as
you
might
tell,
by
his
name
sunrise
really
into
Egyptian
mythology
on
his
band
called
the
orchestra
wanted
to
travel
to
Egypt,
to
play
concerts
in
front
of
the
Great
Pyramid
and
at
the
border.
A
When
he
was
asked
for
his
passport,
he
allegedly
handed
the
Egyptian
border
patrol
a
passport
which
purported
that
his
birthplace
was
satyrs
like
the
knot,
Saturn
Alabama
like
the
planet
Saturn.
So,
needless
to
say,
he
projected
an
Arab
family
in
this
authentic
enough
that
he
did
in
fact
get
admitted
into
Egypt,
and
his
van
was
able
to
perform
some
legendary
concerts
there.
I
guess.
A
It
stands
for
Esperanto
as
in
the
artificial
language,
so
it's
like
starting
to
make
some
connections
between
these
records
and
we're
dedication,
and
so
here's
a
shot
of
the
cover
of
the
first
record
that
Bernard
Stillman
esps
proprietor
released
on
the
label.
It's
called
me
current
Annette
parantha,
which
means
let's
sing
in
Esperanto
and
as
you'll
see
in
Tibet.
A
This
information
is
only
slightly
tangential
to
my
story,
so
the
second
dedication
in
the
notebook
what's
up,
Boris
Boris
was
an
Argentine
writer,
philosopher
poet
and
singer,
who
similar
leads
his
own
raw
transcended
categorization
in
a
lot
of
different
ways.
He
was
obsessed
with
many
different
subject
areas,
including
language,
and
particularly
when
I
saw
boor,
Hayes's
dedication
written
in
Esperanto.
A
Being
someone
who
has
read
a
decent
amount
of
Boreas
in
my
life,
I
was
reminded
of
a
story
that
bore
his
wrote
called
the
analytical
language
of
John
Wilkins
John
Wilkins
was
a
real
person,
but
in
this
story,
Jorge's
provides
a
mostly
fictitious
overview
of
Wilkinson's
philosophical
writings
about
a
universal
language.
Wilkins
wrote
long
tracks
regarding
very
logical
human
design.
Languages
which
were
meant
to
reflect
the
world
in
a
more
realistic
way
than
human
language
has
had
up
until
that
point.
A
A
That's
what
everyone
having
just
broken
the
water
pitcher
and
those
that
from
a
long
way
off,
look
like
flies
so,
while
hilarious,
it's
quite
clear
that
these
ambiguities
in
overlap
in
these
categories
would
leave
one
who's
seeking
logic
and
order
like
wanting
for
a
little
more
right
like
a
dog
right
like
which
one
are
they
right?
Are
they
they're
sort
of
frenzied
innumerable
in
etcetera,
for
example,
I
think
you
get
the
point
and
the
essay
ends
with
this
quote,
which
is
like
a
you
know.
A
Nineteen
fifties
hot
take
attributed
to
Chester
kid,
and
it
sums
up.
Where
is
his
view
very
nicely,
so
I'm
going
to
read
it
for
you,
he
knows
that
there
are,
in
the
soul,
tense,
more
bewildering,
more
numberless
and
more
nameless
than
the
colors
of
an
autumn
forest.
Yet
he
seriously
believes
that
these
things
can
every
one
of
them
in
all
of
their
tones
and
semitones,
and
all
their
blends
and
unions
be
accurately
represented
by
an
arbitrary
system
of
grunts
and
squeals.
A
He
believes
that
an
ordinary,
civilized
stockbroker
can
really
produce
out
of
his
own
inside
noises,
which
denote
all
the
mysteries
of
memory
and
all
the
Agony's
of
desire.
This
is
a
rather
poetic
takedown
of
the
idea
that
a
poem
quote
perfect
human
language
could
ever
really
exist,
and
so,
finally,
the
final
person
in
this
dedication
trio
was
John
McCarthy.
You
know
that
list
guy.
A
A
That's
the
face
I
make
when
my
daughter
asks
me.
If
she
can
watch
more
television,
I
don't
have
the
benefit
of
like
cool
robots
behind
me,
though
so
doesn't
always
work
so
well,
but
sort
of
an
appeal.
He
was
already
saying
again
and
then
here's
the
bottom
half
of
page
13
of
the
list
1.5
programmers
manual,
which
defines
the
core
functions
of
lists
that
are
known
as
eval
and
apply,
and
so
all
of
list
essentially
can
be
defined
in
terms
of
a
vowel
apply
and
a
few
other
primitives
Alan
Kay.
A
Who
is
one
of
the
inventors
of
the
small
talk
programming
language
famously
called
these
Maxwell's
equations
of
software,
saying
that
this
is
the
whole
world
of
programming
in
a
few
lines.
That
I
could
put
my
hand
over
so
K
was
really
fascinated
by
this
idea
that
something
so
massive,
like
programming
languages
could
be
contained
and
expressed
in
this
tiny
half
a
page
of
typeset
writing
acquainting
it
with
Maxwell's
equations,
which
do
the
same
thing
for
physics,
I,
think
I,
don't
know
anything
about
science.
A
So
at
this
point,
I've
got
J
pianos
box
of
records
and
the
notebook
in
which
they've
inscribed
the
perfect
programming
language.
It
felt
like
a
puzzle
that
only
I
was
meant
to
solve.
I
had
to
spend
some
time
piecing
all
this
together
and
it's
by
no
means
a
given
that
I've
interpreted
all
this
correctly
because
I
didn't
you
know,
have
the
benefit
of
an
actual
conversation
with
this
person.
A
Okay,
so
we're
going
to
get
into
the
language
a
little
bit
and
as
I
started,
to
study
this
language.
It
was
clear
from
the
beginning
that
this
was
something
special
right.
This
is
not
like
our
random
dashing
z'
of
a
person
who
had
not
really
thought
this
through.
This
is
something
that
someone
had
spent
a
lot
of
time,
thinking
about
and
I
felt
really
lucky
to
be
the
only
person
on
the
planet
who
had
ever
seen
it
before
as
far
as
I
was
concerned.
A
So
the
first
part
of
the
notebook
covers
the
mechanized
semantics
of
the
language
and
to
gaze
upon
this
operational,
semantics
and
backus-naur
form
seems
to
be
gazing
into
the
very
true
radiant
heart
of
computation
itself.
I
was
very
elegant.
You
can
tell
that
piano
was
heavily
influenced
by
McCarthy's
work
on
lists.
The
notebook
described
a
very
sophisticated
type
system,
one
that
could
express
everything
from
simple
types
to
dependent
types
using
sophisticated
algorithms
for
both
type
reference
and
type
checking
statically.
A
So
the
idea
was,
you
know,
capable
of
thing
you
add
when
you're
writing,
your
programs
tell
you
which
types
you
were
using
helping.
You
determine
after
the
fact
that
the
programs
that
you
wrote
were
actually
well
typed
in
this
under
this
type
system.
Additionally,
the
notebook
has
pages
and
pages
describing
a
flexible,
get
reasonable
macro
system
that
would
let
developers
produce
with
libraries
when
many
other
languages
would
require
changes
to
the
core
runtime
to
create.
There
is
a
built-in
abstraction
to
query
data,
allowing
business
logic
of
applications
to
appear
simple
and
declarative
and
declarative.
A
This,
which
is
an
interest
of
mine
or
programming
languages
in
general,
seemed
also
to
be
a
particular
interest
of
pianist.
It's
an
idea
that
is
heavily
reinforced
throughout
the
pages.
Many
of
the
design
decisions
seem
to
come
down
to
whether
or
not
a
programmer
could
write
clear,
side-effect,
free
order,
independent
code
in
this
programming
language
additionally,
piano
was
interested
in
concurrency
is
an
important
part
of
the
design
of
the
perfect
language.
A
So
I
just
started
to
call
this
language
like
the
perfect
language
capital,
P
piano,
wanted
programmers
to
be
able
to
write
sequential
code,
analyze,
resultant
programs
and
make
aspects
of
them
transparently
concurrent.
So
if
you
had
some
routine
that
you
wanted
to
do
in
a
distributed
amongst
a
number
of
different
cores,
taking
it
from
concurrency
to
parallelism,
you
would
have
the
ability
to
do
that
in
the
perfect
language,
and
some
of
the
aspects
of
the
language
were
surprising
to
me,
because
they
reminded
me
of
work
that
some
of
my
colleagues
are
studying
right
now.
A
Distribution,
for
example,
was
handled
by
a
key
set
of
abstractions,
a
robust
scheduler
and
the
idea
that,
similarly
to
concurrency,
you
should
be
able
to
quote
unquote,
take
regular
programs
and
distribute
them
at
will,
amongst
many
hundreds
of
remote
computers,
something
that
McCarthy
and
lists
people,
Carl,
Hewitt
and
the
etc
were
interested
in
the
50s
and
60s.
That
seems
to
be
a
thread
that
this
Jay
piano
was
interested
in
as
well.
A
The
perfect
language
has
facilities
for
object,
orientation,
there's
a
tasteful
amount
of
class
extending
and
a
concept
of
object.
Although
it
doesn't
go
too
far,
it
doesn't
manage
it
managed
to
avoid
getting
mired
in
the
pitfalls
of
sub
classic,
which
is
a
really
challenging
thing,
that
the
interaction
between
object-oriented
programming
and
sophisticated
type
systems
is
one
that
keeps
programming
language
researchers
very
busy,
even
even
in
2016,
but
piano
had
a
pretty
good
sense
of
how
he
wanted
these
things
to
work.
He
or
she
I'm,
not
sure,
and
so
at
its
core.
A
It's
that
you
could
start
with
a
simple
script
that
was
dynamic,
and
that
did
a
few
things
and
over
time
you
could
evolve
and
enrich
the
Lumbee
program
in
your
language
over
time
which
types
as
your
programs
with
types
as
your
programs
domain
concepts
solidified.
So
you
could
start
with.
You
know:
that's
an
argument
that
people
that
prefer
dynamic
programming
languages
often
make
write.
I,
don't
want
to
think
about
types.
A
This
is
truly
the
best
of
both
worlds
in
terms
of
dynamic
and
Static
typing
datatypes,
also
an
interest
of
mine
and
something
that
I
find
lacking
in
certain
programming
languages
that
I
use
available.
Abstractions
for
describing
data
types
are
very
flexible
and
sophisticated
in
perfect.
He
has
concepts
of
G,
ADT's
and
other
union
types.
Sometimes
things
that
allow
you
to
take
small,
primitive
types,
enrich
them
into
more
expressive
types
which
pertain
to
your
domain
model.
Things
like
that.
We're
really
really
interesting
stuff.
A
Additionally,
there
is
pattern
matching
built
in
to
the
syntax
and
a
bunch
of
different
places
in
the
language.
This
is
another
favorite
language
feature
of
mine
and
the
pervasive
pattern
matching
and
destructuring
allows
for
concise.
You
have
readable
code
in
perfect
with
little
boilerplate,
so
if
you
wanted
to
have
a
function
which
accepted
a
number
of
different
input
types,
for
example,
you
have
this
kind
of
nice
equational
way
of
describing
these
functions
that
are
as
very
concise.
A
Additionally,
data
structures,
so
so
piano
was
not
only
interested
in
in
the
interface.
He
was
also
interested
in
implementation,
so
in
the
runtime
system
of
perfect
described
in
the
succinct,
eight
pages
in
the
notebook
shows
that
these
data
types
that
I
mentioned,
are
backed
by
very
sophisticated
data
structures,
making
it
very
easy
to
choose
the
right
data
structure
for
the
right
job.
So
you
would
have
the
ability
to
use
simple,
naive
arrays
when
you
wanted
to
to
use
more
complex,
you
know
finger
trees
or
something
if
you
needed
to
do
something
that
was
more
performative.
A
If
you
had
a
good
idea
of
the
ratio
of
reads
to
writes
in
your
data
structures,
this
language
is
sort
of
like
guides
you
in
choosing
the
correct
data
structures
to
back
your
programs
and
I
I,
keep
saying
this
over
and
over
again,
but
another
really
cool
thing
that
I'm
really
interested
in
that
this
language
has.
Is
that
it
this
big
sort
of
big
language
that
I've
been
describing?
A
Has
this
really
nice
feature
to
it,
which
is
that
it
compiles
to
a
small
core
language
similar
to
you
know,
Haskell
has
a
intermediate
language
called
HASC
that
all
things
can
compile
down
to.
So
similarly,
I
was
very
enthusiastic
to
see
that
this
was
a
central
tenet,
perfect
everything
could
be
compiled
down
to
a
small
set
of
instructions
allowing
for
multiple
backends
to
be
created
for
perfect.
Should
the
time
come
than
anyone
ever
actually
use
it
right.
A
So
a
really
nice
thing
about
a
language
that
compiles
to
a
small
set
of
instructions
is
that
you
can
use
that
instruction
set
as
the
starting
point
to
target
other
languages.
So,
instead
of
having
to
say,
you
know
rewrite
all
of
the
different
parts
of
the
language
in
Java,
you
would
just
essentially
have
to
implement
these
core
functions,
and
then
you
would
be
able
to
produce
code
and
other
programming
languages
and
we're
seeing
this
a
lot
now
very
interesting.
A
Things
are
happening
along
these
lines
in
a
no
camel,
for
example,
where
it's
really
easy
to
target
these
intermediate
languages
and
generate
JavaScript
from
a
camel
code.
That
you
know
is
small
and
concise
and
does
not
contain
anything
that
you
don't
need,
and
so
this
small
core
another
advantage
of
it
was
that
it
was
part
of
pianos
obsession
with
portability.
A
So,
for
whatever
reason,
perfect
was
meant
to
run
on
any
available
architecture
and
though
I'm
far
from
an
expert
on
low-level
computing,
I
can
tell
that
much
care
and
effort
was
put
into
this
area
of
the
languages
design.
So
piano
was
aware
of.
You
know
the
idea
of
multiple
architectures
and
the
impact
that
those
things
would
have
on
how
this
language
is
back-end
would
work.
A
He
was
interested
in
being
able
to
target
all
different
types
of
computers
with
this
programming
language
and
again
another
thing:
that's
cool
about
this
small
core
is
that
it
also
allows
for
programmers
to
potentially
formally
verify
programs.
So
implementations
of
algorithms,
for
example,
can
be
proven
in
perfect
Fritton
properly.
A
So
there
seem
to
be
this
idea
that
you
would
have
an
interaction
or
an
interface
between
this
small
poor
back-end
and
a
seer
improver
of
some
type,
and
you
would
be
able
to
kind
of
move
back
and
forth
between
those
things
and
say:
okay,
well,
I
have
this
algorithm.
It
purports
to.
You
know,
conform
to
the
specific
invariants
and
now
I'm,
going
to
hook
it
up
to
the
sphere
improver.
A
On
the
back
end,
that's
going
to
show
me
that
in
fact,
it
does
and
then
I
can
use
that
information
going
forward
and
know
that,
as
opposed
to
say,
writing
a
bunch
of
unit
tests
to
prove
that
my
algorithm
was
correct.
I
could
lean
on
this
formal
verification
property.
To
help
me
understand
the
correctness
of
my
of
my
programs,
and
so
perhaps
most
shockingly
of
all,
you
know
the
thing
that
most
programmers
seem
to
care.
A
Then
you
would
have
the
ability
to
write
code
in
this
language,
and
you
could
essentially
then
leverage
all
of
these
advantages
that
I've
been
talking
you
through.
So
here's
a
list
to
summarize
the
notebook
covers
these
points
and
some
more
I
just
kind
of
excuse
because
there's
a
lot
in
there,
and
so
you
know
this
is
a
language
that
is
specified.
A
It
has
a
fancy
type
system.
There's
macros!
There's
data
queries
it's
declarative
when
it
can
be
its
concurrent
when
you
want
it
to
be
it's
distributed.
If
you
need
it
to
be,
it
can
be
both
object-oriented
and
functional.
It
can
be
both
static
and
dynamic
it.
It
uses
a
lot
of
pattern
matching
and
lets
you
use
destructuring
to
make
your
programs,
you
know,
read
concisely:
it
has
excellent
access
to
data
structures,
it
compiles
down
to
a
core
language
and
it's
portable
and
verifiable.
A
You
heard
me
describe
this
language
right,
and
so,
if
you
could
use
this
language
right
now,
would
you
raise
your
hands
if
you
would
be
interested
in
trying
this
language?
Okay,
cool,
I,
don't
know
I'm
going
to
say
everyone
Ranger
again,
it's
just
because
I'm
on
video
and
you're,
not
a
few
people
did
so
thanks,
I'm,
going
to
tell
you
why
you
should
not
use
it.
A
The
number
one
reason
why
you
shouldn't
is
because
it's
not
real
much
like
the
languages
that
for
hazed
critiques
in
the
short
story
about
Wilkins
on
this
language
was
completely
made
up
by
me.
For
the
purposes
of
this
talk,
there
was
no
notebook.
There
was
no
box
of
records
and
I
still
don't
have
an
original
copy
of
that
song
right
album,
though
I'm
going
to
do
some
record
shopping
later
today,
and
maybe
I'll
find
one
here
in
Los
Angeles.
A
That
would
be
nice
and
so,
in
other
words,
the
the
quote-unquote
perfect
with
a
capital
P
programming
language,
it
doesn't
exists
right,
I'm,
always
lying
to
you.
I,
like
lying
it's
fun,
and
also
notably,
the
perfect,
was
a
lowercase
P
programming
language
does
not
exist
either
right
and
so
the
reason
why
I
chose
Sun,
Rob
or
Hazen
McCarthy
to
be
the
connections
to
this
language
is
because
I
wanted
us
to
consider
this
idea
of
constructing
languages.
This
is
a
an
idea
that
has
a
rich
and
very
interesting
history
in
humanity.
A
For
as
long
as
we
have
been
speaking
and
writing,
there
have
been
people
who
have
attempted
to
create
languages
artificially,
which
are
intended
to
be
better
than
the
languages
that
we
use
and
Esperanto
is
one
example
of
that.
It's
one
that
a
lot
of
people
have
heard
of,
but
it's
you
know,
and
it's
probably
the
most
popular
of
these
constructed
languages,
but
there
are
a
lot
of
other
ones
that
are
very
interesting
that
go
back.
A
You
can
find
examples
from
you
know
illuminated
manuscripts
from
the
fifteenth
century
which
describe
perfect
logical
languages
like
this,
and
so
this
is
a
warning
for
us
to
heed
when
we
approach
programming
language
design
and
when
we
evaluate
what
languages
we
use
when
we
write
programs,
because
you
could
look
back
at
that-
celestial
important
of
benevolent
knowledge
that
bore
haisa
books
in
his
story
about
Wilkins
and
apply
the
same
idea
right
to
a
random
person
named
Jay
piano
who's.
Writing
the
these
tenets
of
a
perfect
programming
language
in
a
notebook
right.
A
The
kind
of
thinking
that
leads
us
to
search
for
the
perfect
programming
language
is
the
same
kind
of
thinking
that
might
end
up
classifying
all
types
of
programs
into
14
arbitrary
categories,
those
concerning
numbers
with
more
than
18
words,
fast-slow
Fibonacci
sequence,
easy
to
reason
about
used
to
transfer
money
written
by
people
named
Jay,
underground,
dangerous,
confusing,
etc.
I
had
to
keep
that
was
stolen
and
that,
from
a
long
way
off,
looks
secure
and
as
much
as
the
awesomeness
that
I
described
in
the
perfect
language
would
make
for
a
very
cool
language.
A
There
was
a
lot
missing
there
right.
There
was
a
lot
that
was
not
in
that
notebook
that
I
made
up
right.
In
fact,
none
of
this
isn't
that
I
made
up,
but
any
language
that
doesn't
think
about
these
things
about
the
human
side
of
how
programs
are
written
than
why
they
are
written
is
severely
missing
the
point.
So
things
like
errors
that
are
friendly.
A
The
things
like
how
people
are
actually
going
to
like
write
these
programs
in
their
editors,
that
they
use
package
management,
libraries,
existing
programs
and
idioms
documentation
and
testing
right,
like
all
of
those
gnarly
things
that
are
actually
what
people
who
write
programs
spend
their
time
doing.
There
was
there's
no
mention
of
this
in
the
idea
of
a
perfect
programming
language.
A
We
need
to
completely
get
this
idea
out
of
our
heads
that
there
is
some
perfect
language
out
there
and
it's
just
somehow
different
than
the
one
that
we're
using
right
now
right,
I,
say
the
perfect
language
is
the
one
you're
using
right
now
for
all
their
differences
in
languages
and
spoken
languages
do
have
this
included.
They
are
measured
by
what
they
are
capable
of
creating
and
even
our
most
maligned
languages
are
capable
of
creating
fabulous,
fabulous
things.
A
So
if
we
shouldn't
be
focusing
on
perfection,
what
should
we
focus
on?
What
can
we
learn
about
languages
that
have
been
successful?
What
makes
a
language
successful
a
languages
do
not
succeed
because
they
are
perfect
right.
That's
obvious,
I
think
the
existence
of
this
conference.
The
fact
that
we're
all
sitting
in
this
room
together
proves
us
that
proves
to
us
that
I'll
be
on
perfection.
Somehow
we've
been
capable
of
achieving
quite
a
lot.
A
We
don't
measure
the
success
of
our
projects
based
on
which
languages
we
use
to
create
them
right.
There's
a
kind
of
old
chestnut
about
the
engineering
manager.
Kind
of
you
know
wagging
their
finger
as
a
programmer
telling
them.
You
know
the
customer
doesn't
care
that
you
know
Yatta
Yatta
code
review
or
something
right,
in
other
words
like
when
you
log
on
to
not
you,
but
when
a
normal
person
that
doesn't
know
how
computers
work
log
on
to
a
website
and
something
works
or
does
not
work
there.
A
They
don't
care
about
the
language
or
the
technology
that
was
used
in
the
implementation
of
that.
It's
just
not
important.
Doesn't
matter
and
languages
succeed
because
of
the
tools
that
we
create
to
support
them.
To
me,
the
most
exciting
advancements
in
languages
recently
are
those
that
push
the
boundaries
on
how
developers
actually
use
them
right.
So
you
see
languages
that
have
been
around
for
a
long
time
like
PHP
JavaScript,
Java,
Ruby
Python
closure
is
old
enough
now
to
sort
of
be
in
this
list.
A
These
languages
are
not
killing
themselves
to
be
perfect
and
it
sometimes
drives
people
crazy
that
the
people
that
are
at
the
helm
of
these
languages
aren't
doing
what
they
want
them
to
do
with
that
language
right
like
how
can
they
not
see
how
much
more
perfect
this
language
can
be
right,
you're
missing
the
point:
if
that's
what
you're
spending
your
time
focusing
on,
so
we
should
recalibrate.
That's
the
point
of
this
talk.
We
should
not
aim
for
perfection.
A
We
should
not
cut
down
other
languages
and
communities
for
working
on
things
that
we
don't
care
about
right.
It
doesn't
help
if
you
write
a
framework
and
you're
talking
about
how
the
other
framework
is
more
complex
than
your
framework.
Let
them
make
their
complex
framework
when
they
come
up
with
a
good
idea,
which
they
inevitably
will.
You
will
steal
that
idea
and
use
it
for
yourself,
and
that
is
how
technology
works.
A
That's
actually
how
we
learn
and
get
better
and
make
things
that
allow
us
to
do
really
cool
like
sit
in
a
room
in
a
1920s
Theatre
in
Los
Angeles.
Listening
to
some
like
you-
and
perhaps
most
importantly,
you
know,
like
I,
said:
let's
care
about
what
other
people
are
working
on
and
steal
liberally
and,
let's
all
listen
to
Boris
on
ROM,
read
more
Boris
and
study
more
McCarthy.
So
thank
you
for
listening
to
my
story.
Sorry
for
the
bait-and-switch
I
hope
it
was
worth
it.
Thank
you
very
much.