►
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
B
B
D
A
A
D
B
A
C
C
C
A
C
Yeah
I
was
in
Greece
yeah.
That
was
that
was
that
a
holiday?
No
well
yes,
and
no,
it
was
a
I
spoke
at
a
conference
called
boxed.
Vo
XX
e
d,
thessaloniki
very,
very
nicely
well-run,
well
organised
conference
and
then
took
an
extra
that
was
on
Thanksgiving
Day
cuz
I
worked
on
Thanksgiving
Day,
because
that's
how
I
roll
turns
out,
they
don't
celebrate
Thanksgiving
in
Greece,
what
not
sure
what
that's
of
that
and
then
and
then
mo
and
I,
went
to
Athens
for
a
day
and
looked
at
old
things.
Oh
that's
awesome!
That's.
A
A
And
I
enjoyed
it
very
much,
but
I
wouldn't
mind
seeing
what's
left
there
by
actually
being
there.
That
would
be
super
old,
yeah,
very
old.
What's
called
the
game
to
watches
back,
everything
is
normal.
You
can
all
relax
I'm
back.
This
is
my
my
my
tick
watch
II
I
check
out
in
Kickstarter,
which
is
there
an
established
company,
it
turns
out,
but
they
wanted
to
make
an
Android
watch,
and
so
they
had
their
own
platform.
So
this
is
was
like
a
hundred
dollar
Android,
wear
SmartWatch
and.
A
Life
amazing,
so,
at
the
end
of
the
day,
when
I
take
it
off
it's
about
46%
left.
So
if
I
wanted
to
keep
going,
I
could
like
it's
a
72%
now
I've
been
wearing
it.
I
put
it
on
at
like
9:00
o'clock
this
morning.
It's
really
good
and
it's
actually
much
more
responsive
than
the
the
Huawei
watch.
I
was
running
before
before
the
I
stopped
wearing
the
watch,
because
the
these,
like
the
haptic
engine,
stopped
working
and
wearing
a
watch
that
doesn't
give
you
alerts.
A
Oh
it
doesn't
vibrate
when
you're
gonna,
let
it's
kind
of
pointless,
so
I
stopped
wearing
it,
and
then
I
tried
that
Hawaii
the
band
pro
and
it
was
very
cheap,
which
was
very
good
and
it's
fitness.
Things
are
good,
but
I
really
wanted
it
for
alerts
more
than
anything
else,
and
it
just
didn't
like
it
doesn't
have
a
full
touchscreen
with
an
alert
comes
in
it
doesn't
auto
scroll.
You
have
to
like
tap
it
to
read
everything
which
is
it
just.
It
was
terrible
for
that
type
of
thing.
In
the
end,
yeah
it'd
be
great.
A
B
We're
talking
nerdy
stuff,
the
did
you
guys
see
the
snap
dragon
announcement
thing
today.
A
A
B
Talking
about
the
the
first
ones,
it's
it's
I
mean
it
does
emulate
the
or
translate,
or
whatever
the
36,
apps
yeah.
A
So,
apparently,
is
pretty
good
thing.
I
think
I
think
this
is
the
beginning
of
we'll
see.
I
mean
obviously
not
for
us.
These
laptops
aren't
designed
for
power
users
they're
designed
for
normal
core
normal
users
and
they're
also
always
connected
because
they're
LTE.
So
even
when
they're
closed
they're
always
connected
to
the
network
and
they're
completely
instant
on
that's
like
an
iPad,
you
hit
a
button
and
they're
on
instantly.
There's
no
like!
Oh
it's
here,
it's
connected
standby.
You
hit
the
button
and
then
like
it
comes
on
within
three
seconds.
It's
not
right.
A
Apparently
so
yeah
we'll
see
we'll
see
but
interesting
new
development
there,
which
I
thought
was
pretty
cool.
So
ya
know
this
is
the
pixel
to
excel
I
got
my
upgraded.
My
pixel
thing,
I'm
very
happy
with
it.
So
far,
I
have
north
and
Excel
well,
I
had
the
pixel
excel
and
I
gave
that
to
my
son,
and
so
now
I
have
the
pixel
to
excel,
and
my
wife
got
the
pixel.
She
had
the
Nexus
5x,
which
was
like
2
years
old
that
went
to
my
daughter.
A
A
C
C
B
C
A
A
C
A
Yep
yep
yep,
so
supportive
I
am
absolutely
supportive
now
other
than
that,
the
only
I
was
just
saying
to
jump.
Before
we
go
down
there.
My
house
is
decked
out
with
Google
homes.
Now
I
have
five
throughout
the
house,
because
if
I
bought
two
of
the
home
minis
when
I
bought
this
and
then
I
didn't
realize,
I
was
gonna,
get
two
free
ones.
Then
I
got
two
free
ones
and
I
already
had
the
original
view
climb.
A
So
I
have
five
in
the
house
now
anyone
know
they're
all
they're
all
spread
out
the
house,
so
I
can
just
yell,
hey
Google
at
any
point
now
and
I
can
turn
off
the
Christmas
tree
and
I
can
change
the
thermostat
or
I
can
play
music
or
I
can
ask
what
the
weather's
gonna
be.
You
all
know
is
it's
pretty
good,
hey,
Google
spy
on.
A
D
D
A
B
C
B
B
C
B
Listen
we're
going
back
to
his
mom's
blog
posts,
he's
been
doing
a
series
on
localization
configuration
with
asp
net
core,
so
he's
he's
kind
of
built
up
from
first
of
all
getting
localization
cultures
from
adjacent
file.
Then
he
moved
into
actually
like
using
those
localization
resources
and
in
this
one
he's
he
is
dynamically
reading
the
cultures
at
runtime
from
the
json
file.
So
when
you
go
in
and
call
in
to
use
request,
localization
he's
reading
from
you
know,
what's
actually
in
his
request
culture
provider
yeah.
B
C
A
B
Yes,
Steve
so
Steve
Gordon
wrote
about
gotcha
here
with
environment
variables,
with
asp
net
core
to
O
and
Linux,
and
the
deal
is
when
you
have
delimited
like
nested
environment
variables,
the
colon
delimiter
works
in
a
lot
of
places,
but
not
everywhere,
and
so
on.
The
newer
Linux,
the
newer
Debian.
So
this
is
Debian
stretch.
You
need
to
underscore
x'
as
a
delimiter.
B
Yes,
so
he
talks
about
how
he
troubleshot
that
and
he
actually,
he
said:
hey
I
went
and
tried
to
look
up
where
this
was
documented,
that
the
delimiter
no
longer
works
for
because
he
used
to
work
with
the
Jessie
images
and
I
dug
around
myself.
This
morning,
too,
and
I
didn't
see
a
change
log
thing
where
they
said:
here's
I
I
did
there.
There
is
in
the
asp
net
github
issues
right.
There's,
there's
well.
A
I
can
only
speak
to
the
a
spinet
side.
We've
always
supported
the
double
underscore
as
a
delimiter,
because
we
know
that
there
are
platforms
that
don't
allow
:
environment
variable
names,
but
we
don't
keep
track
of
which
platform
from
version
to
version
supports
one
of
the
other.
But
you
know
the
our
config
system
supports
either
so
that
you
can
use
double
one.
Let's
go
when
you
have
to,
because
you
can't
use
:.
It
might.
A
A
C
B
B
A
I
just
want
to
check
it
is
docked
in
our
docks.
So
if
you
go
to
Ethernet
core
docks,
I
got
a
fundamentals
configuration
there
is
a
bullet
point
there
around
in
configuration
considerations.
It
says
if
a
colon
can't
be
used
in
environment
variables
on
your
system,
replace
the
column
with
a
double
underscore
yep.
D
A
B
B
So
this
is
a
this
is
by
niraj
and
he's
writing
about
multi-tenant
applications.
This
is
a
I.
Would
call
this
a
pretty
like
a
you,
know,
naive
or
a
pretty
simplistic
approach.
This
is
just
using
an
action
filter
and
it's
reading
it
out
and
then
applying
it
to
things.
So,
basically
he
builds
up
to
you
when
he
says
sample
dot,
localhost
or
hello
versus
localhost.
B
He
does
different
things
based
on
this
action
filter,
so
I
would
I
would
call
this
out
more
as
a
reminder
of
you
can
use
action
filters
to
do
quite
a
lot
of
stuff
people
that
have
built
out
full
multi-tenant,
like
there's
a
lot
of
stuff
to
think
of
as
far
as
identity
and
all
that
kind
of
stuff
right,
but
but
yeah.
So
I
thought
this
was
a
pretty
cool.
You
know,
application
of.
If
you
took
an
action,
filter
and
and
solved
solved
a
problem,
there
stays
our
dilatory
all
as
always
right.
B
This
is
following
in
his
pattern
of
writing
a
hundred
pages
to
completely
and
thoroughly
document
something
this
is.
This
is
really
good
and
in-depth.
So
he's
talking
about,
I
hosted
service
and
background
service
class,
and
he
you
know
I
won't
won't
go
through
all
this,
but
just
this
is
this
is
pretty
in-depth
documentation
about
how
to
go
out
how
to
build
on
background
services,
so
very
cool.
B
And
and
I'll
of
course,
he's
really
good
about
you
know
just
diagrams
and
like
spelling
stuff
out
in
detail,
so
very
nice,
so
Tim
he's
been
building
out
this
simple
commerce.
So
this
is
a
commerce
application.
I
like
this
because
he's
this
is
going
beyond
the
very
simple
you
know:
store
Add
to
Cart
and
then
the
rest
is
an
exercise
for
the
reader.
This
is
a
this
is
a
full
commerce
site
and
he's
building
this
out
and
and
here's
showing
how
to
deploy
it
to
AWS
elastic
beanstalk
and
Postgres.
B
C
B
Okay,
so
here
we've
got
from
believe
it's
Fabian,
Fabian
yep.
So
here
he's
two
from
from
Fabian
one
is
credence
in
angular
with
asp
net
core
and
hate
OS.
So
that's
a
hypertext
as
the
operating
something
of
something
I
get
pretty
far
through
that
engine
of
applications
stayed
there.
We
go
so
I
call
it.
B
So
so
this
is
a
you
know,
pretty
full
walking
through
implementing
that.
So
this
is
application
state
using
ASP
net
core.
So
a
lot
of
lot
of
code
here,
but
so
this
is
pretty
cool
scene,
seeing
as
he's
kind
of
work
through
that
pattern.
Here
then
he's
also
applied
that
to
pagination
so
there's
the
angular
material
paginate
ER,
which
I'll
actually
he's
got
a
link
to
at
the
end
of
this
doing
anything.
C
A
B
A
B
B
This
is
this:
is
a
current
one
talking
about
setting
up
bamboo
as
a
build
server
and
build
tasks
for
asp
net
core,
so
nice,
so
bamboo
is
from
Atlassian
when
I
haven't
worked
with
but
nice
to
see
that
written
up
andrew
lock
back
again
so
who's
talking
about
using
docker
hub
to
automatically
build
docker
images
for
asp
net
core.
So
he's
talking
about
you
know
integrating
that
in
with
both
the
doctor,
tooling
and
then
also
docker
hub.
So.
A
B
B
B
A
B
B
Definitely
aim
a
shortfall
so
yeah,
so
this
is
the
beginner
course
going
through.
You
know
getting
started
up
through.
You
know
basic
cred
and
logging
and
stuff
and
the
next
one
going
through
more
building
out
an
application.
This
is
more
of
a
going
beyond
hello
world
to
hey.
Let's
actually
build
an
app
so
working
with
things
like
Web,
API,
swagger,
publishing,
etc,
and
then
Jeff
and
I
did
a
bunch.
That's
not
a
good
picture
either.
There's
really
no
way.
Yeah
I
can't
be
done.
B
B
In
Photoshop,
something
in
I,
don't
know
so
you
know
like
one
flair,
for
instance
I'll
just
you
had
more
lens
flare.
So
here
we
talked
about
some
stuff,
like
advanced
internals,
with
middleware
advanced
configuration.
We
did
some
things
about
running
in
production,
so
some
some
pretty
cool
stuff
here.
Overall,
the
advanced
configuration
there's
some
neat
stuff,
both
configuration
both
here
and
authentication
authorization
stuff
were
some
of
my
favorites
Jeff
did
a
really
cool
job
here,
but
so
anyhow
tons
of
great
stuff.
This
is
fresh
content.
We
just
did
this
in
what
was
it
late,
October?
B
It's
all
2.0,
good
stuff,
late
arrivals
people
ping
me
after
I
kind
of
pre
leaked
what
I
was
planning
on
talking
about.
This
is
a
global
exception
handler
for
dotnet.
So
this
is,
if
you
want
something
that
kind
of
configures,
you
know
global
exceptions,
it's
going
to.
Instead
of
return,
it
returns
a
500
for
every
exception.
So
and
then
he
goes
through
and
shows
shows
how
to
kind
of
set
things
up
and
configure
it
and
handle
specific
exceptions.
So
interesting.
A
A
1X
there
wasn't
into
oh,
they
added
back
the
api's
on
the
abdomen
type
that
allowed
you
to
register
handlers
for
globally
uncaught
exceptions,
which
was
a
big
ask.
So
yeah
like
this
is
use
exception
handler.
So
that's
this
seems
to
be
capturing
exceptions
to
do
with
the
request
pipeline
only
and
which
is
about
our
do
them.
Usually
so,
okay.
A
I
know
other,
like
libraries,
whether
it's
a
cap,
insights
or
Nicks.
So
when
he's
hey,
yes,
not
exceptional,
you
know
one
I,
don't
know.
I
can't
remember
the
name
of
all
these
languages
observer
or
not
that
one,
the
other
one
he's
listening
so
he'll
tell
me
like
I'll
watch
it
in
the
lag
go
so
he'll.
Tell
me
what
I'm
talking
about
his
library
that
does
this
yeah.
B
B
Alright,
so
here
he's
sending
you
know,
real-time
chat,
messages
and
stuff
using
angular,
so
he's
got
it
kind
of
plumb
through
angular.
Sometimes
that
can
be
a
little
more
complicated,
but
I've
seen
some
really
cool
things.
People
have
built
out
using
angular
with
signal
are
in
the
past,
so
so
this
is
pretty
cool
with
that.
So
as
always
tons
of
code
from
him
tons
of
links,
so
very
detailed-
and
you
know,
there's
a
repo
for
all
that
as
well.
No.
C
C
C
A
C
A
C
I
think
that
it
is
a
combination
of
people
who
are
comfortable
and
familiar
with
open-source
taking
our
stuff.
You
know
60%
of
quote-unquote,
our
stuff
and
40%
of
the
great
things
that
they
find
in
the
community
that
John
talks
about
the
containers,
the
micro
services,
the
open
fast,
all
the
different,
great
things
that
there
are
out
there
yeah.
B
C
C
If
you
don't
understand
the
full
stack
and
get
support
on
the
full
stack,
then
you
need
to
get
support
in
the
full
stack,
and
that
means
either
only
use
Microsoft,
which
I,
don't
think,
is
a
isn't
necessarily
a
good
thing
or
support
a
whole
lot
of
open
source
projects
and
I
see
that
a
lot
of
companies
out
there
in
the
community
just
aren't
prepared
to
do
that.
Like
no.
A
This
seems
to
be
a
topic.
That's
come
up
more
and
more
recently,
which
is
like
what
is
the
true
cost
of
participating
in
open
source
or
even
assuming
open
source
like
and
there's
been
I
think
there's
to
be
a
resurgence
of
the
idea,
or
even
just
an
initial
realization
and
a
lot
more
people
champion.
This
idea
that
if
you
consume
open
source
you're
now
part
of
the
project,
you
didn't
buy
a
product,
you
help
the
customer
you're
part
of
the
team.
It.
C
Seems
that
some
people
will
go
and
clone
that
bring
it
into
the
project,
not
talk
to
their
bosses,
and
then
an
open-source
project
has
become
a
piece
of
your
production
code,
and
no
one
mentioned
it
right.
Right.
A
And
then,
when
the
Heron
does
hit
the
fan
at
some
point
in
the
future,
if
it
does-
and
this
is
where
a
lot
of
the
original
fodder
and
open
source
happened-
I
mean
I
mean,
as
is
always
the
case,
some
fighters
often
based
in
real
concerns
and
it
gets
lost
in
the
whatever
and
I
think
there
is
a
real
concern
for
certain
types
of
customers
and
companies
which
is
like
well.
What
happens
when
the
unthinkable
happens
for
this
open
source
library?
Are
we
prepared
to
react
for
that?
A
C
D
C
Was
a
bug
a
well-known,
now
legendary
bug
in
the
rolling
file
appender,
it
did
not
roll
just
the
logs
get
bigger
and
bigger
and
bigger
and
bigger,
but
the
log4net
was
strongly
strongly
typed
yeah
and
then
I
fixed
the
bug.
The
people
who
worked
on
log4net
cared
not
yep
would
not
take
the
equivalent
of
the
pull
request,
which
was
a
diff
file
emailed
in
a
mailing
list.
C
They're
deep
and
profound
apathy
over
the
course
of
nine
months
caused
me
to
eventually
fork
log
for
net
and
sign
up
my
own
thing
right
and
then
I
spent
the
following
five
years
dealing
with
my
own
Fork
of
log4net
with
its
own
private.
You
know
with
its
own
strong
name
right
now.
That's
a
that's
a
that's
a
that's
an
impudent
again,
a
pimp
youment!
Is
it
a
word
that
there
is
a
Jacques
Hughes?
That
is
a
bad
thing
for
strong
naming
to
be
clear,
but
also
the
open
source
process
get
didn't
exist.
C
A
Got
nested
dependencies
where
you
become
this
is
like
left
pad
right.
Why
don't
you
become
the
the
the
potential
for
that
to
create
a
house
of
cards?
Type
of
effect
is,
is
really
I
mean
now,
certainly
other
situations
that
can
cause
that
to
be
a
real
problem
are
somewhat
limited,
but
they
do
happen.
As
you
just
pointed
out,
you
had
a
serious
you
had
a
bug.
It
was
affecting
your
app
the
product
journalist,
for
whatever
reason
were
interested
in
taking
the
fix.
So
what
are
you
to
do
now?
A
If
you
had
17
libraries
in
your
solution
that
were
also
depending
on
that
that
other
library,
where
you'd
identified
the
bug
you
forking
it
now
means
that
you
have
17
libraries
that
aren't
using
the
version
that
you
fought
and
unless
you
go
off
and
fork
those
as
well,
because
it's
somewhat
viral
now
the
strong
naming
thing
is
no
longer
a
concern
in
dotnet
core,
because
it's
not
part
of
the
identity
check
anymore,
but
there
are
still
other
ways.
Obviously
there
that
can
cause
issues,
and
you
know
either
put
in
each
platform.
A
I
think
has
there
has
there
but
technical
behaviors
that
affect
how
this
affects
apps
on
that
platform
like
in
you
know
in
ode,
your
dependencies
come
in
a
source
code,
so
you
know
it's
a
little
different.
They
all
get
loaded
at
runtime
and
there's
no
strong
type-checking,
and
so
you
can
get
away
with
a
lot
more,
but
the
problem
still
exists,
as
we
saw
with
left
pad
when
it
was
removed.
People's
builds
failed
because
that
dependency
just
disappeared
one
day
and
they've
taken
steps
to
stop
the
happening
again,
but
yeah.
That's
a
really
interesting
discussion.
A
I've
noticed
the
other
one
that
comes
up
has
come
up
recently.
Is
this
idea
of
LTS
versus
sort
of
tip
or
current
right,
and
so
a
lot
of
platforms
with
their
operating
systems
or
open
source
application
platforms
like
node,
have
a
concept
of
a
long
term,
servicing
branch
or
version
versus
a
tip
or
a
current
version,
and
they
have
different
support
life
cycles
and
and
bars
or
like
what's
the
universal
term,
we
use
the
term
bar
at
Microsoft,
but
like
the
criteria
for.
B
A
So,
like
you
know,
if
I
want
to
make
this
change,
it's
like
well,
we
only
am
take
these
types
of
changes
in
this
branch.
We
want
new
thing
on
this
API.
You
have
to
do
in
this
branch
right
and
we
have
a
similar
thing
in
dotnet
core,
but
that
also
impacts
all
the
third-party
libraries
because
it's
like
well,
let's,
let's
take
a
spin,
eight
core
as
an
example.
A
If
we
have
an
LTS
and
we
have
a
current
and
they're
different
versions
and
you
as
a
library
author
want
to
integrate
with
us,
so
you
want
to
offer
banana
package
on
new
Georg.
That
adds
a
tag
helper
right,
there's
a
easy
way
to
integrate
with
a
to
connect
or
you're
using
our
API,
and
you
might
want
to
use
an
API
that
we
introduced
in
the
new
version
in
the
current
version.
It's
not
in
the
LTS
version.
What
do
you
do?
A
Do
you
create
two
packages,
one
for
LTS
customers
and
one
for
that
has
its
own
problems
in
net.
If
they
use
the
same
type
name,
they've
got
tight
conflicts.
Do
you
have
one
package
with
different
versions
so
that
that
works?
That's
usually
the
default
thing,
except
that
the
new
get
system
is
kind
of
designed
to
offer.
You
updates
all
the
time
right.
A
C
C
Brings
up
the
even
larger
issue,
which
is
the
lots
and
lots
of
people
out
there
are
bringing
in
new
versions
of
stuff
just
because
there
are
new
versions
right,
like
we
joke
about
NPM,
it's
easy
to
joke
about
NPM.
When
was
the
last
time
you
made
a
new
hello
world
application
and
then
got
a
clean,
restore
of
NPM
yeah.
B
A
I
have
14
other
things
that
depend
on
that
library,
because
it's
so
low
down,
JSON,
dotnet
or
logging
library,
whatever
else
whatever
it
might
be,
but
something
in
a
screen
it
and
then
like
for
humans
to
be
able
to
determine
how
could
my
app
possibly
change
by
updating
this
is
difficult
right.
It's
it's
not
really
something
you
can
rationalize
about
with
that.
A
This
is
this
whole
paradox
at
that
I
want
libraries
to
be
maintained
if
I
wanna
take
your
dependency,
I
kind
of
like
the
idea
that
updates
are
coming
out
right
because
it
means
someone's
working
on
it
and
someone
cares,
but
at
the
same
time
now
I
have
to
think
about.
When
am
I
supposed
to
update
this
and
if
I
report
a
bug.
A
Here's
an
example:
if
I
report
a
bug
in
a
library
that
I'm
using
and
I
haven't
updated
it
for
a
year
because
of
the
reasons
we
just
articulated
like,
why
should
I
update
if
I
don't
need
the
new
version
and
then
I
come
across
a
bug
and
the
owner
goes
well.
Yeah
I
fix
that
bug
like
three
months
ago.
He
is
the
version,
but
they
did
that
in
a
major
release,
and
so
they
broke
stuff
right
there
following
semver
fix.
A
C
A
C
A
Interesting
thing:
I
don't
know
of
any
NPM
packages
that
do
maybe
they
do.
Maybe
they
do,
but
again,
because
javascript
is
a
little
bit
different.
You
can
get
away
with
a
bit
more,
but
now
it's
giant
doesn't
maintain
any
LTS
release,
so
the
active
development
of
node
all
happens
in
current.
There
is
a
consortium
of
companies
who
maintain
LTS
like
the
Java.
A
If
they
discover
an
issue
themselves,
they
can
fix
it
in
LTS
and
then
send
a
PR
to
get
it
integrated
into
current
and
they
take
that
burden
on
because
base
value
in
it.
Now
that
works
great
for
large
platforms
like
node
or
Microsoft
platforms,
will
we
have
a
team
who's?
You
know
who's
paid
to
do
that.
But
what
does
that
mean
for
the
wider
ecosystem
within
the
community
if
you're
a
random
NPM
package
author?
Is
it
up
to
you
to?
A
Is
it
just
expected
that
you
should
maintain
an
LTS
version
of
your
package
and
a
current
version
so
that
customers
can
choose
and
that's
a
lot
of
work
and
there's
a
lot
of
stuff
to
keep
up
with
so
I?
Don't
know,
there's
been
some
good
articles
about
this
recently
and
we're
doing
some
thinking
about
this
on
our
own
side,
because
it
does
affect
the
ecosystem.
A
Having
these
trains
is
something
that
library
authors
in
particular
have
to
think
about,
and
then
customers
on
the
airline
application
owners
have
to
think
about
it
in
the
terms
of
what
experience
there,
we
have
to
think
about
it.
What
experience
to
customers
get
if
they're,
using
LTS
or
current,
and
there
are
packages
in
their
graph
that
have
updates.
A
You
know
to
your
point
before
what
you
don't
want
is
for
someone
to
suddenly
update
some
third
party
package
that
they
were
using
very
successfully
and
then
it
somehow
lifts
half
of
their
app
into
current
and
leaves
the
other
half
in
LTS.
So
it's
an
interesting
discussion
and
I
like
that.
This
is
being
discussed
on
blogs
and
I've
seen
it
discussed
in
some
places
in
the
community
recently
and
I.
Think
we'll
continue
to
have
this
discussion
and
figure
out
what
it
is.
C
A
A
That's
based
on
a
net
standard,
1x
platform,
so
dotnet
4.6
anything
before
four
six
one
or
don't
net
call
1x.
When
you
were
so
library
new
get
says:
yeah
I'm,
gonna
install
banana
for
you,
but
I'm
going
to
install
this
other
two
pages
of
dependencies
as
well,
and
we're
now
seeing
feedback
from
community
like
library,
author
owners
who
are
getting
feedback
from
their
customers
saying.
Why
am
I
seeing
all
these
dependencies
coming
in
my
app
now
that
you've
moved
to
net
standard,
let's
ember
to
solves
it?
A
But
if
the
customer
isn't
on
earnest
and
a
new
platform,
yet
it
doesn't
solve
it
for
them.
Obviously,
so,
as
I
said,
if
you're
under
four
six
or
if
you're
on
don't
make
all
1x
till
you
still
see
this
thing,
and
so
we
would
really
love
to
get
to
getting.
Everyone
moved
to
done
in
standard
two
as
quickly
as
possible.
A
C
A
Back
to
just
other
customers
to
be
very
clear,
Nick
has
his
own
cobbles
with
being
an
application
owner
building
the
Stack
Overflow
app,
which
are
very
well
founded
and
where
you
know,
we
need
to
figure
out
how
to
work
through
those
things
as
well,
which
is
all
good
anyway,
with
anything
else
that
you've
got
from
the
community
that
we,
you
think
we
should
talk
about.
Otherwise,
I
can
give
a
brief
up,
know
what
we're
doing
and.
A
I
wasn't
a
quick
update
that
we've
been
promising
promising
is
a
strong
word.
I've
been
mentioning
two
point
one
for
a
while.
We
are
getting
much
much
closer
to
having
some
news
about
that.
In
fact,
we
hope
to
do
a
video
session
before
the
holidays,
our
channel
9
session
with
Scott,
not
you,
our
boss,
Hunter
Richard
and
myself,
sort
of
going
over
the
top,
the
high-level
scope
for
2.1,
and
hopefully
some
dates
for
that
as
well.
A
Some
rough
dates-
and
we
you
will
people
who
are
watching
our
github
repos,
will
see
issues
in
the
two
one
milestone
now.
So,
if
you
want
to
go
and
have
a
look,
you
can
try
to
see
some
leave
the
features
that
are
being
worked
on
there
are
there.
Arts
I
will
say
it.
So
the
big
features
like
signal
or
is
being
worked
on
signal
are
still
committed
for
2.1.
So
we
have
a
bunch
of
folks
working
on
that.
A
We
do
have
some
big
sort
of
resource
intensive
items
that
is,
they
take
engineering,
a
lot
of
engineering
time
to
work
on
that.
Don't
directly
relate
to
new
features,
and
so
2
1
will
be
a
release
where
there
are
some
new
features,
but
perhaps
not
as
many
as
you
would
usually
see
for
a
milestone
this
size
simply
because
we're
spending
a
lot
of
engineering
time
working
on
more
foundational
issues
which
I'll
explain
so,
for
example,
the
dotnet,
what
we
call
the
inner
loop
performance
I
think
Scott
you've
talked
about
in
a
loop.
This
turned
out.
A
Fourth,
you
can
see
on
your
blog
post
recently,
which
is
basically
we
use
in
a
loop
to
talk
about
all
the
things
that
developers
do
multiple
times
a
day
like
I
make
a
CO
change.
I
save
it
I
build
I,
relaunch
the
application.
I
make
a
CO
change
on
my
test,
I
rediscover
test,
I
rerun
tests,
those
type
of
things;
okay,
I,
add
a
package
I
have
to
restore
and
then
I
build
and
I,
see,
compilation,
errors
and
I
make
changes.
Those
are
all
formed.
A
The
inner
loop
once
you've
committed
to
source
control,
you've
kind
of
you've
exited
the
inner
loop.
At
that
point,
so
you
could
kind
of
claim.
It
is
oh
anything
that
our
developer
does.
While
they're
in
their
active
workspace
before
they
push
in
just
back
into
source
control,
is
inner
loop,
don't
net
core,
whether
in
vs
or
the
command
line,
has
some
parts
of
that
experience
that
we're
nowhere
near
happy
with
that?
A
Is
it's
just
too
slow
and
so
we're
doing
a
massive
effort
across
lots
of
teams
right
now:
msbuild
Roslyn,
the
compiler
teams,
the
project
system
and
Visual
Studio
the.net,
CLI,
the.net,
core
SDK,
the
a
snit
core
team.
What's
that
six
I
think
it's
actually
seven,
I'm
teams
who
are
working
together
to
try
and
isolate.
Where
are
we
losing
that?
Most
of
that
time?
Right
now
for
the
various
scenarios
that
we
care
about
and
and
really
trying
to
chip
away
to
improve
that
we've
made
some
good
progress.
A
Some
of
that
was
released
in
vs
15-5
yesterday,
but
we
have
a
long
long
long
way
to
go
yet
to
really
get
to
the
type
of
performance
that
we
want,
which
is
you
know
for
very
large
solutions.
We
have
a
couple
of
representative
solutions
right
now.
We
have
the
Rosslyn
solution,
which
is
the
solution
that
actually
builds
the
Rosslyn
compiler
and
we
have
orchard
core
which
I
think
when
debita
recently
actually
done
mm-hmm
yeah,
which
is
a
large
solution.
A
It
has
like
a
hundred
and
twenty
projects
in
it
and
it's
all
dotnet
core
and
we're
using
those
two
as
kind
of
our
real-world
test
solutions
to
test.
What
is
the
performance
like
and
we're
doing
it
both
from
the
command
line
and
in
Visual
Studio,
because
those
characteristics
are
quite
different,
and
so
we
have
a
long
way
to
go.
We
have
made
some
progress.
We've
gotten
some
work
in
progress
right
now
which
we're
hoping
to
get
some
good
gains
from
once.
It
finishes,
but
I
think
we're.
A
A
I'm
also
doing
a
lot
of
work
around
improving
our
build
system
so
that
we
can
build
new
releases
much
quicker
than
we
could
before,
which
is
important
just
to
tighten
that
feedback
loop
for
ourselves,
so
that
when
we
make
changes,
we
can
get
a
full
build
out
very
quickly,
and
then
we
can
verify
them.
We
can
experience
them
those
type
of
things.
So
there's
a
lot
of
work
going
into
that
as
well,
and
then
we're
making
some
foundation
well.
B
A
Have
the
full
stack
built
every
day
or
multiple
times
a
day
to
the
point
that
doesn't
just
produce
assemblies
like
it
produces
installers
it
produces
the
things
that
we
deliver
to
customers,
like
that's,
really
really
important.
Obviously
so
yeah
we're
doing
a
lot
of
that
work
right
now,
which
is
good.
A
Then
we
actually
start
getting
close
to
the
feature
work.
So
we've
also
got
some
work
around
the
the
run
time
store,
which
we
introduced
in
Oh,
which
is
the
part
of
dotnet
core
that
allows
a
spinet
core
apps
not
to
have
to
deploy
the
a
spinet
core
dll's
into
the
application,
all
right.
So
that's
why
a
Snickle
two
applications
are
typically
much
smaller
once
you
publish
them
than
they
were
in
1x.
The
Assemblies
in
there
are
also
optimized
their
pre
jittered
so
that
the
application
can
start
up
faster.
A
There
are
some
issues
that
we've
come
across
now
that
the
runtime
store
is
out
and
we've
had
to
do
updates
to
things
in
the
runtime
store.
We
did
a
203
servicing
released
last
month
in
November
for
the
whole
stack
that
we
would
like
to
improve
and
so
we're
making
a
change
to
how
we
deliver
the
a
spinet
core
assets
rather
than
using
the
runtime
store.
The
runtime
store,
as
a
feature
will
continue
to
exist,
but
a
spinet
core
will
not
be
in
it
anymore.
A
We're
going
to
refactor
the
a
spinet
core
apart
into
a
shared
framework,
which
was,
if
you
remember,
the
dotnet
core
layout
on
disk.
If
you
go
into
Program
Files
on
a
Windows
machine,
you've
got
like
the
dotnet
XE
and
then
you've
got
a
bunch
of
folders.
Next
to
it.
One
of
those
folders
is
called
shared
and
inside
that
there's
a
folder
called
microsoft.net
core
app.
A
There
will
be
a
new
folder
next
to
that
called
Microsoft,
but
a
snit
core
dot,
something
right
and
that
will
be
where
all
the
a
spinet
core
stuff
is
that's
pretty
compiled
and
like
one
big
blob
for
that
server
or
for
that
machine,
I
should
say,
and
so
the
meta
package
that
you
use
right
now
will
still
be
the
mechanism
that
you
use,
that
you
use
to
acquire
that
and
use
that
in
your
application.
So
that
part
won't
change.
A
We
call
that
trampolining
so
that
the
app
will
be
deployed,
assuming
that
a
spinet
core
was
in
the
runtime
store.
But
the
dotnet
host
will
be
updated
to
know
that
if
you're
looking
for
an
assembly-
and
it's
not
in
the
application,
but
the
application
was
deployed
for
a
spinette
core,
then
it
needs
to
go
and
find
them
from
this
shared
framework
over
here
and
that
solves
a
whole
bunch
of
problems
that
we
were
having
specifically
for
using
the
runtime
store
to
deliver
a
spinet
core
like
that.
A
Many
assemblies
which
are
designed
to
update
over
time
and
we're
hoping
you
get
some
good
benefits
out
of
that.
So
there's
a
bunch
of
work
going
on
there
again.
Most
of
this
is
silent,
none
of
the
covers,
but
it
takes
resources.
So
it's
worth
calling
out
and
then
we're.
Then
we
actually
get
to
slow
some
features.
A
We
are
building
a
new
version
of
that
phrase
Nick
or
on
top
of
the
new
MVC.
In
that
same
time,
frame
we're
doing
some
work
around
gdpr,
the
new
EU
regulations,
around
data
protection,
they've
data
privacy,
debit,
whatever
that
stands
for
and
that'll
include,
some
new
API
is
inside
of
our
cookie
policy
middleware,
so
that
it's
easily
implement
cookie
compliance
on
your
site.
That.
A
And
the
idea
is
that
that
will
be
the
default
in
the
templates
now.
So,
if
you
file
new
the
full
template
into
one
it'll
be
set
up
out
of
the
box
with
some
UI.
That's
yes,
fully!
Customizable,
it's
just
in
the
template
and
a
middleware,
that's
installed!
That
does
all
the
basically
you
know
show
that
show
a
notice
to
the
user
to
tell
them
that
the
site
uses
cookies
and
don't
write
any
cookies
until
they
accept
it
doesn't
do
any
of
the
geo
stuff
like
we're,
not
gonna
ship.
A
Any
solution
that
lets
you
determine
if
someone's
in
Europe
or
not
I,
mean
they're
still
up.
That's
still
an
exercise
to
the
reader,
but
at
least
the
mechanism
of
having
some
UI
getting
consent
have
that
plummet
or
the
cookie
system
and
stuff
will
be
there.
What
else
we're
still
doing
a
bunch
of
razor
tooling
improvements?
We've
got
a
long
series
of
work.
A
A
And
then
you
get
all
the
incremental
updates,
so
once
the
app
is
running,
you
can
just
go
back
and
change
the
razor
file
and
the
app
in
the
background
was
warm.
It's
in-app
compilation
system
so
that
if
you
do
change
a
file
after
the
app
is
running,
you
still
get
the
benefits
without
that
instant
recompilation
that
the
Razr
file
wants.
A
The
app
is
running
when
you're
in
in
a
loop,
alright,
so
the
whole
bunch
of
work
going
on
to
do
that
which
will
hopefully,
which
will
end
up
I,
think
making
some
great
experience
improvements
for
the
data
for
developers
there
we're
doing
some
work
in
NCM.
So
the
ace
minute
core
module,
which
is
the
part
of,
is
the
host
a
spinet
core.
Today,
that's
out
of
process,
so
we
actually
launched
on
their
XE.
A
We
now
have
an
in
process
mode
so
that
you
can
actually
have
w3
WP
directly
launched
the
dotnet
core
CLR
and
then
launch
your
application
in
process.
So
there's
a
couple
reasons
why
that's
beneficial
one
it's
much
faster,
so
we
get
about
a
four
times
performance,
throughput
change
for
hello,
world
sort
of
scenario.
So
that's
really
really
good
and
two.
We
think
it's
going
to
be
more
reliable.
So
in
situations
where
the
dotnet
XE
failed
to
start
and
there's
lots
of
reasons,
why
doesn't
it
might
not
start
right?
A
You
may
look
if
any
one
he
was
watching
the
live
stream
I
did
before,
where
I
fixed
the
live,
face,
v-net
website
from
the
bug
that
we
all
found
a
few
weeks
ago,
if
you
have
the
wrong
version
of
the
runtime
installed,
your
app
won't
start
right,
you
say,
run
the
application,
it
says:
well,
this
happen.
E's
1.1,
you
only
got
200
it'll
fail.
If
that
happens,
behind
dot
behind
is
the
process
just
fails
is
doesn't
know
what
to
do.
It's
like.
Well
done.
A
It
actually
didn't
start
sorry
and
you
just
get
a
white
screen
when
we're
running
in
proc.
We
can
now
get
distinct
error
codes
back
from
the
host
from
the
dotnet
host
and
it
can
tell
you:
oh
the
application
didn't
start
because
it
was
missing
this
framework
and
then
we
can
give
you
a
nice
friendly
error
page.
We
can
wipe
that
error
out
to
the
IaaS
logs
like
into
the
event
log
and
into
the
is
valid
request
tracing.
So
it
should
be
much
easier
to
diagnose
issues
with
why
your
dotnet
core
app
didn't
run
behind
I.
A
Yes,
so
we're
hoping
that
that's
gonna
also
I
should
just
be
more
reliable
because
it's
there's
a
lot
less
moving
parts
we
don't
have
to
like
create
sockets
and
then
forward
requests
over
pull
the
other
process
and
stuff.
So
it
won't
be
the
default
because
we
can't
make
a
chase
like
that.
Obviously
in-band,
the
intent
is
for
it
to
be
the
default
for
new
project.
So
when
you
file
new,
a
2.1
project
in
proc
would
be
configured
by
default,
but
for
any
existing
application.
A
It'll
continue
to
be
out
of
process
until
you
change
it
in
proc
and
then
also
if
you
are
doing
in
proc.
There
are
some
caveats:
you
can
only
run
one
application
for
Apple
in
praat
because
we
can't
load
multiple
versions
of
the
dr.
Corsi
all
are
side
by
side
in
one
process
and
we
don't
have
AB
domains
and
things
we
did
in
dotnet
framework.
So
there's
a
there's
a
couple
of
drawbacks,
but
there
tend
to
be
fairly
edgy
like
most
people
don't
use
those
features
so
well.
C
A
Yeah
what
else
is
going
on,
so
we
we're
still
doing
the
kestrel
work
to
move
to
the
sockets
transport
and
set
a
Libby
v.
We
still
plan
to
have
the
libuv
transport.
We
won't
remove
it,
but
the
hope
is
that
we
can
default
kestrel
into
one
to
the
sockets
transport
the.net
manage
sockets
implementation
transport
so
that
we
have
a
common
networking
base
in
dotnet
used.
A
You
know
whether
you're
doing
up
sockets
yourself
or
whether
you're
using
Kestrel
in
a
spinette
core,
and
then
that
way
we
just
get
to
share
the
benefit
of
all
the
great
investments
the
dotnet
core
networking
team
has
been
doing
for
their
cross-platform
networking
support.
We
also
hope
it'll
make
the
performance
between
Linux
and
Windows
a
little
bit
more
consistent.
Today,
Windows
is
a
bit
faster
with
our
libuv
transport.
I
think
the
sockets
work
will.
C
A
A
A
A
brand
new
platform
for
us
so
I
think
it's
not
unfair
to
just
to
blame
the
fact
that
Windows
was
faster
because
our
Linux
implementation
was
just
newer.
Okay,
then,
as
we
as
we
invest
a
lot
more
into
the
Linux
side
of
the
darknet
call
runtime,
the
assumption
is
you'll
I
mean
I'm
talking.
The
difference
is
very
small,
like
it's
not
like
it's
twice
as
fast
running
on
Windows
like
in
some
scenarios,
it's
7%
faster
or
something-
and
these
are
benchmarks
I'm
talking
about.
A
So
it's
not
something
anyone
should
be
concerned
about,
but
it's
just
one
of
the
advantages
we
hope
by
getting
on
top
of
a
common
bass.
Another
thing
we're
doing
is
important,
which
is
we're
looking
to
default
to
HBS
for
new
projects,
so
we're
doing
a.
But
today
we
did
a
blog
post
of
a
couple
of
weeks
ago,
and
if
anyone
read
it,
oh,
they
did
about
setting
a
hub,
a
spinet
core
for
HTTP
for
development.
A
No,
thank
you.
I'm
Dan
and
Glenn
wrote
this
one.
It
is
not
trivial.
Today,
it's
actually
quite
a
lot
of
configuration
to
get
it
correct
if
you're
doing
it
all
behind
is
you
just
use
is
as
wonderful,
easy
HTTP
configuration,
but
if
you're
doing
it
all
in
Kestrel,
it's
quite
a
lot
more
setup
and
so
part
of
the
work
we're
doing
is
to
make
it
a
whole
bunch
of
API
works
and
middleware
work.
A
Some
configuration
and
tooling
work
to
make
it
so
that
a
we
can
make
HTTP
the
default
in
new,
a
spinet
core
templates
and
be
the
code
required
to
do
that
is
not
you
know,
stab
your
eyes
out
horrible.
So,
like
it's
one
line
in
them
in
the
middleware
configuration
it's
a
one
line
in
the
config
and
then
cached
will
automatically
us
find
the
config
bindings
bind
to
HTS
by
default
as
well
as
HTTP.
So
today,
Kestrel
in
lack
of
any
configuration,
will
just
assume
you
want
to
listen
on
localhost
5000.
A
A
So
we
don't
want
that
to
be
the
default
experience
when
you're
building
applications,
they
it's
been
it
core
and
then
for
anyone.
Who's
worked
on
the
web
for
a
while
I
think
knows
that
there
can
be
issues
that
rise
when
you
are
in
production
in
HTTP,
but
you're
not
doing
development
in
HBS,
and
so
you
can
get
differences
that
come
in
that
you
maybe
don't
see.
There
are
rules
that
apply
to
things
that
you
can
do
and
the
side
is
not
HTTP
versus
when
it
is,
and
so
it's
as
better
if
it's
consistent
everywhere.
D
A
Doing
that
work,
and
then
there's
some
other
stuff,
but
that's
really
the
most
of
the
stuff.
That's
committed
right
now
to
get
us
to
the
first
preview,
which
we're
hoping
will
be
early
next
year.
I
think
I'm
breaking
any
rules
by
saying
that
and
then
we
would
hope
to
do
a
second
preview
where
we
page
in
some
smaller
features
after
that
and
obviously
address
their
feedback
that
we
get
from
the
first
preview
and
then,
ideally,
we
would
do
a
true
RC
with
a
go
live
license
after
that,
and
then
an
r
TM
sometime
after
that.
A
B
A
B
A
We
have
an
ongoing
effort
that
I
didn't
mention,
which
is
we
have
a
working
group,
that's
looking
at
data
access
performance
and
we
have
a
github
repository
that
you
can
use
either
you
can
track
our
performance
there.
So
if
you
go
to
github.com,
slash
a
spinet,
slash
data
access
performance,
the
V
team
or
the
working
group
is-
is
doing
their
work
inside
there.
A
Now
that
the
the
we're
using
the
tech
empower,
Fortunes
benchmark
is
kind
of
the
the
litmus
test
you
know
we
traditionally
have
performed
pretty
badly
in
that
what's
more
interesting
than
the
fact
that
we
perform
badly
in
it
is
that
all
of
our
entries
in
it
perform
about
the
same?
It
doesn't
matter
if
you
use
EF,
if
you
use
dapper,
if
you
use
ATO
Don
that
directly,
if
you
use
razor,
if
you
use
string
concatenation
it
sucks
everywhere,
it's
all
about
20,000
requests
per
second
within
a
thousand
requests.
Where.
A
A
Yeah
yeah
there
it
is,
and
so
so
far
what
we've
done
is
Andrew
here
on
the
EF
team,
along
with
a
Seb
Sebastian
from
the
orchard
team,
they
have
written
a
prototype
data
access
driver
for
Postgres
in
completely
in
managed
code
and
we've
codenamed
a
Peregrine.
So
we
have
Kestrel
as
the
web.
Server
and
Peregrine
is
a
data
access
library
just
to
help
us
do
this
initial
investigation
and
what
we've
basically
determined
is
that
there's
nothing
wrong
with
c-sharp
the.net
core
runtimes
ability
to
talk
to
a
database.
A
If
you
just
consider
see
how
the
language
in
the
runtime
all
the
performance
issues
are
somewhere
between
c-sharp
is
the
language
and
the
database
I
mean
that's
a
very
large
area
right
say:
do
dotnet,
it's
the
the
driver
for
a
do:
donate
you're
using
its
the
application
code.
You've
written
there's
a
lot
of
things,
but
we
have
a
version
of
the
fortunes
tech
empower
benchmark
in
Channel
me
right
now
that
uses
this
new
sort
of
Peregrine
driver
is
that
even
a
driver
actually
just
a
low-level
library
right?
A
If
that
makes
sense
right,
it's
not
a
data,
it's
not
a
generic
data
access
library,
it's
a
Postgres
library
that
can
do
very
limited
things,
but
that
proves
a
whole
bunch
of
things
that
proves,
if
you
remove
all
the
abstractions
and
all
the
API
for
data
access,
things
are
perfectly
fine.
Things
are
super
super
fast,
and
so
now
we
need
to
go
through
the
process
of
kind
of
adding
back
functionality
or
wrapping
some
of
the
existing
net
abstractions
around
it
like
ATM
and
then
finding
out
how
much
you
lose
like.
A
Where
are
we
losing
all
this
performance?
Are
we
losing
because
of
the
memory
allocation
that
if
it
gets
forced
upon
you
because
of
the
shape
of
the
edit
or
net
API?
Are
we
losing
it
because
of
the
implementation
of
the
Postgres
driver?
Are
we
losing
it
because
of
something
else,
and
so
we're
doing
that,
so
we've
made
good
progress
and
people
like
I
said
who
are
interested.
They
can
follow
along
with
that
progress
on
that
repo,
but
you
know
we're
serious.
We've
got
people
dedicated
to
it.
A
Nothing
is
off
the
table
like
we
don't
know
exactly
where
the
issue
is
right
now
we
just
know
that
if
we
talk
directly
to
Postgres
using
sockets
and
c-sharp,
it's
really
fast.
So
now
we
have
to
go
back
and
figure
out.
What
is
it
about
everything
else
in
the
stack
that
makes
it
slow
and
then
we
can
make
a
decision
about
what
we
have
to
do.
For
example,
maybe
we
end
up
with
a
new
a
do
Donette
driver.
A
Maybe
we
end
up
not
using
a
GeoNet
and
there's
like
a
fast
driver
for
EF
when
you
talk
to
Postgres
and
then
the
ATO
Dannette
compliant
driver,
maybe
it's
a
we
help
the
Postgres
they
do
to
net
driver
owner
who
is
a
an
open
source
library
contributor.
It's
not
owned
by
us,
it's
owned
by
someone
in
the
community.
Maybe
we
work
with
them,
which
we
are
already
actually
to
to
improve
their
driver.
A
We
don't
know
where
they
were
the
issue
where
the
box,
the
issue
is
yet
so
good
investigation,
though
happy
with
the
progress
so
far,
yeah.
So
someone
else
about
the
roadmap.
My
goal
is
to
publish
a
2018
roadmap
in
the
same,
in
the
same
spirit
as
the
vs
code
roadmap.
So
for
those
who
haven't
seen
the
vs
code
roadmap
I
really
liked
the
way
Chris
did
it.
So
it's
it's
thematic.
It's
not
a
list
of
issues
it's
like
in
2018.
These
are
the
things
that
we
care
about.
A
These
are
the
things
that
we
think
we're
going
to
go
off
and
and
investigate
so
I
plan
to
do
a
2018,
a
spinet
core
roadmap
of
that
ilk.
So
it
won't
be
a
list
of
issues
or
a
bullet
list
of
features.
It'll
be.
These
are
the
things
that
we
care
about
micro
services
and
performance
and
cloud
and
blubber
or
whatever.
It
is
right
and
I'll
dude
we'll
do
some
blurbs
for
each
of
those.
Is
this
the
looks
like
someone's
showing
the
vs.
A
C
A
So
if
you
just
go
and
like
look
at
the
2-1
milestones,
you'll
be
able
to
just
enumerate
all
the
features
and
enhancements
and
kind
of
see
the
body
of
work
that
we're
doing
so
I'm
yeah,
so
that
that'll
all
happen
over
the
next
month
and
I
also
hope
that
when
we
do
this
channel
9
video
with
Hunter
and
Richard,
that
hunter
will.
Let
us
talk
about
our
proposed
dates.
I
like
to
be
able
to
tell
people
these
things,
but
a
lot
of
that
will
depend
on
our
status
and
stuff.
So
I
need
to
go
yeah.