►
Description
Join members from the ASP.NET teams for our community standup covering great community contributions for ASP.NET, ASP.NET Core, and more.
Community links for this week: https://www.theurlist.com/aspnet-standup-2020-08-04
A
A
A
B
I
just
started
to
type
in
the
chat
that
I
didn't
realize:
james's
car
james
montemagno's
car
is
named
ty,
oh
and
so
in
this
update
little
known
fact
in
this
update,
we
are
actually
adding
sound
effects
to
time
james's
car
so
that
every
time
you
turn
on
the
car
it
says
just
do
it
really
loudly
everybody
around.
You
can
hear
it.
A
B
A
A
A
B
A
And
got
all
the
links
yeah,
I
think.
Okay,
let's
see
if
this
all
works,
so
what's
fun
here
is
we
are
still
both
alive.
So
don't
pick
your
nose.
It
is
time
to
show
the
fun
okay.
So
all
these
links
will
be
posted
in
the
show
notes.
A
There
it
is
no
no
there,
it
is
there's
the
link,
I'm
using
the
magic
of
internet
and
my
computer,
so
I
will
be.
I
will
be
posting
that
now
I
need
to
oops
escape
where's
my
escape
button,
whoa,
okay,
good.
Let's
stop
all
the
zooming,
I'm
getting
busy
very.
A
A
You
know
some
previous
posts,
he'd
done
and
stuff,
and
he
wrote
it
all
up
as
a
you
know,
like
some
practical
examples
using
this,
so
you
know
this
is
a
great
way
if
you
are
both
learning
about
asp.net
core
or
if
you
want
some
more
deep
information
he
all
of
these
things.
He
goes
kind
of
deep
on
stuff,
so
you
know
you
can
you
can
grab
this
enormous,
wonderful
pdf,
but
so,
for
instance,
for
any
of
these
topics.
He'll
talk
about.
A
You
know
gosh
whatever
it
is,
if
he's
talking
about
authentication
and
authorization
and
then
he'll
go
deep
on
a
lot
of
he'll
he'll,
explain
the
basics,
but
then
he'll
also
go
deep
on
stuff.
So
just
wonderful
stuff.
He
also
this
is
pretty
neat.
He
goes.
He
has
automated
all
this
stuff
to
generate.
So
he
generates
word
documents
from
his
xh
or
from
his
html
docs
and
stuff.
It's
pretty
cool
too.
A
So
if
other
people
want
to
do
something
similar
he's
written
that
up
too
all
right,
chris
saint
e
always
kicking
out
the
amazing
blazer
stuff.
So
here's
talking
about
client-side
pre-rendering,
so
he
talked
about
yeah.
What
is
pre-rendering-
and
you
know
like
you-
can
you
can
just
send
all
your
stuff
down
to
the
client
and
have
it
do
the
rendering
for
for
web
assembly
based?
A
A
Okay,
he
creates
this
new
app
using
blazer
pre-rendering.
There
is
a
setting
in
here
when
you
configure
in
the
host
and
and
so
when
he
goes
through,
and
he
does
see
adblazer
framework
files
and
configures
the
pre-rendering
and
then
there's
also.
Where
is
it
I
read
through
this
over
the
weekend?
I
need
to
make
sure
I
remember
there's
different
settings
for
the
pre-rendering
that
you'd
need
to
look
at,
so
you
can
pre-render
and
let
me
see,
there's
different
settings
to
dig
into
on
this
and
but
yeah.
A
A
A
A
So
in
the
the
get
repo
it
it
talks
through,
here's
what
the
syntax
is-
and
this
is
about
as
clear
as
you
can
explain
it,
but
so,
for
instance,
if
this
is
an
example,
that's
running
every
monday
and
let's
change
it
to
run
every
when
or
tuesday
right
and
then,
if
we
want
it
instead
of
every
second
tuesday,
we'll
run
it
every
third
tuesday,
you
know,
and
then
we
can
do
print
more
and
stuff.
A
So
this
is
great
for
being
able
to
kind
of
look
at
in
depth
and
and
test
these
out
by
the
way
glenn
when
you're
typing
it's
pretty
loud,
you
may
want
to
get
your
business
or
I'll
just
boot,
you
from
the
stream.
I.
B
Don't
know:
okay,
I'm
going
to
keep
tracking
everybody,
and
then
everybody
that
judd's
going
to
have
to
do
this
thing
about
thai
dessert.
It's
going
to
be
great.
B
A
You
it's
pretty
neat
yeah,
I
would.
I
would
make
all
kinds
of
crazy
promises.
Tai
will
integrate
with
your.
I
don't
know
something:
another:
okay,
here's
a
cool
post
from
pratip
on
the
aws
blog,
and
this
is
about
running
blazer
on
aws
serverless.
So
it's
just
neat.
You
know
it's
a
it's!
A
it's
client
agnostic!
You
know
it'll
run
on
any
server
and
it'll
run
on
static.
A
Like
people
have
been
doing
this
on
github
pages
or
whatever
you
do
have
you
know
you
do
have
to
publish
first
and
some
things
like
that,
but
it
does
run
on
multiple
platforms,
so
here
practice
walking
through
setting
this
up.
You
know
and
all
the
things
you
know
creating
the
application
and
then
it's
neat
that
you
can
configure
it
to
integrate
with
different
backends,
so,
for
instance,
here
deploying
out,
but
then
once
it's
deployed
out,
you
can
integrate
then
more
deeply
in
with
aws.
A
So
so
this
is
cool,
and
here
in
this
example
integrating
with
the
serverless
video
downloader.
So
it's
neat
stuff
there
yep
yeah,
yeah,
okay,
this
is
a
beefy
post.
This
is
one
that
you
just
remember
is
out
there
unless
you're
glenn,
who
actually
understands
all
this
stuff
in
depth.
Yeah.
B
I
looked
through
that
I
haven't
like
read
this
in
depth,
but
I
did
skim
through
it
again.
Oh
and
I
realized
just
now
that
he
paul
he
does
call
out
steve
gordon's
series
of
posts,
which
is
great,
there's
a
there's.
A
lot
of
detail
in
here
on
what
http
client
factory
does
in
the
in
depth
about
how
it
actually
works
right.
So
this
is
more
about.
This
is
more
about.
Oh
I'd
really
like
to
understand
what
this
thing
is
doing
under
the
covers
more
so
than
just.
B
I
want
to
understand
what
it's
for
and
how
I
would
use
it,
but
I
think
at
the
beginning
there
you'll
notice
he
called
out.
If
you
don't
know
anything
about
http
climb
factory,
already
go
check
out.
Steve
gordon's
blogs
series
about
it,
which
are
very
good
as
well.
If
you
want
to
now,
learn,
learn
about
it,
steve
actually,
when
ryan
noak
and
I
were
first
working
on
http
client
factory,
steve
got
in
touch
and
we
used
to
actually
exchange.
B
We
used
to
talk
to
each
other
in
order
for
him
to
be
able
to
generate
content
for
the
thing
as
we
were
building
it.
Oh
wow,
he
wrote
most
of
the
official
docs
as
we
were.
Writing
the
as
we
were
writing
the
code.
We
had
a
bunch
of
markdown
files
in
the
repo
with
stuff
in
them
and
he
would
look
at
them
and
look
at
what
we
were
doing
and
he
regenerated
a
bunch
of
content
for
us
and
it
was
amazing.
It
worked
out
really
well
for
us
and
and
him.
B
I
think,
because
I
believe
he's
done
some
pluralsight
courses
on
that
now
as
well
and
he's
becoming
a
big
content,
generator
which
is
good,
and
so
what
this
is
is
kind
of
that
next
step
right.
I've
learned
about
http
client
factory,
I've
kind
of
learned
what
it
can
do
for
me.
Let
me
really
get
into
depth
about
what
it
does.
B
So,
if
you
scroll
down
here
a
bit
further
down
to
kind
of
the
next
diagram-ish
down
to
the
kind
of
next
diagram,
or
so
we
start
seeing
like
the
more
about
the
handler
pipeline,
how
we
clean
up
the
handlers
when
we
do
it,
like
all
the
details
of
that.
So
if
you
really
want
to
understand
how
all
that
works,
go
check
this
out
and
and
have
a
look
and
browse
through
the
codes
and
and
see
how
it
goes.
Yep.
A
B
Some
cool
tricks
in
there
too,
like
how
we
do
we're
like
we're
using
a
timer
to
like
keep
track
of
everything
and
then
clean
it
up
right.
So
you
get
the
same
instance,
because
the
way
http
client
factory
works
is
when
you
get
a
new
http
client.
You
reuse
existing
handlers,
the
point
here
being
that
when
you
just
knew
up
a
http
client,
you
also
knew
up
an
underlying
type
that
eventually
turns
into
actually
binding
to
a
port
on
your
machine.
B
B
So,
no
so
that's
right,
so
you
don't
want
to
knew
them,
so
you
don't
want
to
ideally
knew
them
up
all
the
time,
but
not
knowing
that.
But
if
you
keep,
but
if
you
new
one
up
every
time
the
problem
here
is,
it
takes
a
little
while
for
that
underlying
port
to
be
released
after
a
socket
after
it,
http
client
goes
away
as
well.
B
B
So,
but
when
people
so
then
the
natural
tendency,
then,
is
to
grab
one
and
just
keep
it
around
forever,
because
that
binds
one
port,
but
you
just
keep
one
port
and
you
can
reuse
http
clients,
as
you
know,
okay,
now,
the
problem
with
that
is
that
in
that
case,
if,
if
the,
if
the
underlying
types
never
get
recycled,
you
end
up
in
situations
where
things
like
dns,
never
refresh
for
that.
How
should
you
be
clients
and
things
like
that?
A
bunch
of
different,
oh.
A
B
What
http
client
factory
does
is
it
kind
of
walks
this
middle
path?
Now
for
that
that
first
problem
that
problem
of
not
of
holding
on
to
one
forever
and
never
recycling
it?
There
is
now
a
property
in
http
client
itself
to
fix
that.
So,
if
you're
just
trying
to
avoid
these
two
problems
that
you
don't
necessarily
have
to
go
into
http
client
factory,
now
you
can
just
use
http
client
and
set
this
property
and.
B
Absolutely
so
what
the
http
client
factory
does
is
everyone?
Ever
you
go
create
new
http
client.
You
get
a
new,
a
brand
new
http
client
instance,
but
the
underlying
the
handling
handlers
back
that
actually
processes.
The
requests
is
reused
for
a
while
and
then,
after
a
certain
amount
of
time,
we'll
start
giving
out
a
new
one
and
then,
when
nobody's
using
the
old
one
anymore,
we'll
clean
it
up.
A
B
That
accepts
a
http
client
and
then
register
that
with
di,
and
then
you
can
test
it
and
use
it
and
completely
encapsulate
your
usage
of
a
patient
client
for
a
particular
endpoint.
You
can't
really
like
you
can
that's
that's.
That's
the
big
feature
that
it
gives
you
that
you
don't
get
from
just
http
client
and
of
course,
when
we
built
http
client
factory,
you
also
didn't
have
the
http
client
property
to
work
around
the
problem.
But
now
you
do
so.
A
You
know
so
the
whole
thing
with
type
clients-
I
remember
being
really
excited
about
this
and
actually
like
teaching,
workshops
and
stuff
and
being
like
these
type.
Clients
are
amazing.
A
B
A
B
Rat
that
basically
wraps
all
the
endpoints
for
your
api
and
you
just
grab
it
and
add
it
right
and
you've
got
the
unit
test
for
it
in
your
nuget
package.
You
know
it
works
like
you
could
imagine
doing
something
like
that.
If
you
wanted
to
the
other
thing,
is
you
could
also,
but
you
could
also
publish
open
api
documentation
and
generate
client
free
from
that,
the
one
of
the
things
that's
nice
about
typed
clients
as
well
is
you
can
do
things
like
entirely
transparently
encapsulate
something
like
say,
caching,.
A
B
A
One
question
is
like
so
typed
clients
are
amazing
with
http.
Is
there?
Is
there
a
similar
sort
of
thing
with
grpc
that
would
fit
in
or
is
grpc
a
different
beast
where
it
just
isn't.
B
B
Caching,
the
type
client
or
or
retries,
or
any
bunch
of
things
like
that
that
you
can
implement
in
the
pipeline,
handle
like
the
poly
integration,
for
example,
that
works
with
http
client
factory,
where
you
can
start
without
poly,
then
everybody's
using
hdb
clarifactory,
and
then
you
add
poly
in
one
place
in
the
beginning
of
your
project,
and
now
everybody
has
a
retry
policy
right
right,
which
is
super
nice
and
you
can
get
that
from
a
bunch
of
different
ways
like
handler.
Stacks
are
not
new
and
stuff,
but
it's
it's
good.
A
Cool
all
right
for
I'm
not
really
sure
this
restarted
the
stream.
I
believe
it's
it's
not
for
people
watching
on
youtube,
I'm
not
sure.
What's
going
on
after
the
after,
I
finish,
the
community
links
I'll
hop
over
there,
so
I
actually
just
have
a
couple
more
community
links
to
share
so
we've
got.
This
is
just
neat.
This
is
james,
doing
the
magic
things
with
grpc
and
he
added
it
to
this
benchmark.
So
this
is
a
public
benchmark
and
it's
cool
to
see
the
oops.
B
What's
really
interesting
to
me
about
this
picture
about
the
what
the
story
that
this
chart
tells
yeah,
because
it
tells
a
couple
of
interesting
stories,
but
the
one
that's
super
interesting
is
how
close
all
of
this
top
end
are
right.
How
close
they
are,
how
it's
all
flat
with
a
bunch
of.
B
Yeah
and
so
there's
a
few
there,
because
I
don't
know
what
the
causes
for
this
are,
and
I
know
investigations
will
be
happening
but
like
we
saw
a
bottleneck
like
that
appear
in
postgreson.net,
for
example,
and
it
turned
out
that
the
postgres
driver
that
everybody
was
using
in.net
needed
work
which,
because,
like
you,
had
dapper
dapper
the
rm
performing
at
the
same
level
as
like
a
full
entity
framework
or
any
framework,
or
things
like
that,
but
yeah.
B
So
I
don't
know
why
it's
flat,
but
I
don't
know
if
it's
completely
saturating
the
the
network
or
if
it's
or
what
the,
because
I
haven't,
spent
much
time
looking
into
it.
But
the
fact
that
it's
flat
is
always
interesting
because
it
indicates
potentially
either
everybody's
hit,
the
max
all
of
that
hardware
set
up
or
like
there's
something
that
everybody
is
using.
That
is
that
is,
that
is
it's.
B
But
yeah
super
nice
to
see
grpc
on
on.net,
be
so
fast
and
be
so
nice
and
you
can
see
if
you
go
back
to
that
picture.
The
other
thing
somebody
else
asked
was
what
about
when
I'm
on
the
net
framework,
which
I
believe
is
the
c-sharps
underscore
gipc
line
in
that
chart,
yeah.
A
B
That
you
could
use
on
full
framework,
which
is
still
not
bad,
not
bad,
but
yeah,
and
I
think
that's
because
it
I'll
see
james
might
if
james
was
going
to
be
here
and
since
he's
not
here
I'll,
just
talk
about
it
because
he's
not
here
to
defend
himself.
I
think
that's
because
that
one
wraps
the
wraps,
the
native
core,
the
native
kind
of
c
library
that
uses
for
grpc,
which
means
not
only
is
it
doing
all
of
grpc's
work.
B
But
it's
also
like
invoking
effectively
like
it's
got
this
invocation
layer
as
it
translates
between
the
c
core
and
and.net,
whereas
the
other
one
is
written
in
all
all.net
right,
written.
A
Yeah
yep
and
here
here's
the
repo
and
this
actually
this
includes
the
this
includes
the
you
know.
I
don't
know
the
the
spreadsheet
basically
of
data
yeah,
there's
a
output,
one
of
these
things
stats.
I
don't
know
it's
linked
in
here,
yeah.
B
A
B
I
mean
you
need
a
reasonable
target
right.
Yeah,
you
can
you,
can
you
can
chase
you
can
chase
whatever
you
want.
It's
just
a
matter
of
putting
a
putting
a
target
in
somebody's
yeah
say
like
chase.
This
number
make
it
better
right.
Well,.
A
You
know,
and
there's
I
mean
that
is
a
thing
microsoft
is
good
at
and
focused
on,
is
give
us
a
metric
that
we
can
track
over
time.
You
know,
and
that's
that's
just
so
baked
into
how
microsoft
does
stuff.
So
it's
it's
cool
to
have
like
hey
we're,
focused
on
this
and
like
this
you
know
the
the
I
don't
know
the
screens
up
in
the
you
know
the
work
areas
and
stuff
yeah
yeah.
A
A
B
Yeah
so
dev
time
azure
function
support
this,
so
deployment
doesn't
work,
yet
the
intent
is
that
we
would
deploy
to
kubernetes,
rather
than
so
the
way
this
works
is
you
have
a
functions
project
in
your
in
your
kind
of
solution
and
you
try
run
it
and
we
will
run
the
the
the
local
functions.
B
You
know
tooling,
to
boot
up
your
functions
and
have
them
run,
and
we
do
the
same
config
thing
that
we
do
right
now
when
you
deploy,
I
don't
think
they
go
anywhere
in
the
next
version.
We
will
deploy
them
to
the
cluster
using.
Is
it
kita
keda?
I
think
it
is
the
ubilift
the
functions
thing
that
meant
that
supports
running
functions
on
kubernetes.
A
It
just
this
all
brings
me
back
to
when
project
k
and
everything
started
with
k
and
stuff,
and
now
we've
got
the
kubernetes
living
the
same
life.
So
wonderful,
that's
right!
All
right!
Well,
that's
it
for
me
in
the
community
links.
Let
me
pop
over
to
this
here
view
so
yeah.
Well,
first
of
all,
thanks
for
being
on
the
show
right
after
I
know,
you
and
the
whole
team
was
really
busy
last
week
with
dot
net
micro
services.
A
Cool
yep,
so
a
lot
of
stuff,
that's
and
I'll
share
that
in
the
thing
too,
that's
focus.netconf.net
and
the
slides
and
videos
and
stuff
are
still
getting
posted
from
it.
Yeah.
B
I
believe
it
may
well
have
been
the
big
at
least
the
biggest
focus
event.
The.Net
can't
focus.
I
think
I
haven't
checked
it.
I
haven't
actually
read
the
email
in
detail
or
checked
with
with
beth,
but
I
think
it
was
bigger
than
the
blazer
one,
which
is
all
we
really
want.
Yeah.
B
A
Okay,
cool
cool,
so
so
I
don't
know
if
there's
any
just
other
than
project
tai,
which
we're
going
to
dig
into.
Was
there
anything
big
that
that
jumped
out
to
you
or
that
you
won
the.
B
Durian
dynamic
of
you
mean
yeah
yeah.
If
any
of
you
haven't
checked
it
out,
you
should
definitely
watch
at
least
the
first
three
sessions,
which
is
the
keynote
with
scott
hunter
and
david
fowler.
The
second
talk
it
was
brendan
burns
on
maybe
on
just
on
his
thoughts
around
doing
doing
microservices
or
not
doing
micro
services.
Where
he
does
talk
about
you
know,
maybe
you
don't
need
to
and
the
third
session
from
kelsey
hightower
to
see
somebody
who's
not
got
any
idea
about
what
donet
is,
but.
B
So
throw
if
you
watch
those
three
talks,
they'll
be
really
good.
He
was,
I
know
it's
it's
great,
because
kelsey
just
comes
from
go
and
so
he's
used
to
like
using
vm
we've
known
like
syntax,
coloring
and
like
it's.
It
made
no
sense
to
me,
but
it's
great
that
he's
able
to
successfully
do
that
and
give
some
of
his
opinions
and
talk
a
little
bit
about
what
it
was
like.
B
He
just
touches
briefly
on
what
it
was
like
diving
into
dinette,
because
one
of
the
things
you
we
have
in
dot
net
is
it's
it's
one
of
our
greatest
strengths,
but
it
can
also
be
very
intimidating.
Is
that
any
app
any
platform,
any
type
of
thing
right,
there's
just
so
much
stuff,
yeah,
so
jump
in
and
you're,
like
man
like
okay,
you're
right,
then
so
some
people
see
that
and
they
go
wow.
This
is
amazing.
I
can
see
I
can
make
a.
B
I
do
ai
an
ml
and
I
can
do
a
web
app
and
I
could
do
a
do
a
like
mobile
app.
I
could
do
everything
right
and
you
can
transfer
your
skills
across
all
those
app
types
and
that's
great
and
it's
super
powerful
once
you're
there,
but
I
think
kelsey
definitely
is
one
of
those
people
who
looked
at
it
and
went.
I
don't
know
where
to
be
in
now,
there's
so
much
stuff.
How
do
I
make
sense
of
all
this,
and
so
that
was
good.
B
Talk
about
some
of
those
experiences
and
then
just
talk
about
how
what
his
dev
flow
is
like
and
how
he
relate,
how
he
uses
very
simple
tools
to
keep
things
going
and
then
the
q,
a
section
was
probably
the
most
valuable
thing
about
his
talk,
so
good
for
definitely
for
an
alternate
to
the
main
to
like
what
we
expect.
What
we
see
every
day
is
kelsey's
talk
worth.
B
Brendan's
is
very
good
for
talking
about
organizational
structure
and
challenges
you
might
have
with
microservices
and
things
like
that
and
a
lot
of
good
good
discussions
about
that,
and
he
spends
a
lot
of
time.
You
know
talking
and
thinking
about,
microservices
and
kubernetes
and
how
people
should
use
that.
So
it's
definitely
some
good
insights.
There.
A
I
I
thought
his
presentation
was
so
smooth
and
fun
and
you
know
like
you,
can
get
really
like
you
tell
somebody
hey
presentation
on
microservices
and
kubernetes
and
it
might
sound
a
little
boring.
Maybe
you
know,
and
then
he
just
made
it
like
right
off
the
bat
he's
like.
Why
do
you
care
about
microservices?
You
don't
okay
show
over
you.
B
B
A
B
A
Yeah
right
well
right,
yeah
and
that's
you
know,
that's
something
that
I've
I've
learned
a
lot
over
time,
just
in
general,
with
any
kind
of
tech
is
like,
if
you
oversell
it
or
sell
it
to
the
wrong
people,
you're
nobody's
going
to
be
happy,
they're
going
to
be
upset,
they're
going
to
be
complaining.
You
know
it's
like
it's
really
good
up
front
to
just
be
like
here's,
where
it
fits.
Here's
where
it
doesn't
yeah.
B
And
I
think
in
general,
like
what
I
know,
all
of
our
all
of
our
of
the
dot
net
customer
base,
talking
to
being
one
being
a
donate
developer
for
so
many
years
before
I
started
to
work
on
the
team
and
then
interacting
with
so
many
so
many
of
presumably
all
you
folks
who
are
watching
this
conferences-
and
things
like
that
is,
I
think
that
message
of
don't
necessarily
chase
the
greatest
and
greatest
tech
like
do
what
makes
sense,
for
you
is
probably
something
that
our
that
we
can
all
relate
to
and
understand
so
yeah.
B
Those
three
were
really
good.
The
first
three
sessions
now
ever
all
of
the
rest
are
very
good
too,
but
what
are
the
difference
is
those
first
three
are
kind
of
very
generic,
very
general
hunter
and
fowler
gave
a
good
overview
of
just
in
general.
What
we're
doing
in
donating
microservices.
There's
a
few
there's
a
few
cool
demos.
In
there
we
did
some
some
single
file
stuff.
I
saw
some
thai
stuff
things
like
that
after
those
first
three,
then
everything
kind
of
dropped
into
okay.
B
Let's
talk
about
one
half
an
hour
about
steel,
toe
or
half
an
hour
about
orleans,
half
an
hour
about
messaging
and
asynchronous
right,
so,
like
just
kind
of
tech,
bam
bam,
bam,
bam,
bam,
one
after
another,
so
they're
all
really
good
as
well,
but
they're
all
kind
of
focused
on.
B
That's
probably
what
I
would
do
I
watched
most
of
it
live
as
I
was
curating
and,
as
I
was
talking
to
people
on
twitter,
but
that's
probably
what
I
would
do
right
now
is
watch
those
three
cherry
pick
the
half
an
hour
ones
that
seem
interesting.
I
watch
mine,
so
I
get
the
most
watched
views.
Obviously,.
A
B
Okay,
that's
that's!
That
is
fair,
okay,
so
so
just
started.
I
was
distracted
slightly
by
the
comments
in
the
in
the
twitch
chat
to
try
and
make
sure
that
make
sure
that
I'm
not
missing
a
cool
question
on
that
topic
we
just
had.
Generally.
I
think
people
are
agreeing
with
us
that
we,
deaf
friends,
don't
let
friends
be
buzzword-driven
developers
which
is
good
to
code
soon.
What
do
we
got?
B
B
Delayed
and
I'm
like
so,
the
tie
is
a
way
to
for
those
of
you
who
haven't
seen
it
go.
Watch
some
of
the
further
first
go
watch
the
first
session
of
dot
com
micro
services,
definitely
and
my
talk
on
and
my
talk
at
either
build
or
dot
net
conf.
But
it
is
interesting
because
I
want
I'm
going
to
talk
a
little
bit
about
the
history
I
guess
of
tai
and
what
we
were
setting
out
to
achieve
and
what
I
think
we've
actually
achieved
so
far.
B
I
think
over
basically
the
christmas
kind
of
lull
and
two
of
our
architects
and
a
couple
of
other
people
actually
as
well,
but
mainly
the
two
of
our
two
architects,
david
fowler
ryan
noack,
effectively
went
landed,
error
separately
and
built
two
different
prototypes.
B
A
B
And
ryan
is
right:
getting
to
kubernetes
is
hard.
Yep
fowler,
then,
is
focused
on
well.
We
haven't
even
solved
running
multiple
things
at
the
same
time
on
your
local
machine.
Yet.
A
Yeah
yep,
a
thing
that
I
like
you
know
about:
all
of
them
is
working
in
public
david
really
does
this,
where
he'll?
Just
you
know,
tweet,
stuff
and
they'll,
be
like
okay,
I'm
doing
this
thing
and
then
I'm
part
way
through
I'll,
be
like
man.
This
is
hard.
Why
is
it
so
hard
to
set
this
up?
And
you
know,
do
you?
Do
you
really
like
doing
this?
This
way?
A
You
know
it's
just
kind
of
it's
it's
cool
and
then
you'll
have
that
discussion
with
people,
but
it's
it's
cool
that
it
is
driven
by
actually
building
applications,
and
some
of
this
goes
back
to
some
of
my
favorite
things
about
asp.net
in
general,
like
when
damien
was
working
on
some
stuff.
It's
like
that's
how
kind
of
razor
pages
evolved
and
you
know
some
other
features,
it's
it's
kind
of
a
hey.
This
is
too
hard.
Let's
make
it
better.
You
know
yeah.
B
B
Okay,
there's
a
few
interesting
things
that
come
out
of
that
and
I
guess
we
probably
should
have
show
where
I
guess
I
show
a
bunch
of
this
for
those
of
you
who
have
tried
thai
and
done
the
survey
which
any
of
you
who
try
thai
should
because
it
helps
us
a
lot
and
if
you're
willing
to
have
us
talk
to
you
fill
out
the
last
question,
because
we
do
actually
go
and
talk
to
many
people
who
fill
out
the
survey.
Interestingly,
we
I
even
met
a
developer.
We
did
a.
B
We
did
called
them,
because
they'd
put
their
name
down
for
thai
to
talk
to
them
about
how
they
did
microservices
and
then
discovered.
It
was
one
of
the
developers
who
we
had
went
to
help
way
back
when
we
were
building
the
first
version
of
dot
net
call
we
went
to,
and
he
was
one
of
the
people
who
sat
with
us
and
we
helped
import
an
application
to.net
core
back
in
like
pre-100
days
at
donette
core.
It
was
great
there.
A
A
B
Yeah,
so
the
we
we
murdered
as
the
merger
of
these
two,
these
two
concepts,
but
one
of
the
things
that
that
survey
data
showed
us
was
most
people
aren't
really
using
deploy
right
now,
probably
like
70
percent
of
people
who
try
it
high,
aren't
actually
deploying
they're
just
using
the
local
developer.
Experience
there's
several
reasons.
Why
that
why
that
is
right,
one
is
like:
if
you're
just
kicking
the
tires,
maybe
you're
not
going
to
go
to
the
trouble
of
setting
up
a
community's
cluster.
B
B
Maybe
you
don't
want
the
cost
of
running
a
kubernetes
cluster
somewhere
and
then
trying
to
deploy
to
it
so
you're,
just
trying
out
the
local
stuff
right
or
maybe
you've
already
got
a
good,
mature
way
of
getting
to
kubernetes,
and
as
such,
you
don't
need
any
of
that
stuff
right
like
there's
a
lot
of
reasons
why
people
might
not
want
the
developer
flow,
the
product,
the
deployment
flow,
but
in
general
the
developer
photo
seems
to
make
sense
because
in
general
you
need
to
run
multiple
things
and
have
dependencies
the
thing
about
the
developer
experience
is:
it
doesn't
actually
have
anything
technically
to
do
with
microservices
as
much
as
it
does
just
running,
multiple
things
or
depending
upon
things.
B
It's
all
the
dependencies
and
all
that
yeah
it's
the
dependencies
right.
So
if
you're
gonna
use
something
like
a
redis
or
a
sql
server
or
a
message
queue
that
you
can
run
in
a
container
on
your
local
dev
machine
or
a
log
aggregator,
maybe
or
something
right
anything
that
you
can
run
in
a
container
that
you
want
to
have
running
at
the
same
time
as
your
epic
production
tyre
can
stitch
all
that
together
now
docker
compose
can
also
stitch.
B
A
A
B
B
A
B
And
so
the
yeah,
so
meshiko
not
really
right,
like.
I
think
the
point
of
this
little
tangent
is
to
say
that
the
developer
experience
that
we're
not
he
asked,
are
we
still
talking
about
microservices
and
no
not
really
like?
What
I'm
saying
is
that
the
tire,
the
developer
experience
of
depending
upon
containers
and
building
your
app
turns
out
to
be
actually
fairly
nice
with
a
whole
bunch
of
apps
for
the
whole
sorts
of
applications?
B
So
just
if
you're,
just
because
you're,
not
necessarily
interested
in
microservices,
don't
dismiss
tie
as
a
thing
for
micro
services.
That's
the
reason
we
tried
to
build
it,
but
I
believe
that
the
developer
experience
part
of
it,
at
least,
if
not
the
deployment
part
of
it.
The
deployment
part
of
it
is
obviously
very
focused
on
microservices
or
at
least
running
on
kubernetes,
but
the
development
part
of
it
is
not
the
development
part
of
it
has
just
run
multiple
applications
at
the
same
time
or
run
a
single
application
with
the
container
dependencies
yeah.
A
Well-
and
you
know
the
dependencies
too,
it's
like
it's
it's
very
easy
to
think
like
if
you're
focused
on
microservices
you're
thinking
of
microservices
only
but
like
you're
pointing
out
dependencies
include
things
like
you
know.
Regular
people
are
building
front
end
and
back
ends.
You
know
they
may
have
an
api
back
end
and
a
you
know,
angular
front
end
or
an
angular.
A
A
B
B
Look
at
that
professional.
We
know
what
we're
doing.
This
is
great
science,
okay,
so
in
order
to
in
order
to
just
to
just
to
highlight
that
difference
between
something
like
what
docker
compose
gives
you
versus
something
like
what
ty
gives
you,
and
also
to
show
something
that
I
know,
people
asked
a
bunch
of
questions
about.
B
During.Net
conf
was
debugging,
so
if
you
dock
a
compose
run,
you
can
run
all
your
local
container
dependencies
and
all
your
projects
together,
but
everything
has
to
be
containerized
to
run,
which
means
the
process
is
running
inside
a
container
which
is
then
an
isolation
boundary
between
your
local
machine
and
the
container.
So
getting
a
debugger
in
there
can
be
hard
right
and
we
have
some
visual
studio
tooling
to
try
and
help
with
some
of
that,
but
but
it
doesn't,
but
not
necessarily
within
the
compose
as
well.
So
I
have
a
sample
app
here.
B
It's
the
same
sample
app.
I
was
using
in
the
in
the
talk.
I
did
I'm
going
to
tyron,
and
so
this
is
going
to
spin
up
a
whole
bunch
of
stuff.
So
by
doing
thai
run,
it
gives
me
a
voting.
So
this
is
a
the
what
this
app
is.
I
guess
I
should
say,
because
not
everyone's
going
to
be
familiar
with
it,
it
is
an
app
where
you
can
go
and
vote
right.
B
You
can
pick
if
you
like
dogs
or
you
like
cats,
if
that
that
adds
my
vote
to
a
redis
instance
that
is
running,
then
there's
a
worker
which
picks
up
all
those
things
from
redis
and
that
keeps
up
to
update
to
postgres
database
and
there's
a
results
app
that
just
displays
the
results
from
the
postgres
database.
A
Yeah,
it's
like
there's
so
much
here
if
you're
managing
mo
and
I've
done
recently,
like
some
client
visits
and
work
with
people
that
were
doing
consulting
for
large
companies
and
they
like
it
was
crazy.
They're
like
I've,
got
nine
console,
windows
open
and
I've
got
all
this
stuff
trying
to
manage
all
this,
and
here
you've
got
it
all
right
here
in
one
automatically
created
dashboard.
So
this
is
cool.
B
Yeah
and
see
replica
account,
so
you
can
kind
of
simulate,
replicas
and
stuff
like
that,
and
this
is
using
identity
server
to
do
off.
So
one
of
my
things
is
auth
and
it
has
an
auth
service
specifically
anyway.
So
that's
the
rough
outline
of
the
app
now
what
I
was
going
to
show
you
was
I've
got
this
whole
thing
running
and
I
want
to
debug
something.
So
I
can
do
this
right.
Don't
net
attach
play
vote
attach
to
vote.
B
Let's
put
a
break
point
here.
I've
got
a
break
point
here,
which
is
this
is
on
the
index
page.
I
have
the
on
post
of
the
index
page
for
votes,
and
now,
if
I
come
back
here
and
I
click
dogs,
I
hit
a
break
point
right.
So
the
difference
there
you
go,
and
this
is
the
normal
debugging
experience
right.
So
the
only
difference
between
this
and
any
other
time
that
you
try
to
debug
any
net
application
is
that
the
is
that
the
you
had
to
attach
instead
of
f5
right.
A
B
Right
right,
so
we
can
detach
from
that,
but
that
is
one
of
the
big
differences
between
being
able
to
do
use
something
like
docker
compose
versus
something
like
thai.
Where
you
can
you
can
you
can
just
attach
because
they're
just
processors
running
on
your
machine.
B
Yeah,
we
don't
run
your
app
code
in
a
container
unless
you
go
tyron
dash
dash
docker,
which
is
which
gives
you
the
which
actually
containerizes
all
your
stuff
locally.
A
B
Exactly
which
means
you
don't
it's
a
lot
not
as
slow
to
start
up,
because
you're
not
doing
a
docker
build
in
order
to
do
your
dev
loop
and
also
you
could
just
attach
like
a
normal
big
benefits.
You
get
out
of
that.
Now,
yes,
cisco!
Yes,
you!
It's
not
not
actually
get
service
url.
So
the
question
was:
can
I
also
use
get
service
url
to
get
the
address
of
the
redis,
so
there's
actually
a
special
get
connection
string
method.
B
B
It's
just
environment
variables,
so,
as
ty
is
launching
all
of
your
processors,
it
sets
a
bunch
of
environment
variables
with
various
config
data,
and
then
these
methods
just
stitch
that
together
to
grab
it
out
and
give
it
to
you
in
the
format
that
you
want
in
the
thai
yaml
for
redis.
I
actually
say
like
this:
is
our
connection.
A
So
a
few
other
questions
here,
one
was
asking,
is
ty
still
an
experiment.
B
Yes,
tai
is
still
an
experiment
until
november
and
so
we'll
work
out
what
to
do
with
it.
After
that.
A
B
B
End
at
the
end
of
an
experiment,
there's
a
couple
of
reasons
why
we
call
something
experiment:
one
would
be
we're
not
really
sure
what
it's
going
to
be.
So
we
don't.
We
want
to
be
able
to
reserve
the
right
to
completely
pivot,
right,
yeah
or,
like
just
say,
okay.
This
was
a
great
idea.
Let's
take
a
bunch
of
this
learnings
and
roll
it
into
the
product,
or
so
I
don't
think
this
will
happen,
but
let's
paint
this
picture
for
you.
What
if
we
said,
okay
deployment
doesn't
really
matter.
B
A
B
Then
everybody's
using
it
and
then
we
decide
to
move
it
into
donette
right,
so
you
want
to
give
yourself
some
extra
freedoms
in
to
potentially
change
everything
around
or
pivot
or
break,
as
you
said,
make
lots
of
breaking
changes.
Now,
in
general,
we
don't
try
to.
We
try
not
to
make
lots
of
breaking
changes
because
they're
painful,
even
if
it
is
an
experiment-
that's
not
really
an
excuse,
but
we
do,
but
we
didn't
want
people
to
like
take
serious
bets
on
it
when
we
were
still
working
out
what
it
is,
yeah.
B
Okay,
that's
the
main
reason
why
we
do
something
to
call
something:
an
experiment
while
we're
working
it
out
and
we,
but
we
want
to
do
it
in
the
open,
but
we
don't
want
people
to
get
confused,
so
we
need
some
label
to
say
this
is
a
thing
we're
doing
we're
doing
it
in
the
open
you
all
can
participate.
We
want
you
to
tell
us
talk
to
us
about
it,
but
we
don't
want
you
to
like
think
this
is
definite.
This
is.
This
is
definitely
something
that
we're
going
to
do.
A
Right:
okay,
two
more
quick
questions
does
is
the
dashboard
out
of
the
box.
A
B
So
one
of
the
you
could
imagine
that
this
tire
could
use
helm
for
its
deployment
part
right
so
because
what
it's
doing
right
now
is
basically
generating
kubernetes
yaml
files
and
then
running
them
using
cube's
ctl
right
it
could
you
could
imagine
a
world
where
it
required
helm
and
it
generated
helm,
charts
and
then
can
use
helm
to
get
to
the
kubernetes
cluster,
but
in
general
those
things
are
all
going
to
do
more
than
what
ty
does,
because
it's
so
much
more
focused.
B
I
right
now
I've
viewed
right
now.
The
richness
of
deploy
with
tai
basically
leaves
it
at
a
point
where
it's
really
great
to
get
started
and
learn
a
bunch
of
stuff,
but
I
don't
think
we
have
the
richness
for
it
to
be
a
real
kind
of
production
ready.
You
know,
deployment
tech,
whereas
something
like
how
more
scaffold
is,
and
it
has
been
for
a
long
time
and
they
have
lots
of
features
because
of
that
that
you
know
ty
doesn't
have,
will
it
become
as
rich
and
full
feature
to
those?
B
B
I
I
think,
if
I
was
to
summarize
it
very
very
concisely,
I
think
the
answer
is
probably
no
links
to
thai
tutorials
there's
some
yeah
yep,
oh
and
somebody
did
some
stuff
cool
all
right,
so
that
was
that
I
was
attaching
and
doing
debugging.
Where
are
we
for
time?
You
have
to
keep
reminding
me
because
I'll
talk.
B
B
A
B
Deploy
dash,
I
so
I'm
not
actually
going
to
deploy
this.
This
would
normally
deploy
this
to
a
kubernetes
cluster.
I'm
not
actually
going
to
do
that.
I'm
just
going
to
get
it
started
and
then
I'm
going
to
go
to
the
file
system
and
look
at
the
files
that
get
generated
as
we
go.
So
I'm
telling
it
I'm
telling
it
glenn
c,
which
means
it's
going
to
think
it's.
It
thinks
it's
going
to
go,
publish
all
these
docker
images
that
it
generates
to
kubernetes
if
you
haven't
seen
tai.
B
Yet
we
didn't
really
give
you
the
like
wow
demo
here
that,
like
what
we
talk
about
is
the
kind
of
path
to
wow,
and
the
moment
is
really
when
you
realize
you
just
generated
a
docker
image
for
your
application
and
have
it
running
on
kubernetes,
but
you've
never
seen
a
docker
file
or
a
docker
compose
file
or
anything
there's
one.
You
can
do
it
without
any
ammo
right.
You
can
just
have.
A
B
A
B
An
app
see
this
temporary
directory
here
that
was
created
in
my
temp.
It
has
the
vote
app
right,
there's
a
dockerfile
and
it's
all
the
published
output
of
my
voting
application,
and
so
what
I
did
was
drag
the
dockerfile
over
into
vs
code
and
it's
gone
right,
because
now
the
image
is
done
so
on
my
local
docker.
I
have
that
image
and
it's
just
pushed
it
to
to
docker
hub.
B
So
if
you
were
to
go
find
me
on
docker
hub,
you
would
find
the
docker
image
that
actually
you
would
find
this
docker
image
because
it
just
pushed
it
to
docker
to
docker
hub
ready
to
go
to
kubernetes,
but
now
I'm
going
to
cancel
the
deployment
because
it's
not
actually
going
to
push
it
to
a
kubernetes
cluster,
so
this
docker
file.
So
what
was
said
what
happens
is
under
the
covers?
Is
it
does
dot
net
publish
into
that
temp
directory
that
you
just
saw?
B
Then
it
generates
this
file
on
disk
which
you've
never
you've,
never
had
this
file
anywhere.
But
this
is
a
docker
file,
but
we
know
it's
an
asp.net
application.
We
know
what
version
of.net
the
project
is,
so
we
can
generate
all
of
this
and
then
we
can
copy
the
published
output
into
the
docker
image
and
build
it
by
locally
using
your
local
document
and
then
push
and
that's
what
we
did.
A
B
B
Yeah,
I
believe
in
here
you
can
actually
give
us
a
docker
file
and
we'll
use
that
one
instead
and
things
like
that
right
because
we're
just
writing
docker
build
on
it.
So,
okay,
the
okay,
so
there's
a
couple
of
questions
I'll
get
past
before
I
continue
on
that
tangent.
I'm
not
sure
we've
ever
showed
that
much
before,
but
that's
basically
how
the
kind
of
magic
works-
and
you
can.
You
can
imagine
like
doing
the
same
thing
for
the
files
configuration
files
needed
to
go
to
kubernetes
or
at
least
automating
the
kubernetes
command
line.
Experience.
B
Is
the
voting
app
owns
open
source?
Yes,
so
the
voting
app
is
double
the
open
source
doubly
open
source,
because
it
is,
it
is
in
our
tire,
repo.
It's
currently
in
a
branch
we
haven't
merged
it
yet,
but
it
is
also
just
a
straight-up
copy
of
a
voting
sample
from
the
docker
ecosystem.
Right
like
it's.
If
you
watch
some
of
the
older
docker,
keynotes
and
docker
conferences
and
stuff,
they
show
this
voting
sample
when
talking
about
the
benefits
of
docker
and
how
docker
can
help
you
build
micro
services
and
stuff
like
that.
B
So
we
just
kind
of
stole
it
completely
from
that
sample,
and
it's
all
done
that
that
demo
is
usually
polyglot,
and
the
point
is
usually
that
each
of
those
bits
is
a
different
tack
like
the
dogman
voting.
App
is
a
I
don't
know,
say
python
and
the
workers.net
and
the
results
app
is
java
or
something
we
just
converted
it
all
to
dotnet
and
used
it
with
ty.
So
it's
in
the
thai
repo.
If
you
want
to
go,
look
at
it.
A
If
you
don't
mind,
just
super
quick,
we
had
a
question
on,
like
some
people
are
new
to
this,
and
I
just
wanted
to
I'm
sharing
here.
This
is
from
the
docs,
which
I
link
to
the
docs
and
I'll
include
that.
But
this
is
as
simple
as
it
gets
like.
This
is
the
simple
hello
world
demo.
A
It's
basically
create
a
new
net
app
and
do
tie
run
front
end
and
that's
all
you
need
and
you've
already
got
this
dashboard
and
then
adding
to
the
solution,
adding
in
a
back
end
with
a
micro
services
and
tie
run,
and
it
just
discovers
because
it's
in
the
solution-
and
you
do
tie
run
you
get
a
dashboard
that
has
both
the
front
end
and
back
end
like
it's.
That's
the
kind
of
magic
of
what
glenn's
going
deep
into
right
now
is
the
like.
B
Yeah
meshico
there's
no
reason
why
I
didn't
use
the
building
terminal
other
than
I
didn't.
I
mean
this
keyboard
doesn't
have
a
tilde
key
which
doesn't
make
it
super
easy
to
get
to,
but
other
than
that.
That's
not
really
a
reason,
though,.
B
I
have
a
ducky,
I
have
a
ducky,
I
forget
its
exact
name,
but
it's
got
it's
10,
keyless
and
f
keyless.
B
So
I'm
trying
to
struggle
it
and
it
has
sa
profile
keys,
so
they're,
big
and
chunky.
So
it
works
pretty
good.
I
mean
I
just
tend
to
remap
things,
so
the
f
keys
aren't
really
a
problem.
B
A
B
B
One
of
the
ducky
me
is,
I
think
it
is
that
I
have
for
those.
I
don't
believe
it's
the
ducky
shine,
but
it
is,
it
has
big
it's
the
big
ducky
shine
is
full
full
size
right.
This
one
is
is
much
smaller,
so
it's
10
keyless
it's
it's
kind
of
weird,
like
it's
10
keyless,
but
it
it
and
missing
the
f
keys,
but
it
does
have
like
home
and
ends
home
and
end
down
up
here
and
a
lot
of
the
ones
that
are
missing.
All
of
these
keys.
B
A
Cool,
I'm
sorry
to
distract
and
while
we're
at
it
would
you
like
to
talk
about
bananas.
B
We
best
not
start
a
new
topic,
then
not
only
is
going
to
take
a
whole
hour.
We
could
talk
about
avocados,
but
I
know
you
don't
like
them,
so
we
won't
do
that.
B
Okay,
so
let's
see
what
other
questions
do
we
have
do
we
just
round
it
out
with
a
few
more
questions.
A
B
So
you
should
try
it
try
the
local
dev
experience.
You
should
be
able
to
make
a
functions,
app.
There's
an
example
in
the
github
repo
for
thai.
B
You
should
be
able
to
go
run
it
so
I'd
like
you
to
go,
run
it
and
try
it
and
do
the
survey.
If
you
can
like
or
just
tell
us
if
it
was
good
or
bad.
You
can
always
tweet
me
about
tai
and
I'll
probably
respond,
probably
about
95
chance
that
I'll
respond
to
you.
If
you
tweet
me
about
tai,
the
probably
higher
than
that
actually
I
have
about,
I
have
thousands
of
unread
emails,
but
no
unread
tweets
for
so.
A
B
B
The
so
try
it
out
and
let
us
know
how
it
goes,
because
it
is
new
and
we're
still
working
on
it
like
we
didn't
get
to
deployment,
as
you
saw
and
so
play
around
with
it,
and
give
us
a
give
us
that's
the
best.
The
best
thing
you
could
do
in
general
with
tai
is
use
hi
and
tell
us
how
it
goes
or
tell
us
how
it
does
or
does
not
match
your
expectations
or
what
you
expect
it
to
do,
that
it
doesn't
like
just
talk
to
us
about
it
and
try.
A
A
B
Yeah-
and
so
I
didn't
talk
about
this
much
but
that
that's
one
of
the
things
that
we
will
be
working
on,
eventually
we're
a
very
small
team,
so
we'll
see
not
sure
when
this
will
happen
yeah,
but
we
I
expect
there
to
be
tooling
in
the
various
ides
like
that
lets
you
run
a
thai
app
and
attach
to
a
process
kind
of
imagine
via
a
click
rather
than
a
you
know,
attach
more
manual
process
right
like
attaches
a
few
clicks
in
any
of
the
ids,
including
vs
code,
whereas
we
can
probably
build
something
more
expensive
more.
B
You
can
imagine,
for
example,
in
vs
code
having
a
tie
button
on
the
left,
which
would
list
all
of
the
apps
list
all
the
process.
All
the
projects
you've
got
and
you
could
or
maybe
run
and
it'll
list
all
of
the
list,
all
of
the
processors
that
are
running
and
you
can
just
click
one
and
it'll
attach
to
it
or
something.
There's
things
like
that.
You
can
imagine
tooling,
like
that.
B
You
could
go
as
far
as
having
everything
that
the
dashboard
does
in
tooling,
if
you
really
wanted
to,
but
with,
but
more
unique
value
is
just
getting
the
debugger
attached.
So
we
expect
some
tooling
like
that
to
exist
over
time
and
that's
where
people
are
asking
about
vs
mac
and
vs
code
and
vs
visual
studio.
There
is
no
actual
special
support
for
thai
in
any
of
the
editors
right
now,
but
over
time
we
expect
to
hopefully
get
that
for
all
of
them.
A
A
Cool,
I
I
just
I'm.
I
love
sharing
your
docs
because
they're
amazing
this
one
just
for
people
that
don't
know
like
the
getting
started
experience
you
need
net
core
31,
you
install
a
global
tool,
yeah
or
or
update.
If
you
already
have
it
installed,
and
then
you
do
this
basic
tutorial
and
you
can
do
this
in
literally
five
minutes
under
five
minutes
like
and
that's.
B
A
He's
done
a
lot
of
really
cool
videos
too,
on
the
the
azure
site,
where
he
explains
kubernetes
and
he's
writing
on
a
whiteboard
and
stuff,
and
it's
just
it's
really.
B
Yeah,
that's
part
of
that
shares.
He
used
that
under
the
same
banner
as
shanes.net
microservices
cloudnative.net
show
right
something
like
that.
Don't
they
have
like
the
cloud
native
show.
That's
so
that's
what
we
called
it.
The
idea
started
out
as
like
cloud
native.net.
I
think.
B
A
Cool,
let
me
see,
I
don't
know,
questions
wise
or,
if
there's
other
anything
else,
you
want
to.
B
A
I
was
just
saying:
do
these
tutorials
it's
like
a
lot
of
the
time,
I'll
hear
about
something
neat
and
then
I'll
say:
oh
you
know.
Maybe
I
should
do
this
tutorial
and
it's
like.
I
don't
have
an
hour
to
do
this.
You
can
do
in
five
minutes.
It's
crazy,
easy,
like
and
I
dummy
tested
these
tutorials
for
you
on
a
live,
twitch
stream
and
made
all
kinds
of
mistakes,
and
then
you
guys
fixed
it.
B
Yeah,
we
did
oh
last
point
that
I
was
going
to
make,
because
I
mean
I
did
the
debugging
thing,
because
lots
of
people
were
asking
about
debugging
or
said
they
had
struggled
a
lot
to
get
it
working.
So
I
wanted
to
show
it
working
easily,
at
least
once
another
common
question
and
misconception
is
that
we
have
special
support
for
redis
or
postgres,
or
any
of
these
things
that
we
show
in
our
demos
we
don't
they're,
just
containers
that
we
run
so
like
what
we
do
with
redis.
B
You
can
replace
redis
with
anything
that
runs
inside
a
container
and
if
you
need
a
connection
string
or
something
there's
that
connection
string
support
like
we
don't
actually
do
anything
special
for
them
at
all.
Ty
doesn't
know
about
them.
It
doesn't
special
case
them.
It
doesn't
have
any
support
for
them
right.
So
it's
not
that
it
supports
red
essay
or
like
postgres.
B
It
supports
running
containers
and
then
giving
you
configuration
to
be
able
to
talk
to
them.
If
you
need
to
talk
to
them
well,
it
does
have
some
special
support,
for
is
things
like
dapper,
where
there's
an
extension
specifically
built
for
dapper.
So
that
we'll
run
all
of
the
dapper
side,
cars
along
with
your.net
application
and
some
things
like
that,
what
we
can
another
thing
that
I
was
going
to
mention.
B
A
Nice,
that's
funny,
apparently
look
somebody
said
the
stream
broke
on
youtube
again,
but
it
seems
to
be
going
for
me.
One
last
question
is
you
know
you
mentioned
like
there's
the
extensibility
for
stuff
like
dapper.
A
B
Not
really
like
there
is
an
extension
concept
within
tai,
but
it's
not
like
a
polished
experience
where
you
can
kind
of
acquire
things
and
add
them
to
the
mix
or
anything
like
that.
But
it's.
B
That's
the
thing:
if
your
thing
is
just
like,
even
the
dapper
support
right
like
you,
could
make
dapper
work
without
it
just
by
running
the
apps
and
putting
in
the
side,
cars
and
having
them
all
attached
and
stuff
right
like
so,
I
think
at
least
that
you
could
make
that
would
work
without
it.
I'm
pretty
sure
you
can
yeah,
so
I'm
not
sure
that
you
would
need
an
extension.
It
would
super
depend
on
what
you
were
doing.
A
B
Oh,
that's
another
thing:
yes,
brave
brave
cobra.
Thank
you
so
much
because
you
reminded
me
about
the
thing
that
I
was
going
to
talk
about.
We
don't
have
I've
said
before
occasionally,
probably
badly
that
we
don't
have
great
support
for
multi
repo,
like
we
tend
to
expect
you
to
run
everything
at
the
same
time.
B
It's
not
strictly
true
right.
We
do
have
a
dock
and
if
you
go
to
the
github
again,
if
you
have
the
github
open,
still
john
the
github
page
in
docs
somewhere,
so
in,
if
you
click
there
is,
there
is
some
reference
stocks.
If
you
click
reference,
oh
scroll
down,
actually
just
scroll
down,
because
I
think
we
have
it
on
this
main
one
yeah
we
have.
B
We
have
multi-repo
scenarios
here,
see
we
have
a
dock
for
it,
the
last
dock,
and
we
so
here
we
talk
about
how
and
if
you
scroll
down
a
bit,
we
talk
about
different
things.
You
can
do
to
kind
of
have
multi-repos
and
have
them
work.
B
Yeah,
because
we
know
that
obvious,
we
know
that
some
people
like
multi-repos
or
some
people,
don't
so.
I
would
call
this
support
like
we
have
some
stuff,
I'm
not
sure
it's
necessarily
the
right
stuff
yet
or
whether
it
needs
polish
or
whether
it's
definitely
like
we've
nailed
it
like.
We
haven't,
I'm
not
as
confident
that
we've
nailed
this,
as
I
am
confident
that
the
dev
experience
for
a
single
repo
is
pretty
good
is
going
to
be,
is
going
to
be
very
good
right.
B
A
All
right:
well,
that's
probably
a
good
place
to
wrap
up
I'll
update
the
community
links
thing
that
I
shared
to
also
include
these
link
over
to
the
docs
and
then
for
people
that
are
following
along
like
the
repo
right
and
then
like.
As
you
mentioned,
they
can,
they
can
tweet
you,
your
twitter
handles
right
there
on
the
screen.
A
Yep
and
then
we
are,
you
know,
planning
on
kind
of
making
things
more
regular,
because
we
we
have.
These
kind
of
standard
shows
right.
We
have
like
micro
services,
focus
and
blazer
focus
and
stuff,
so
we're
planning
on
kind
of
making
those
more
of
a
regular
rotation.
And
so,
if
people
watching,
if
you
have
ideas
or
things
you'd
like
to
know
more
about
microservices,
there's
some
areas
where
we
could
dig
into
more
too
so.
B
Yeah,
absolutely
I
mean
there's
more,
we
can
we
can.
We
can
there's
lots
more,
we
can
talk
about,
but
people
if
people
tell
us
what
they
want
to
hear,
we
can
always
talk
about
what
they
actually
want
to
hear.
Is
that
just
whatever
we
feel
like
wonderful
show
about
keyboards,
I'm
sure
I'm
sure
there.
A
You
go
yep,
my
keyboard,
they
probably
all
do
now,
but
mine
runs
arduino
has
an
arduino
board
inside
it.
You
can
customize
the
firmware.
You
can
see
it's
important
all
right.
Well,
I
think
that's
it.
Unfortunately,
now
with
our
new
system,
we'll
need
to
do
some
more
work
to
create
a
dramatic
zoom
out
effect,
but
I
I
do
have
thanks
for
watching.