►
From YouTube: hyper TSC Meeting April 6, 2021
Description
This is a recording of the hyper April 6 Technical Steering Committee meeting. We talk about the current status of hyper and the upcoming focus to a v1 release. Also some insights on how our Free Developer service for hyper will work.
If you are interested in finding out more about hyper, you can visit https://hyper63.com to learn about the vision and https://github.com/hyper63/hyper63 to dig into the open source project.
A
A
A
It
came
out
a
while
ago
from
the
inventor
of
rabbitmq
or
whatever,
and
it's
like
it's
just
super
nice,
because
it's
all
tcp
ports
and
you
just
send
it
to
a
port
and
it
takes
care
of
the
batching
and
cueing
for
you,
and
then
it
just
admits
to
any
workers.
And
it's
just
just
a
really
nice
pattern,
and-
and
I
hammered
it
when
I
was
working
on
the
implementation
and
what's
nice
about
it-
is
there's
no
zero.
There's,
no
single
point
of
failure
right.
A
A
A
A
Yep,
so
so
this
is
the
design
they
built
these
graphics
and
then
kind
of
created
this
kind
of
flow,
which
is
really
cool.
Where
you
know
you
got
all
these
challenges
flow
into
hyper
and
it
flows
down
to
all
the
the
ports.
And
then
you
click
on
each
port
and
you
can
kind
of
see
what
each
port's
all
about.
A
So
I
thought
that
was
pretty
cool
click
and
then
they've
got
got
this
and
then
the
faq
and
one
of
the
cool
things
was
was,
is
like
wanted
to
be
simple,
pragmatic
and
and
clean.
So
they
really
focused
on
on
trying
to.
You
know,
use
graphics
to
draw
that.
So
there's
not
a
lot
of
photos,
it's
just
a
lot
of
neat
little
graphics.
B
B
A
There
is
and
we'll
work
on
the
copy,
but
I
basically
gave
the
copy
of
the
the
current
site
and
we'll
work
on
improving
the
copy
to
to
get
that
in
one
of
the
things
they
did
is
they
did
like
a
a
graphic
of
all
the
kind
of
the
benefits,
but
it
just
kind
of
looked
really
busy
right.
It's
like
all
the
illites
right
and
that's
like
well.
A
I
think
I
can
get
that
into
writing,
but
but
I
think
we
can
reword
some
of
these
things
and
and
put
open
source
in
there,
but
they're
not
copy
people.
They
were
pretty
much
all
design
focused
on
graphic
design
so
but
yeah
and
I'm
looking
forward
we'll
have
these
graphics
that
we
can
repurpose
on
other
pieces
too.
So
so
that's
cool
to
have
these
custom
graphics.
A
A
A
The
the
ts
is
is
wonderful.
If
every
single
library
you're
gonna
ever
use
is
ts
right,
but
but
then
there's
some
things
that,
like
the
currying
is
you
just
feel
like
you're
writing
so
much
more
code
than
you
have
to,
but
but
anyway,
I'm
sure
we'll
continually
to
re
revisit
that.
But
but
a
lot
of
people
have
found
that
they
they
don't
run
the
checker
as
much
as
they
just
run
es
build,
and
you
know
maybe
they
have
it
in
their
ci
to
run
the
checks.
A
It
was
requested
by
a
customer,
but
both
customers
are
signed
on
for
a
year,
which
is
great
and
both
both
of
them
will
be
in
production
by
the
end
of
this
month,
which
is
really
good
and
I
feel,
like
the
rest.
Api
is
kind
of
stabilizing
there's
a
couple
of
tweaks
that
I
would
like
to
make,
but
I
think
overall
it
seems
to
be
pretty
stable
and
I
think
the
the
data
search
and
cache
and
storage
and
queues
are
are
all
doing.
Well
again,
storage
is
one.
A
A
It
can
be
added
as
part
of
the
service
implementation
and
not
attached
to
core,
so
you
could
use
elastic
or
sentry
or
whatever,
whatever
monitoring
solution
you
want-
and
I
just
picked
elastic
mainly
because
I
had
some
experience
with
it
before,
and
I
also
want
to
work
on
so
it's
using
the
apm
side
of
elastic.
But
I
also
want
to
work
on
sending
the
logs
into
elastic
and
use
the
elk
stack
for
that.
A
I
think,
will
really
make
that
onboarding
experience
so
much
easier
for
for
developers,
so
kind
of
play
played
around
with
kind
of
these
types
of
things
where
it
really
kind
of
narrows
hyper
to
to
a
a
smaller
use
case.
If
you
will
so
so,
like
most
of
the
time,
you're
gonna
have
one
database
with
one
cache,
with
one
search
with
one
queue
right.
A
A
And
and
it's
kind
of
funny
it's
it's,
you
know
you
kind
of
build
out
this
generic
thing
and
then,
as
you
get
ready
to
launch
it,
you
you
narrow
it
down
to
become
more.
You
know
kind
of
opinionated,
but
the
the
cool
thing
is
is
putting
the
opinions
in
the
client
and
not
in
the
service.
That
way
it
can
kind
of
serve
many
opinions.
If,
if
those
use
cases
come
up
at
least
that's
the
goal.
D
A
A
Right
and
and
there's
really
this,
you
know
and
there's
there's
more
more
battles
going
on,
but
but
the
the
the
main
battles
are.
You
know
the
spa
versus
ssr
versus
serverless
right.
Those
are
like
the
the
main
battles
and
then
there's
you
know
maybe
nosql
and
postgres,
but
but
I
think
you
know
that's
kind
of
you
know
if
you
choose
postgres
you're,
choosing
postgres
and
you're
you're,
choosing
an
orm
you're
choosing.
A
You
know
all
of
these
tools
to
manage
that
postgres,
whereas
you
got
a
spa
where
you're
you
know,
building
out
an
api
and
separating
your
presentation
from
that
api,
pretty
clearly,
then
you
have
ssr
and
and
with
server-side
rendering
you
can
do
the
same
thing
and
I
think
actually
that's
the
best
kind
of
workflow
is,
is
to
build
the
ssr.
Just
like
you
would
build
a
spa
or
jam
stack
app
and
have
your
api
separate.
A
A
A
A
A
So
whatever
your
github
account
is
that's
the
name
of
your
app
and
then
this
connection
stream
is
everything
you
need
to
connect
to
your
hyper
instance
right.
So
you
copy
that
and
then
you
come
to
your
client
and
you
use
an
environment
variable
to
paste
that
string
in
and
then
the
hyper
client
actually
reads
that
that
string
and
parses
it
using
the
node,
url
parser
and
then
will
generate
a
token
for
you
when
you
make
requests.
A
So
all
you
have
to
do
is
and
I'll
actually
make
this
hyper
colon
colon.
Probably
so
it
doesn't
look
like
an
actual
url,
because
this
one
is
not
an
actual
url,
it's
more
of
just
a
one
line
config,
so
you
don't
have
to
have
five
environment
variables
and
then,
with
that
connection
string
you
can
come
in
and
you
can
choose
to
define
models
or
if
you
don't
want
to
define
models
you
don't
have
to.
A
A
B
You
know
one
line
and
and
maybe
add
a
model
unless
there's
like
a
like
an
easy,
easily
removable
default,
one
where
it's
like
you're
you're,
already
set
up
and
ready
to
go,
you
don't
have
to
do
anything
else,
but
also
have
having
something
like
this.
Where
they've
already
got
a
project
here,
you
go.
A
Get
pod,
I
o
basically
allows
you,
you
can
put
a
button
like
the
launched
heroku
button
on
your
readme
and
then,
if
someone
clicks
that
it
will
actually
fork
that
through
github
and
then
spin
that
project
up
for
them
in
a
visual
code
in
their
browser,
with
everything
ready
to
go
and
there's,
there's
others
like
this,
but
I
really
like
get
pod,
but
because
I
mean
it
literally
is
just
like
visual
studio.
It's
a
it's.
A
It's
got
a
you,
you
have
a
full
terminal,
you
have
you
know
everything
you
need
to
to
hook
into
it.
It's
it's
made
by
the
same
guys
that
made
eclipse,
so
they
they
know
ides
and
stuff.
I
might
play
around
with
that,
because
I
think
asking
someone
to
clone
something
and
run
it
on
the
machine,
and
you
know
do
all
that
is
even
just
a
little
bit
too
far
right.
It's
like
click
on
this
and
you
know:
you're
kicking
the
tires.
A
B
That's
fabulous
yeah,
oh
my
gosh.
I
mean
because
that
that
goes
back
to
the
thing
that
I
we
I
brought
up
earlier
early
on,
about
removing
the
barriers
to
entry
and
just
letting
people
like
play
like
having
a
demoe
test
area
that
they
could
play
in.
But
this
takes
it
even
a
step
further
than
that,
where
it's
like.
Here's,
your
own
demo,
retest
area,
one
click.
A
A
If
you
install
that
extension
in
in
your
browser,
then
you'll
get
this
get
pod
button
and
on
every
single
repository
that
you're
working
on
and
then
you
just
click
the
get
pod
button
for
whatever
branch
you're
looking
at
or
whatever
and
it'll
just
spin
up
that
workspace
with
that
code
or
that
pr,
you
can
make
specific
changes
to
that
and
submit
it
and
it
never
messes.
With
your
desktop.
A
A
A
The
one
workflow
I
would
like
to
go
over
is
the
multi-tenant
workflow
and
see
if
you
guys
can
poke
some
holes
in
it.
So
so,
basically,
what's
happening
is
it's
one
kind
of
hyper
service?
That's
multi-tenant
for
the
dev
kind
of
dashboard
and
the
way
that
I'm
coding
it
is
that
when
the
hyper
client
actually
goes
to
dev
with
a
request
right,
it
goes
to
list
movies
or
create
a
movie
or
whatever.
A
A
A
D
A
D
C
A
So
yeah
that
that's
captured
at
registration
time
here
and
you
know
actually
this
this
string
will
be
kind
of
close
to
the
audience
string
and
and
then
once
it
verifies
the
token.
It's
basically
making
sure
that
in
a
multi-tenant
system
that
you
know
you're
on
the
right
tenant
right
trip,
I
can't
make
a
request
to
t
wilson,
63
or
robert
pierce
can't
make
a
request
to
trip
out
right.
So
so
it'll
make
sure
that
you're
on
the
the
right,
tenant
and.
A
A
A
E
E
A
A
A
So
I'm
making
a
get
request
to
dashboard,
which
is
acting
as
the
auth
authorized
body.
If
you
will
the
identity
management
service,
so
I'm
to
take
the
key,
which
is
part
of
the
sub
of
the
token
and
I'm
going
to
call
the
dashboard
service
and
it's
going
to
give
me
the
secret.
Basically,
it's
going
to
give
me
the
secret.
It's
going
to
give
me
the
key,
and
it's
going
to
give
me
the
app
and.
A
E
A
E
A
E
So
if
I'm
understanding
this
correctly,
does
this
all
it
sounds
like
this
also
sets
up
like
for
in
the
future.
If
you
wanted
to
have
multiple
sets
of
keys
each
with
their
own,
like
set
of
scopes
like
maybe
one
set
of
keys,
could
only
access
like
cash,
or
maybe
you
want
to
segment
permissions
across
right
across
ports
or
maybe
like
specific
scopes
per
port,
but
I
think
this
sort
of
sets
up
eventually
moving
to
that,
because
you
could
just
it's
just
a
jwt,
and
so
you
could
just
have
scopes
as
claims
on
the
token.
A
Exactly
exactly
it
sets,
it
sets
it
up
for
for
the
scope
so
that
you
can
have
you
know
fine
grain
access,
like
maybe
a
a
read-only
user,
or
you
can
only
access
the
port
and
not
the
queue.
I
mean
the
the
search
and
not
the
queue
etc.
So
so
all
of
those
things
come
into
play-
and
you
can
also
you
know-
have
multiple
apps
like
right
now.
This
is
just
for
the
free
tier.
You
have
t
wilson,
you
know
your
name,
but
then
in
other
tiers.
B
D
A
Cool,
so
that's
that's
everything
that
I
had
so.
My
goal
is
to
try
to
get
all
of
this
up
and
running
by
the
end
of
this
week
and
then
work
on
you
know,
kind
of
doing
the
figma
design
and
pulling
that
into
the
webpage
the
new
website,
putting
all
that
together
so
that
that's
really
the
focus
of
april
is
is
those
two
things
is
to
have
a
end-to-end
kind
of
workshop
guide
platform
with
hyper
and
have
the
new
website
and
the
only
other
things
that
are
kind
of
entering
the
picture
is.
A
The
other
thing
is
scheduled
jobs
right,
so
the
queue
does
a
very
simple
push:
pull
that
works.
Great
for
serverless
environments,
but
I'm
still
struggling
with
the
scheduled
jobs
and
I
sort
of
feel
like
I'm
overthinking,
that
one
just
because
there
are
some
use
cases
that
that
I
need
to
do
that.
But
I
don't
think
that
that's
a
a
major
use
case
for
like
startups.
For
example
off
the.
E
D
A
But
if
you
don't
have
an
off
solution,
then
you
probably
want
it
in
the
box.
So
I
kind
of
see
it
kind
of
both
ways
and,
like
you
said,
tyler,
the
scheduled
jobs
always
come
up
and
and-
and
the
thing
is,
is
how
do
you
you
know
it
takes
a
a
lot
to
think
through
that.
In
the
effort
of
you
know,
infrastructure,
agnostic
right,
like
support
amazon,
google
digitalocean
render
et
cetera.
So
it's
it's
not.
A
B
A
Cool
any
thoughts,
any
other
thoughts
that
I'm
missing
really
trying
to
work,
that
road
map
to
get
to
some
sort
of
announcement
towards
the
end
of
april
early
may
to
announce
whether
it's
private,
beta
or
public,
beta
or
whatever.
Just
to
you
know,
try
to
get
some
developers
to
start
to
build
hobby
apps
with
it
and
then
really.
The
the
road
map
to
v1
is
is
still
more
targeted
in
october,
where,
where
we'll
have
like
a
full
sas
platform
built
out.
B
Yeah
are
you
thinking
of
engaging
of
engaging
more
like
individual
developers
and
getting
them
to
drive
stuff
at
their
companies,
or
are
you
like
from
the
bottom
up
or
because,
let's
face
it,
we're
at
the
bottom
or
the
other
way
around?
Like
approaching
you
know
exec
level
people
and
having
them
like
dictate.
A
A
little
above,
but
I
think
for
for
this
period
right
for
this
phase
of
beta,
if
you
will,
for
lack
of
a
better
term,
is
to
really
try
to
engage
at
the
developer
level
and
and
get
get
feedback.
That's
that
someone
has
actually
used
the
product
and
gives
you
feedback
right.
Test
drove
the
car,
opposed
to
looked
at
some
presentation
or
documentation
and
then
feedback,
because
I
think
that's
good
early.
But
but
at
some
point
you
need
people
driving
the
car
to
really
know
how
it's
going
to
perform
right
and
yeah.