►
From YouTube: PSF TSC Meeting 1/20/21
Description
Agenda for this meeting:
https://github.com/Permissionless-Software-Foundation/TSC/issues/2
A
A
A
A
I'm
going
to
present
a
few
of
these
bullet
items
with
the
psf
core
and
joey
and
david
feel
free
to
interrupt
me
if
you
guys
have
anything
to
add.
As
I
go
through
these
items,
we'll
get
to
a
discussion
around
uptime
monitoring
and
transaction
history
by
address
that
joey
requested
to
add
to
the
agenda.
A
Sure
that's
a
good
idea
for
people
who
aren't
familiar
with
us
I'll
start
my
name's
chris
troutner.
I
founded
the
permissionless
software
foundation
as
well
as
fullstack.cache,
which
makes
developer
tools
for
developers,
particularly
javascript
developers,
who
want
to
build
on
top
of
the
bitcoin
cache
and
bcha
blockchains,
and
so
it's
a
wide
range
of
tools.
A
A
The
psf
blog
at
psfoundation.cache
go
to
the
blog
area
and
this
particular
article
is
called
towards
a
roadmap,
and
so
the
first
couple
items
in
the
agenda
speak
to
the
column
on
the
left
called
psf
core,
which
is
comprised
of
the
bch
api
rest,
api,
bchjs,
javascript
library
and
then
the
gatsby
ipfs
web
wallet,
which
is
the
sort
of
core
front-end
engine
for
building
web
apps
and
phone
apps.
That
can
interact
with
the
blockchain.
A
C
Yeah,
just
unfortunately
in
javascript
I
mean
since
since
this
is
primarily
a
web
tool,
library
and
a
lot
of
app
developers
are
going
to
be
working
with
javascript,
even
if
the
api
is
returning.
Good
information
and
number
form
it'd
be
very
easy
for
developers
to
start
screwing
with
those
numbers,
if
they're
not
using
the
right
library,
so
by
keeping
it
in
a
string.
C
You
know
it's
not
without
its
own
issues,
because
if
the
developer
just
takes
it
and
pretends
it's
a
number
they're
going
to
face
the
same
problems,
so
it
still
is
important
that
developers
use
the
big
number
js
functionality
when
they're
working
with
it.
Apart
from
documentation,
it's
just
kind
of
one
of
those
things
that
that
comes
with
the
territory
right.
A
Yeah,
that's
something
I've
struggled
with.
I
mean.
Obviously
I
need
to
update
the
api
documentation
to
reflect
this,
but
other
than
these
videos
and
the
api
documentation.
I
don't
know
of
a
good
way
to
communicate
these
this
type
of
nuance,
so
hopefully
that's
something
that
will
just
get
solved
as
we
move
forward.
C
A
There
is
a
config
folder
and
in
that
is
an
index.js
file,
and
here
is
where
the
environment
variables
get
overwritten.
So
the
way
javascript
developers
will
be
familiar
with
this
format,
where
you,
it
basically
checks
to
see.
If
this
environment
variable
exists,
if
it
does
or
if
it
does
not
exist,
it
will
use
the
default
value,
that's
defined
here
and
if
it
does
exist,
it'll
overwrite
that
value
with
the
environment
variable.
A
So
anyone
who
runs
an
instance
of
bch
api,
the
rest
api.
You
can
set
the
rate
limit
for
anonymous
users,
which
is
50,
so
it's
a
thousand
divided
by
this
number.
So
this
works
out
to
be
20
requests
per
minute.
A
thousand
divided
by
50
is
20.,
and
then
the
whitelist
rate
limit
are
people
who
authenticate
with
a
jot
token
and
in
the
case
of
fullstack.cache,
there's
a
the
process
for
getting
a
jot
token
and
then
and
then
the
the
rate,
in
the
rate,
limits
increase
to
a
hundred
requests
per
minute.
A
So
a
thousand
divided
by
ten
is
a
hundred,
and
then
this
new
environment
variable
here
are
whitelist
domains.
So
if
you
have
a
specific
domain
that
you
want
to
apply
high
rate
limits
towards,
like
in
in
my
case
with
fullstack.cache,
it's
anything
coming
from
the
fullstack.cache
domain
or
the
psfoundation.cache
domain
I
want
to-
I
want
those
to
be
in
my
whitelist.
C
Just
kind
of
the
same
same
theme,
you
know
eventually
we'll
need
to
get
that
stuff
documented
as
well.
I
mean
right
now:
it's
not
that
big
of
a
deal
because,
if
somebody's
using
it-
and
they
just
don't,
do
anything
about
those
variables,
the
app
uses
the
same
defaults
that
it
used
to
use
in
the
past.
So
it's
really
only
something
that
a
more
sophisticated
user
who's
trying
to
mess
with
something
needs
to
worry
about,
but
just
some
kind
of
weird
javascript
things
like,
for
example.
C
C
Not
so
it's
the
kind
of
thing
that,
like
if
you're
used
to
running
back-end,
node
apps
as
a
full
stack
developer
like
you'll
you'll
get
it
you
don't
need
you
don't
need
documentation
for
it,
but
if
you're
just
pulling
this
down-
and
you
want
to
run
it
and
you're-
like
oh
I'd
like
to
add
some
domains-
you're
probably
gonna,
run
into
that
issue.
So
not
really
a
new
thing,
but
just
the
same
same
theme.
That
will
need
some
kind
of
a
reference
for
the
stuff
going
forward
right.
A
A
It's
called
slp
avex
bridge,
it's
in
the
psf,
github
repo
and
so
far
that
library
can
mint
new
slp
tokens
and
burn
slp
tokens,
which
are
the
two
primary
functions
on
the
bitcoin
cash
side
and
now
he's
working
towards
setting
up
an
avalanche,
full,
node
and
then
creating.
And
then
he
will
create
tokens
on
the
avalanche
side
and
create
a
library
that
does
the
same
thing
on
the
avalanche,
side,
burning
tokens
and
minting
tokens,
and
so
once
that
basic
functionality
is
in
place,
we
will
be
able
to
do
asset
transfers
across
chains.
A
A
A
Okay,
next
item
on
the
agenda
is
slpdb,
which
is
the
indexer
that
controls
the
the
use
of
slp
tokens
there's
it's
recently,
james
cramer's
recently
released
a
new
version.
That's
optimized
to
work
with
nft
tokens,
non-fungible
tokens.
There
was
a
problem
that
was
discovered
a
few
weeks
ago
by
sweet
io.
A
They
are
a
phone
app
that
is
making
non-fungible
tokens
on
the
bitcoin
cash
blockchain
and
they're
producing
large
quantities
of
them,
and
it
was
through
that
process
that
we
all
collectively
discovered
that
slpdb
is
does
not
process
nfts
the
same
way
it
wasn't
optimized.
In
the
same
way,
it
used
a
lot
of
memory
and
it
caused
the
majority,
if
not
all,
of
the
slp
db
instances
around
the
world
to
essentially
stop
working.
They
stalled
out
at
specific
block
heights,
and
so
james
has
addressed
that
he's
released
a
new
version.
A
I've
stood
them.
This
new
version
up
at
full
stack
dot
cash
for
both
bch
bcha
and
testnet,
and
it
is
running
great
and
haven't
really
had
a
chance
to
do
any
sort
of
benchmark
testing
with
it,
but
it
did
sync
from
genesis
and
and
seems
to
be
running
fine
and
I've
updated
the
docker
slpdb
github
repo,
which
is
part
of
our
psf
core,
to
use
this
new
version,
and
I
will
be
releasing
database
snapshots
soon.
A
A
And
so
this,
the
cost
of
doing
this,
is
going
to
be
cheaper
for
the
psf,
because
they
we
pay
these
infrastructure
costs
and
it's
also
going
to
be
faster.
And
it's
going
to
allow
us
to
update
the
snapshots
more
often.
So
this
this
will
be
a
pretty
significant
upgrade
once
we
get
it
going.
Plus
we
get
the
opportunity
to
be
at
the
forefront
of
file
coin
as
the
technology
progresses.
A
Okay,
so
with
that,
let's
move
into
the
discussion
of
uptime
monitoring,
so
joey's
been
setting
up
sort
of
a
mirror
image
of
the
psf
core
on
the
bitcoin
abc
side
for
their
own
usages,
and
so
that's,
why
he's
been
making
so
many
valuable
contributions
to
our
code
base
and
we
got
in
we.
Let
me
introduce
the
the
build
verification
test
system
which
is,
which
is
what
I
built
for
to
monitor
the
infrastructure
at
fullstack.cache.
A
A
What
this
is?
It's
a
it's
a
brief
at
a
glance
summary
of
the
health
of
all
the
infrastructure
running
at
fullstack.cache,
so
it
it
does
these
liveness
tests.
It
checks
the
full
node,
it
checks
the
fulcrum
index
server,
it
checks
the
slp
db
and
it
does
that
for
mainnet
for
bch,
abc
and
testnet
and,
like
you
can
see
right
here,
it's
sending
me
a
warning:
hey
the
slp
db
on
test
net
has
fallen
behind
these
all,
say:
okay,
which
means
the
tests
passed.
A
It
runs
unit,
it
runs
liveness
tests,
unit
tests,
integration
tests
and
end-to-end
tests,
and
it
does
it
every
two
hours.
It
runs
it
basically
tests
every
thing
that
is
possible
to
test
and
then
gives
a
gives
a
status
update.
If
these
things
said
you
know
failed,
then
I
could
go
up
here
to
these
links
up
here
and
click
on
them
to
drill
down
into
the
logs
and
find
out
it
precisely
which
which
tests
failed,
and
so
this
is
available.
This
is
also
open
source,
and
this
link
here
will
take
you
to
that
repository.
A
I
haven't
done
a
very
good
job
documenting
this
because
so
far
it's
been
a
tool
for
me,
but
I'm
I'm
definitely
interested
in
better
documenting
it
and
collaborating
on
the
code
with
other
people
if
it's
valuable
to
other
people
joey.
What?
What
obstacles
have
you
had
in
terms
of
uptime
monitoring
and
what
features
are
you
looking
for.
C
So
we
need
to
have
some
sort
of
idea
if
the
back
end
has
degraded
performance
or
is
down-
and
I
guess
features
for
that-
are
one-
you
know
kind
of
immediate
notification,
if
say:
slp
isn't
anymore
over
some
other
components
off
or
like
a
test
started,
failing
or
some
reason
to
believe
that
the
api
is
down,
and
the
second
thing
would
just
be
logging
at
that.
So
we'll
have
some
sort
of
reliable
information
in
terms
of
like
hey
our
uptime
is
98.3
percent
early.
C
A
Yeah,
I'm
I'm
curious
to
hear
what
solutions
you
come
up
to
for
that
problem,
because
I
I
too
want
to
monitor
up
time.
It's
it's
not
so
much.
The
problem
doesn't
lie
so
much
in
the
actual
monitoring,
but
in
the
reporting-
and
I
mean
we
could
just
put
the
numbers
in
the
database-
that
would
be
the
conventional
way
to
do
it.
I've
thought
about
using
the
bitcoin
files
protocol
to
actually
write
like
a
small
json
blob
to
the
blockchain.
C
I
I
get
how
it's
cool,
but
yeah
it'll.
I
like
that.
If
the
problem
is
that
people
think
we're
misreporting
our
uptime,
maybe-
and
I
I
guess
that
could
be
a
problem
for
some
operators,
but
the
way
the
incentives
will
line
up.
It's
like
if
you're
running
the
software.
It's
because
you
want
higher
up
time
and
you
want
to
improve
it.
So
I
don't
see
it
as
such
a
big
risk,
but
I
I
get.
I
would
be
me
yeah.
A
Okay,
so
yeah?
So
that's
that's
the
system
that
that
I've
got
in
place
joey's
going
to
explore,
what's
most
appropriate
for
for
his
business
and,
like
I
said,
I'm
happy
to
accept
pull
requests
and
collaborate
and
if
there's
continual
interest,
I
I'll
make
a
bigger
effort
to
better
document
the
system
that
I've
built.
A
C
Like
if
you're
using
this
back
end
to
power,
something
like
a
wallet,
there's
pretty
specific
and
standardized
information
that
you
need
to
get
about
each
transaction
and
right
now,
there's
a
very
basic
transaction
history
endpoint
that
just
returns
a
list
of
txids
and
block
height
and
that's
a
start,
but
it
means
as
an
app
developer.
You'd
have
to
call
that
endpoint.
Then
you'd
have
to
organize
those
tx
ids.
Somehow
by
whatever
your
wallet
is
using,
then
you
have
to
get
more
information
for
each
txid.
C
Unfortunately,
the
amount
of
information
you
need
for
each
txid
is
kind
of
a
lot
and
not
straightforwardly
organized
right
now
in
bch
api.
So
some
things
you
need
to
know
is
hey.
Is
this
a
bch
transaction
or
is
this
a
token
transactions
one?
And
if
it's
a
token
transaction,
then
there's
a
whole
other
endpoint?
C
You
have
to
hit
to
figure
out
what
kind
of
token
transaction
it
is
and
if
it's
cash
transaction
there
is
a
details
endpoint
you
can
hit,
but
that's
missing
some
information
about
what
the
inputs
were,
and
if
you
don't
have
that
input
information
you
then
have
to
make
another.
You
know
at
this
point
kind
of
recursive
api
request
on
the
information
you
do
have
about
the
info
utxos.
C
So
it
means
it's
possible
for
an
app
developer
right
now
to
get
this
transaction
history,
but
it's
a
pretty
standard
thing
that
an
app
developer
or
user
of
vch
api
would
want
to
do,
and
it's
not
great
that
they
have
kind
of
this
significant
and
undocumented
end
around
burden
to
get
everything
to
work
right,
so
the
information's
in
the
back
end
it
just
it
takes
some
work
to
get
in
and
point
together.
That
will
then
give
app
developers.
Something
like
you
know,
get
history
for
address
or
an
array
of
addresses
and
just
return.
C
Everything
organized
with
like
this
is
an
incoming
vch
transaction
like
this
is
an
outgoing
token
transaction.
This
token
id
this
quantity
was
sent
from
here
to
here
and
that's
the
kind
of
information
that
users
are
used
to
seeing
when
they
pull
up
transaction
history
on
their
wallet
right.
They
don't
see
oh
well
utxo
index,
one
from
address.
Why,
when
like,
they
don't
need
that
they
just
need
to
say.
Oh
somebody
sent
me
some
money
or
I
sent
some
money
and
that
that's
it
so.
C
A
Yeah
and
that
at
the
api
level,
that's
that's.
Let's
take
a
minute
talk
about
this,
because
this
speaks
to
a
trade-off
that
I'm
constantly
wrestling
with
and
that
we
wrestled
with
also
when
we,
when
we
were
both
at
bitcoin.com,
there's
this
question
of.
Where
do
you
put
the
load
on
the
on
the
client
side
or
on
the
server
side?
So,
for
instance,
one
one
way
that
I've
dealt
with
this
or
or
I
have
made.
A
C
A
That
makes
a
worse
user
experience,
but
every
user's
treated
fairly
for
the
resources
that
they're
consuming,
and
so
that's
that's
where
it's
a
trade-off,
and
so
I
see
your
point
and
I
could
definitely
see
the
advantage
in
creating
an
an
endpoint
on
bch
api.
That's
just
like
give
me
the
address
of
the
sender
of
this
transaction.
That's
that's
the
question.
A
We
want
to
ask
the
server
and,
and
then
the
server
can
can
make
all
those
those
calls
which
it's
more
efficient
at,
but
it's
going
to
impact
the
ability
to
scale
the
server,
and
so
I
don't
totally
know
what
the
right
answer
is.
I
mean
I
think
it's
particularly
in
the
short
term
and
at
low
scale.
The
answers
clearly
put
the
load
on
the
server
improve
the
user
experience,
but
at
scale
that
decision
is
going
to
come
back
to
bite
us.
C
You
want
to
limit
your
api
calls
anyway
and
you're,
trying
to
make
the
app
more
efficient,
something
like
transaction
history
like
at
least
when
cash
tab
has
that,
because
server
costs
are
expensive,
whether
those
rate
limits
are
not
right,
like
you,
don't
want
to
just
be
hitting
it
with
everything
all
the
time.
It
would
be
to
store
a
lot
of
transaction
history
in
in
the
database
right
like
past
transaction.
C
I
don't
think
it
really
encourages
that
much
as
much
innovation
as
we
could
like
app
developers
some
of
these
things.
They
should
have
the
resource
for
it.
So
you're
like
right.
That's
how
that
piece
of
the
wallet's
gonna
work.
I
get
the
transaction
history
I'll
build
the
rest
of
it,
so
I
think
it's
more
of
a
resource
pricing
problem.
I
guess,
as
a
summary.
A
C
A
Right
well
and
the
rate
limits
are
essential.
That
was
where
I
think
we
we
went
horribly
wrong
with
developer.bitcoin.com
was
that
we,
we
didn't
take
rate
limits
too
seriously
soon
enough
and
a
lot
of
our
downtime
issues
were
just
due
to
abuse.
Just
because
we
didn't
have
the
rate
limits,
people
took
everything
they
could
take,
and
so,
when
I
started
building
the
psf
core,
those
rate
limits
were
like
number.
One
thing
that
I
built
and
and
as
and
I
was
pretty
gun
shy
to
open
up.
A
Like
I
don't
know
when
I
first
started
it,
the
anonymous
rate
limit
was
like
three
per
minute.
I
mean
it
was
super
super
conservative,
because
I
had
all
these.
You
know
ptsd
from
the
from
the
the
rate
limit
abuse
at
bitcoin.com.
So
it's
taken
me
a
while
to
get
brave
enough
to
increase
the
the
free,
the
free
tier
up
to
20
20
per
minute
right.
C
A
A
Okay-
well,
that's
great,
I
think,
lasting
on
the
agenda.
I'll
just
give
a
brief
review.
So
the
coin
party
hackathon
just
finished-
I
participated
as
a
contestant
that
was
pretty
fun.
It's
the
first
hackathon
I've
ever
actually
attended
as
a
contestant.
I've
always
either
been
a
judge
or
a
mentor
or
both,
and
that
was
a
lot
of
fun.
A
We
built
a
game
called
geo
drop,
dot
cash
and
it's
a
map
based
scavenger
hunt
very
similar
to
pokemon
go
except
instead
of
running
around
and
collecting
pokemon
you,
the
the
canonical
use
case,
is
like
a
coffee
shop
that
wants
to
promote
their
business
by
say,
dropping
like
free
coffee
tokens
around
their
their
neighborhood,
and
this
is,
I
think,
it's
been
one
of
the
best
received
projects
that
were
built
in
the
hackathon
and
I'll.
Just
leave
this
here.
A
I'm
going
to
put
some
more
links
in
the
about
page
to
some
of
the
videos
that
was
published
on
how
to
play
it
and
what
went
into
making
it,
but
it's
a
great
example
of
some
of
the
tools
at
fullstack.cache
we
used
the
bch
wallet
starter,
which
is
a
gatsby
starter.
We
figured
out
how
to
compile
this
into
a
native
android
app,
which
works
great
but
but
real.
A
You
can
use
the
cash
drop
page
to
create
a
campaign
and
and
drop
your
own
tokens
for
people
to
find
and
then
and
then
hitting
the
the
play
button
will
take
you
to
the
play
screen
where
it'll
show
you,
where
you
are
and
it'll
give
you
directions
to
the
to
the
closest
point
to
you
and
when
you
get
close
enough,
the
collect
button
becomes
enabled
and
you
get
an
slp
token
and
you
get
3000
satoshi's,
which
is
enough
to
then
like
send
that
slp
token
somewhere.
So
you
don't
have
this
problem.
A
A
What
this
is
it's
very
similar
to
the
slp
token
bridge
that
we're
building
they
were
able
to
take
slp
tokens
and
represent
them
on
the
ethereum
chain
as
wrapped
slp,
which
is
the
rage
right
now
for
like
moving
btc,
you
wrap
it
and
all
these
different
blockchain
assets
can
be
ported
over
to
ethereum
by
wrapping
them.
So
they
took
that
same
idea
and
they
applied
it
to
slp
tokens.
A
A
Let's
say
we
had
five
people
on
the
com,
com
and
five
people
on
the
technical
steering
committee
and
would
require
five
of
ten
members
to
sign
a
message
in
order
to
bring
new
tokens
into
existence
so
that
no
one
party
you
know
can
you
can
control
that
or
shut
that
down,
and
then
we,
those
newly
minted
tokens,
can
then
be
brought
across
to
whatever
chain
we
need.
We
need
to
use
them
on
so
that
that
was.
A
That
was
what
really
struck
me
with
this
project
and,
lastly,
I'll
speak
real
briefly
to
the
incenseable
web.
So
the
mission
of
the
permissionless
software
foundation
is
free
speech,
censorship,
resistant
and
economic
action.
This
really
spoke
to
all
three
of
them.
They
combined
james
cramer's
bitcoin
file
protocol,
which
is
a
way
of
storing
small
amounts
of
data
on
chain
up
to
10
kilobytes,
and
you
can
do
more.
That's
a
reflection
of
the
50
chain
limit
and
and
then
what
shimari
did
the
guy?
Who
who
did
this?
A
He
took
james's
idea
and
he
created
a
new
yaml
metadata
format.
So
yaml
files
are
used
in
like
docker
compose
they're
used
in
many
different
software
projects
to
sort
of
describe
at
a
very
high
level
what
the
program
is
supposed
to
do
so
in
docker
compose
you
use
this
for
composing
several
different
docker
containers
and
get
them
all
working
together.
A
And
I
actually
had
a
chance
to
talk
to
amri
about
this
idea
a
little
bit
and
because
I'm
concerned
with
this
idea
of
on-chain
data
bloat,
I'm
seeing
more
and
more
market
signals
of
people
who
want
to
do
it.
Even
given
the
constraints
that
exist
today,
which
I
think
are
fine
and
I
do
not
hope
ever
change
and
he
said,
there's
there's
nothing
magical
about
putting
data
on
the
blockchain,
because
eventually,
when
we
have
a
hundred
times
more
transaction
volume
than
we
do
now,
most
infrastructure
will
probably
run
pruned
nodes
and
pruned
indexers.
A
I
thought
that
was
a
really
good
point
as
well,
which
is
why
I'm
I
am
so
excited
about
the
work
I'm
doing
with
filecoin,
because
I
think
that
is
the
place
where
the
data
should
go
long
term,
and
so
it's
just
I'm
just
bringing
it
up
here,
because
this
idea
of
uncensorable
data
is
resonates
with
me
and
the
mission
of
the
psf
and
it's
an
innovative
new
twist.
But
it's
not
without
its
concerns.
A
B
A
Of
course,
it
does
so
yeah
anyways
that
was
worth
mentioning.
These
hackathons
don't
come
out
along
very
often
and
when
they
do
there's
usually
a
few
gems
left
behind.
I
thought
I
thought
these
were
that
okay.
Well,
I
think
we
can
wrap
up
this
meeting.
Is
there
anything
you
guys
want
to
add
before
we
wrap
this
up.
B
I
just
want
to
make
a
quick
comment
about
the
geo
geodrop
dot
cash.
I
had
a
chance
to
play
with
it
while
the
hackathon
was
going
on,
and
I
thought
it
was
pretty
unique.
It
reminded
me
of
the
times
in
2014
2015,
when
similar
things
were
attempted
on
the
btc
blockchain,
of
course,
that
that's
not
as
inexpensive
to
allow
that
to
happen
now,
but
yeah,
it's
it
felt
like
old
times.
So
I
just
wanted
to
have
put
across
my
little
smile
for
that
one
right
on.