►
From YouTube: Deprecations future
Description
OpenJS Foundation Collaborator Summit, Berlin, 2019
More details: https://github.com/nodejs/summit/issues/153
So far our user experience was partially meh, when it comes to deprecation warnings. We often struggle with deprecations due to our very heterogeneous users. There is often a gap in the requirements for e.g., CLI users, big companies that want the most reliable and safe way doing things, individuals who e.g., write a small script and so on. Most important though: we mainly reach the wrong people when printing our deprecation warnings. In general there are lots of pros and cons with deprecations.
That's why I would like to brainstorm what we could do to:
* reach the right people
* improve the average experience for all users
* still have a good way to improve the Node.js code base
So far I already thought about e.g., using a Node.js survey to gather further data and to figure out what our users actually believe and also about add an option to opt-into providing data when it comes to deprecations or other usage data.
This is very likely not an easy task and I ask not only Node.js collaborators to join but people who e.g., have run into deprecation warnings before, who found them distracting, great, weird and people who have e.g., strict requirements in companies. I believe we need a lot more feedback on this topic! If anyone has a lot of experience in this field, great, please join this session!
B
Coming
we
are
here
to
talk
about
the
duplication,
to
notice
and
how
we're
dealing
with
that
in
the
future.
So
currently
we
have
a
depreciation
system,
some
rules
for
that.
The
typical
no
justification
works
like
this,
so
you
have
a
teacher
that
is
documented,
ideally,
maybe
not
if
it's
not
a
community
probably
skip.
The
first
step
is
the
first
that
is
to
market
is
duplicated
in
the
documentation.
We
typically
do,
then
assembler
major
change,
sometimes
several
miners.
We
feel
that
makes
more
sense,
but
usually
then
it's
marketed
in
the
documentation.
B
C
B
B
B
Yeah,
so
why
did
I
actually
want
this
session
because,
as
I
already
said,
is
and
then-
and
we
are
often
struggling
with
actually
reaching
the
persons
we
want
to
reach,
and
then
they
are
often
not
efficient
in
a
way
that,
even
if
they
are
printed
out
to
someone
and
people
would
not
follow
up
on
implementing
the
change
that
we
recommend
and
we
can't
afford,
or
we
may
not
to
break
the
ecosystem
all
the
time.
So
there
is
like
a
lot
of
things
that
we
have
to
take
into
account,
while
making
duplications
and
I
mean
right.
B
Now,
it's
more
than
support.
We
should
definitely
try
to
improve
the
experience
and
a
lot
of
areas,
and
this
is
hopefully
what
we
can
figure
out
in
this
session.
So
yeah,
like
some
people,
might
profit
from
duplications
more
than
others,
and
that's
also
something
that
we
have
to
figure
out
and
in
a
way.
We
want
to
also
get
input
from
every
one
of
you
and
to
know,
and
what
your
experience.
A
B
D
B
E
B
Which
we
used
to
like
sitting
as
a
tool
for
anyone
who
doesn't
know
it
and
that
it
runs
Adam,
node
modules,
new
test
Suites
against,
occur
in
a
specific
node.js
module,
and
then
we
check
if
the
test
suite
still
passes
with
dead
node
model
and
adjustment,
it
is
a
tool
to
pretty
much
run
a
regular
expression
over
whatever
one
you
want
to
choose.
We
have
like
separate
and
dump
step
from
the
whole
ecosystem
and
it's
all
node
modules
out
there.
B
We
have
them
sorted,
I,
hear
by
the
top
list,
pretty
much
so
all
modules
that
have
a
specific
requirement
of
likes
and
it
would
be
preferable.
So
we
explicitly
trying
to
not
break
any
of
those,
and
then
we
can
just
run
it
against
a
whole
system
to
see
if
specific
API
is
used
in
a
while
a
lot.
But
of
course
we
do
not
know
anything
about
private
code,
and
we
also
do
not
know
anything
and
like
then,
when
you
get
the
output
from
just
minute.
It's
also
that
you
have
to
first
look
through
every
entry.
B
If
it
actually
it
finds
a
lot.
Sometimes
there
is
transpile
code.
Sometimes
it's
a
different
API,
which
is,
and
just
has
a
similar
name
as
a
note
for
API,
and
we
have
to
first
of
all
filter
out
all
false
positives,
and
this
is
not
easy
like
it's.
It's
really
really
tough
situation,
and
sometimes
we
have
duplications
which
make
a
lot
of
sense
in
a
way
for
reason
like
they've
guys
broken,
but
otherwise
we
just
do
it
because
that
we
have
two
versions
of
it
with
the
same
name
and
well.
B
F
F
B
B
B
G
Like
it
will
send
some
requests
to
a
controlled
by
us,
but
like
people
may
open
to
use
it
in
their
pipelines
like
what
people
would
do
with
coverage.
Of
course
they
would.
You
know
after
they
run
the
test,
it
will
use
a
module
that
provided
by
some
current
service
providers
and
send
requests
to
the
servers
and
plan
coverage
that
I
want
I'm
pondering
whether
we
could
do
something
sooner
to
attach
and.
G
B
I
B
C
B
Be
it's
not
really
doing
what
we
want
to
have
or
we
have
it
duplicated
and
then
someone
might
come
up
with
a
PR
and
then
we
officially
have
the
stree
steps
or
documentation
and
duplication,
runtime,
duplication
and
then
removal,
which
is
also
like
we
have
very
like
when
does
it
happen?
Does
anyone
really
do
it
Juiz
later
on,
say
this
was
a
bad
idea.
J
B
K
I
B
G
J
J
B
Yeah
the
main
theme
that
I'm
coming
from
is
we
see
a
lot
of
data
as
the
community
log
data
in
the
open
source
space
that
help
form
on
it
and
he
is
incisions.
But
from
my
you
know,
guesstimate,
that's
what
we
less
than
10
percent
of
meters
into
actual
the
world
most,
that
is
in
private
enterprise
businesses.
Teams
of
hundred
of
the
offers,
don't
think
thousands
of
operations.
B
And
you
just
right
now
we
were
already
and
talking
about
implementing
something
in
tools
like
NYC
and
when
you
round
the
cameras
report,
you
could
automatically
to
a
fire.
Every
indication
warning
and
the
duplication
warning
itself,
certainly
even
like
what
you
should
do.
Probably
in
that
case
it's
like
a
often
also
to
have
the
tracing
for
what
code
land,
but
we
could
remove
every
information,
there's
actually
really
about
the
application.
You're
running
and
and
I
mean
do
we
need
to
trace,
in
this
case,
probably
not
yeah,
because
unique.
We
want
to
have
anything
probably.
K
B
B
K
C
B
We
did
we
did.
We
did
a
lot
of
data
to
tell
us.
We
actually
scraped
a
lot
of
data
of
get
out.
We
skipped
off
with
it
off
at
the
end,
did
a
lot
of
internal
white,
charts
and
automation
and,
like
all
sort
of
information
for
us
to
understand,
what's
gone
on
our
ecosystem,
because
I
have
hundreds
and
hundreds
and
thousands
of
applications,
there's
no
way
for
me
to
do.
Calculus
I
needed
that
ring
question
from
Metis
and
I
mean
what
do
you
think
about
our
duplications
in
general?
Do
you
think
they
useful?
B
B
Versions
of
them,
that's
the
second
tier
problems.
First,
your
problem
is
even
getting
this
structure
business
and
pumping
and
keep
it
building
perspective
is
just
getting
over
all
the
riddles
and
things
that
was
a
struggle.
So
people
were
able
to
do
that.
The
next
year
becomes
the
application
challenge
of.
Is
it.
F
B
B
H
B
B
B
B
B
B
A
B
J
J
B
K
B
B
B
K
B
K
H
H
B
B
And
I
totally
in
this
case
and
I
see
we
also
have
here
somewhere
and
then
MIT.
The
point
argued
vacations
useful
at
all,
because
should
be
sometimes
just
say:
okay,
let's
break
the
APR.
Is
then
what
we
want
to
do
instead
for
some
API
is
because
if
people
don't
follow
up
on
two
vacations,
if
it's
just
noise
so
to
speak
on
the
terminal
and
just
annoying
to
people,
is
it
really
like?
This
is
just
depression?
I,
don't
say
this
is
right
or
wrong.
B
You
know,
but
it's
something
I
want
to
get
some
feedback
on
as
well,
because
having
a
strict
policy
about
that
makes
it
difficult,
depending
on
what
API
we
use,
because
sometimes
it's
rarely
used
API
and
if
we
duplicate
that
it's
probably
it
doesn't
hurt
their
system
so
bad.
But
then
we
have
other
API
is
there
he
is
so
often
and
that
removing
it
would
break
it
always.
B
Why
would
it
always
break
because
a
lot
of
modules
are
not
maintained
and
they
are
all
used
in
production,
which
is
a
big
problem
and
probably
something
we
should
discuss
at
some
point?
How
can
we
overcome
that?
Well,
it's
not
only
about
updating
the
note
version
itself;
it's
also.
How
can
we
make
sure
to.
B
A
F
H
B
B
F
F
B
B
B
E
B
C
B
So
we
are
in
the
green
keeper
circle:
I,
don't
Nick!
It
has
done
a
lot
of
manual
outreach
working
like
for
the
buffer
icon,
circular
depreciation
and
like
if
it
feels
like
that
I
wouldn't
leave
that
to
intrude
kind
of
frustrating.
You
have
to
interact
a
lot
with
the
maintainer
themselves,
yeah
it'd.
B
Derrick,
follow
up
on
that
one,
so
I
personally
think
in
documentation.
Duplications
are
normally
completely
useless,
mostly
speak
because-
and
we
already
like
and
last
week
on
a
conference
I've
been
at
the
main
problem-
is
no
chance.
Darts
people
don't
look
at
it
and
if
we
document
that
it's
duplicated,
no
one
will
change
anything.
B
They
would
even
implement
sometimes
the
documents
apart,
and
so,
if
we
did
not
document
an
API
before
and
now
have
it
in
the
dots
where
we
say
hey,
this
API
is
duplicated,
then
some
people
might
actually
use
it
instead
of
not
having
used
it
before,
because
they
looked
at
the
docks.
So
it's
like
the
opposite
of
what
we
actually
want
to
achieve
and
that's
I
really
and
I
think
we
should
not
have
documentation
duplications
in
90%
of
the
cases
most
of
the
time.
They
are
not
useful.
G
B
So
I
guess
this
comes
in
two
types
of
duplications:
yeah.
We
have
different
types
and
that's
the
problem,
because
sometimes
we
have
human
facing
ones
which
are
consuming,
for
example,
in
the
battle
yeah
you're
using
the
repple
you're
in
the
actual
end
user,
and
there
is
something
whyever.
I
don't
know
what
in
the
repple
duplicated
and
now
you
get
to
notification
right
away:
that's
the
user
that
should
receive
the
notification
awesome,
but
then
we
have
the
buffer
one
and
you
know
it's
like
in
which
is
buffer
who's.
B
Actually,
the
person
who
reaches
it
could
dependent
like
it
could
reach
anyone,
and
it
could
be
like
a
deep
down
dependency
where
you
don't
and
have
any
influence
on
the
maintainer.
You
cannot
change
it
and
for
whatever
reason
you
are
also
bound
to
using
their
modules
company
pods,
and
then
we
have
api's,
and
then
I
use
the
modules
used
by
applications
different
frequencies
of
usages.
So
all
these
implications,
I
think
we
should
start
thinking
not
in
the
generic
term
for
duplications
but
more
about
it.
B
F
H
B
B
B
B
Deactivated
automatically,
if
you're
doing
that,
already
right,
if
you're
building
that
into
two,
for
example,
by
again
agent
you
might
as
well
difficult
is
of
course
is
if
this
is
something
that
usually
to
set
up.
Yes,
he
did.
He
might
as
well
just
set
it
up
for
his
no
process.
Yes,
the
problem
is
that
how
do
we
meet?
You
keep
Dennis
name,
so
it
would
be
something
on
top
to
the
feature
itself
is
implemented
in
fourth,
but
and
to
actually
opt
into
it
it's
like,
but
how
do
they
opt
into
if
it
didn't?
D
C
D
B
D
B
We
can
see
like
a
nice
with
colors
and
grab
and
like
a
nice
grass
like
where
would
we
have
this
coverage
and
others
in
super
nice
apply
to
that
kind
of
coverage
is
also
about,
like
which
a
guy
who
uses
I
guess
like
so
it
would
actually
be
nice
for
me
just
for
me,
as
a
user,
to
have
the
same
information
and-
and
so
it
actually
makes
sense
for
for
coming
that
codecop
to
kill
that
in
yes,
because
me,
as
a
user,
I
would
actually
like
that.
So
how
can
we
sell
the
actual
implementation?
B
B
B
A
lot
of
data
that
is
normally
not
accessible
and
then
they
they're
tweeting
about
it.
So
codecop
could
potentially
there
they're
collecting
all
this
data
anyway,
they
could
aggregate
it
and
they
could
send
it
to
to
the
no
foundation
open
gates,
foundation
anonymously.
It
doesn't
have
to
say
which
project
is
using
and
just
say
there
is
ten
percent
of
our
users
running
on
our
system
are
using
this
API.
Yes,
you
know
and
I'm
not
even
sure
they
would
have
to,
depending
on
their
privacy
policy,
but
I'm
sure
they,
because
him
already
have
access.
They.
E
B
B
B
B
A
B
To
actually
reduce
burden
for
us
as
containers
to
sink
about,
and
in
this
case,
if
you
see
that
API
is
actually
not
used
in
the
wild,
then
you
would
be
able
to
just
change
it
would
stop.
Somebody
said
that
this
was
just
continuing
to
be
a
problem
going
forward
and
it's
just
not
grow
I,
don't
think
so
because
in
the
future.
A
J
B
D
No
agree
that
we're
not
doing
anything
like
if
you
look
at
what
front
end
is
doing
like
what
opponents
are
doing
and
they
have
like
this
problem
thanks
a
hundred
because
the
stuff
a
lot
more
often-
and
if
you
look
at
some
of
the
worst
offenders
like
angular,
they
have
pretty
decent
tooling.
That
will
actually
like
migrate.
Your
projects
from
API.
J
D
One
set
they
don't
like
parse
the
syntax,
then
it
sort
of
becomes
problem,
I,
guess,
they're,
looking
past
I've
script,
but
like
the
next
thing,
that's
not
like
directly
possible
by
the
tool
and
the
other
thing
they
have
is
like
compatibility
packages
like
react
on.
We
have
to
compare
and
then
like,
if
you
the
promised
dependencies
like
if
you
have,
for
example,
if
you
want
to
ever
definitely
the
next
step,
you
can
say:
okay,
this
API
not
develop
anymore,
and
then
you
have
a
million
packages
to
choose
next.