[dart-misc] 8/19 DEP meeting notes

['Bob Nystrom' via Dart Misc]

Here's my notes from yesterday's wide-ranging DEP meeting:

https://github.com/dart-lang/dart_enhancement_proposals/blob/master/Meetings/2015-08-19%20DEP%20Committee%20Meeting.md

Cheers!

- bob

[Jan Mostert]

Quote: "I agree. It's not that we need a slower waterfall, it's that we
need more iteration, experimentation and data gathering before a design is
locked down. It doesn't matter how slow you go. If you don't get good
feedback, you aren't going to do good design."

Posting the proposals on the mailing list is already a step in the right
direction.
Adding some examples of how the new ideas are useful will also help - some
of the syntax proposals in the pass didn't make any sense coming from a
heavy Java-orientated background.

Post by 'Bob Nystrom' via Dart Misc
https://github.com/dart-lang/dart_enhancement_proposals/blob/master/Meetings/2015-08-19%20DEP%20Committee%20Meeting.md
Cheers!
- bob
--
For other discussions, see https://groups.google.com/a/dartlang.org/
For HOWTO questions, visit http://stackoverflow.com/tags/dart
To file a bug report or feature request, go to http://www.dartbug.com/new
To unsubscribe from this group and stop receiving emails from it, send an

['Bob Nystrom' via Dart Misc]

Post by Jan Mostert
Quote: "I agree. It's not that we need a slower waterfall, it's that we
need more iteration, experimentation and data gathering before a design is
locked down. It doesn't matter how slow you go. If you don't get good
feedback, you aren't going to do good design."
Posting the proposals on the mailing list is already a step in the right
direction.

Yes!

Post by Jan Mostert
Adding some examples of how the new ideas are useful will also help

I agree totally. There's a reason the proposal template has both motivation
and example sections. I don't think some of the proposals coming from the
Dart team have done a good job of focusing on that, to our detriment. I
hope we can improve that over time.

Cheers!

- bob

[Cristian Garcia]

Is Dart still on ECMA? How does Go evolve as a language? Maybe the proposal
process is too formal, it could start with some sample code and an
explanation (no need for a huge document), then an experimental
implementation, then a formal document, and finally the full implementation.

Post by 'Bob Nystrom' via Dart Misc

Post by Jan Mostert
Quote: "I agree. It's not that we need a slower waterfall, it's that we
need more iteration, experimentation and data gathering before a design is
locked down. It doesn't matter how slow you go. If you don't get good
feedback, you aren't going to do good design."
Posting the proposals on the mailing list is already a step in the right
direction.

Yes!

Post by Jan Mostert
Adding some examples of how the new ideas are useful will also help

I agree totally. There's a reason the proposal template has both
motivation and example sections. I don't think some of the proposals coming
from the Dart team have done a good job of focusing on that, to our
detriment. I hope we can improve that over time.
Cheers!
- bob
--
For other discussions, see https://groups.google.com/a/dartlang.org/
For HOWTO questions, visit http://stackoverflow.com/tags/dart
To file a bug report or feature request, go to http://www.dartbug.com/new
To unsubscribe from this group and stop receiving emails from it, send an

['Bob Nystrom' via Dart Misc]

Post by Cristian Garcia
Is Dart still on ECMA?

Yes, all changes go through ECMA Technical Committee 52
<http://www.ecma-international.org/memento/TC52.htm> before they are
officially in the language.

Post by Cristian Garcia
How does Go evolve as a language?

As far as I know, they don't have a formal open standard process. They are
open source, and work with the larger Go ecosystem, but ultimately
languages decisions are made by the Go team.

Maybe the proposal process is too formal, it could start with some sample

Post by Cristian Garcia
code and an explanation (no need for a huge document),

I don't think the document needs to be *huge* but it does need to be pretty
detailed and precise. We want to push much of the burden of due diligence
onto the person proposing the change. They have the most incentive to get
the proposal to succeed and likely the most time to work on it.

If proposals are just rough sketches, it ends up falling onto the Dart team
to work out all of the details, and we don't have the resources to do that
for everyone's proposal. We'll help, of course, but we need to distribute
the work as much as possible.

The document doesn't have to be *done*, but it should be pretty well
thought out.

then an experimental implementation, then a formal document, and finally

Post by Cristian Garcia
the full implementation.

We are absolutely moving in the direction of having a working experimental
implementation that's been tried on real-world code before we consider a
proposal complete. This is obvious in retrospect, but almost every time we
*haven't* done that, even on seemingly innocuous proposals, we've
discovered problems that didn't come to light until the proposal was
implemented and used.

I think the pipeline should be something like:

1. Send out rough sketch to core-dev.
2. If people seem to think it has merit, author writes proposal detailed
enough to be implementable.
3. Experimental implementation. If we run into problems here, go to 2 or
bail.
4. Try implementation on real programs. If we run into problems here, go
to 2 or bail.
5. Refine implementation to something shippable. If we run into problems
here, go to 2 or bail.
6. Decide if the experiment succeeded and we want to accept the proposal
in the language. If not, go to 2 or bail.
7. Accept the proposal.

By the time a proposal is accepted, we should be absolutely confident there
will be no surprises.

Cheers!

- bob

[Felipe]

Bob, would not be nice to have a separate group for proposals (eg.
***@dartlang.org)?

A place where the community makes its proposals, and if they are feasible
then someone (???) generates a ticket in "???" for further review by the
dart engineers. If the ticket is accepted, then a DEP is created. Otherwise
the proposals will begin and end in a post.


Thanks.

- Felipe

['Bob Nystrom' via Dart Misc]

Post by Felipe
Bob, would not be nice to have a separate group for proposals (eg.

core-dev@ basically is that list. It just doesn't *require* a topic to be
in full proposal form in order to be discussed. That's good because
starting with an informal discussion saves us all a lot of time on ideas
that won't pan out. It would be a waste of effort to write a full proposal
for many of those.

Cheers!

- bob