New EEP draft: Pinning operator ^ in patterns

Nicolas Martyanoff khaelin@REDACTED
Fri Jan 29 09:41:33 CET 2021

On 2021-01-29 09:18, Richard Carlsson wrote:
> Den tors 28 jan. 2021 kl 11:02 skrev Nicolas Martyanoff <khaelin@REDACTED>:
>> If you had started transparently about how you and people at WhatsApp were
>> hoping to drive changes to the Erlang language, starting with a new
>> operator/annotation, you would have received very different responses.
> Would I have? Any suggested change needs to stand on its own merits,
> because there can be no guarantee that anyone will continue to build on it.
> An EEP can explain which further developments are made possible by a
> certain change, but that does not mean that any of them will happen, only
> that there would be a new opening for experimentation.

In general I would agree with you. But as you said yourselves, there are
several propositions you would like to make, they originally come from a plan
with a larger scale, and at least one of them involves scoping.

Lots of people, me included, would be perfectly fine with an optional warning
for existing binding names in pattern matches. Less people are ok with a new
operator/annotation which could end up mandatory. But if this operator was a
first step to a change in scoping rules (as it was clearly said, and tell me
if I'm mistaken, this operator is a necessity for any future possible change
to scoping rules), I suspect a lot more people would have a very different

Context matters. Developers invest a lot of time, money and energy in the
languages they choose for their work; if these languages change, maybe in a
way that makes them reconsider their choice, they would like to know early.

On a different subject, I regret to see so much energy wasted on an annotation
to supposedly make code easier to read, when there is so much potential in,
for example, improving Dialyzer, standard libraries and tooling. Of course,
every one is free to work on what they want, but it is still a bit sad.

Nicolas Martyanoff

More information about the erlang-questions mailing list