code review 5451099: spec: remove contentious composite literal shortening

spec: remove contentious composite literal shortening

[rsc, 2011-12-05 19:02:40]

Hello gri, r (cc: golang-dev@googlegroups.com),

I'd like you to review this change to
https://go.googlecode.com/hg/

[r, 2011-12-05 19:07:09]

LGTM
unfortunately

[gustavo_niemeyer.net, 2011-12-05 19:19:11]

> LGTM
> unfortunately

So, what was it? There was internal agreement that things look bad, or
that it's not worth risking Go 1 with these changes? Was a bit
surprising to see the quiet abandoning.

-- 
Gustavo Niemeyer
http://niemeyer.net
http://niemeyer.net/plus
http://niemeyer.net/twitter
http://niemeyer.net/blog

-- I'm not absolutely sure of anything.

[r2, 2011-12-05 19:22:10]

On Dec 5, 2011, at 11:18 AM, Gustavo Niemeyer wrote:

>> LGTM
>> unfortunately
> 
> So, what was it? There was internal agreement that things look bad, or
> that it's not worth risking Go 1 with these changes? Was a bit
> surprising to see the quiet abandoning.

There was a lot of disagreement about which of the elisions made possible by
this feature were improvements. Given the confusion, it seemed we didn't
understand the consequences of the change well enough to push it out.

-rob

[rsc, 2011-12-05 19:22:26]

*** Submitted as http://code.google.com/p/go/source/detail?r=10d6c39c2685 ***

spec: remove contentious composite literal shortening

R=gri, r, gustavo, r
CC=golang-dev
http://codereview.appspot.com/5451099

[rsc, 2011-12-05 19:28:38]

On Mon, Dec 5, 2011 at 14:18, Gustavo Niemeyer <gustavo@niemeyer.net> wrote:
> So, what was it? There was internal agreement that things look bad, or
> that it's not worth risking Go 1 with these changes? Was a bit
> surprising to see the quiet abandoning.

You make it sound like this discussion was not public:
http://groups.google.com/group/golang-dev/t/696c3b6fefc5085f

Russ

[gustavo_niemeyer.net, 2011-12-05 19:40:52]

> You make it sound like this discussion was not public:

The agreement to abort the plan was not public, so I was merely asking
what was the final understanding of the situation.

> http://groups.google.com/group/golang-dev/t/696c3b6fefc5085f

This thread feels like a healthy brainstorm around the change,
including points going both ways. I'd never expect such a disruptive
change to go in without comments.

-- 
Gustavo Niemeyer
http://niemeyer.net
http://niemeyer.net/plus
http://niemeyer.net/twitter
http://niemeyer.net/blog

-- I'm not absolutely sure of anything.

[rsc, 2011-12-05 19:58:50]

On Mon, Dec 5, 2011 at 14:40, Gustavo Niemeyer <gustavo@niemeyer.net> wrote:
> The agreement to abort the plan was not public, so I was merely asking
> what was the final understanding of the situation.
>
>> http://groups.google.com/group/golang-dev/t/696c3b6fefc5085f
>
> This thread feels like a healthy brainstorm around the change,
> including points going both ways. I'd never expect such a disruptive
> change to go in without comments.

The requirement for making most changes, but
especially language changes, is consensus.
That thread showed that there was, after seeing
a demonstration of its implications, no longer
consensus on the change being a good idea.
Robert, Rob, and Andrew all raised objections,
and I decided not to try to make a case for it,
so I rolled it back.

I'm disappointed too, but taking it out leaves
room for a better idea to come along later.

I note that one of your examples is still allowed:

>    []*Page{{Top: r}, {Top: s}, {Top: t}}

I only rolled back the shortening of struct fields,
not the eliding of &T like in the above.

Russ

[gustavo_niemeyer.net, 2011-12-05 20:11:09]

> The requirement for making most changes, but
> especially language changes, is consensus.
(...)
> I'm disappointed too, but taking it out leaves
> room for a better idea to come along later.

Understood, and that's helpful to get the picture, thank you.

> I note that one of your examples is still allowed:
>
>>    []*Page{{Top: r}, {Top: s}, {Top: t}}
>
> I only rolled back the shortening of struct fields,
> not the eliding of &T like in the above.

Oh, that's great! I had missed it.

-- 
Gustavo Niemeyer
http://niemeyer.net
http://niemeyer.net/plus
http://niemeyer.net/twitter
http://niemeyer.net/blog

-- I'm not absolutely sure of anything.