lists.llvm.org/pipermail/llvm-dev/2015-October/091218.html

Maybe we could have something similar in D community

On 10/13/2015 6:36 AM, Daniel Kozak wrote:
> lists.llvm.org/pipermail/llvm-dev/2015-October/091218.html
>
> Maybe we could have something similar in D community

No. People who need to be told what decent behavior is won't pay attention to such a document.

On Tuesday, 13 October 2015 at 19:13:07 UTC, Walter Bright wrote:
> On 10/13/2015 6:36 AM, Daniel Kozak wrote:
>> lists.llvm.org/pipermail/llvm-dev/2015-October/091218.html
>>
>> Maybe we could have something similar in D community
>
> No. People who need to be told what decent behavior is won't pay attention to such a document.

:-)

On Tuesday, 13 October 2015 at 19:13:07 UTC, Walter Bright wrote:
> On 10/13/2015 6:36 AM, Daniel Kozak wrote:
>> lists.llvm.org/pipermail/llvm-dev/2015-October/091218.html
>>
>> Maybe we could have something similar in D community
>
> No. People who need to be told what decent behavior is won't pay attention to such a document.

I was considering saying that we shouldn't add one unless the need presents itself, and for the most part, our community has been well-behaved, but your response is spot-on. About the only reason that I can think of why we might really benefit from such a document would be if we did have a problem, and we needed a way to fairly justify kicking people out of the newsgroup or from other official D channels, then we'd have a set of rules to point to that the person had violated rather than it just being the say-so of someone in charge. But fortunately, we haven't generally had problems like that. Occasionally, we've had a bad apple show up and cause problems, but none have stayed around long term, and things are normally pretty civil around here.

- Jonathan M Davis

On Tuesday, 13 October 2015 at 19:13:07 UTC, Walter Bright wrote:
> On 10/13/2015 6:36 AM, Daniel Kozak wrote:
>> lists.llvm.org/pipermail/llvm-dev/2015-October/091218.html
>>
>> Maybe we could have something similar in D community
>
> No. People who need to be told what decent behavior is won't pay attention to such a document.

Yes, I such such codes of conduct not as a solution to, but and indication of a problem.

On Wednesday, 14 October 2015 at 09:50:31 UTC, Marc Schütz wrote:
> On Tuesday, 13 October 2015 at 19:13:07 UTC, Walter Bright wrote:
>> On 10/13/2015 6:36 AM, Daniel Kozak wrote:
>>> lists.llvm.org/pipermail/llvm-dev/2015-October/091218.html
>>>
>>> Maybe we could have something similar in D community
>>
>> No. People who need to be told what decent behavior is won't pay attention to such a document.
>
> Yes, I such such codes of conduct not as a solution to, but and indication of a problem.

It is inevitable that even in a nice and respectful community there are misunderstandings based on the lack of body language and audible tone of voice (e.g. in the case of irony, tongue in cheek remarks etc.). Not to mention the fact that different people and cultures have a different understanding of what's appropriate|insulting|disrespectful and what isn't.

A code of conduct might do more harm than good, when people keep pointing to it saying "Look ma, he said XYZ to me! Ban him!". I once was on a forum like that and it soon became unbearable with people taking offense at every answer they got and complaining to the admin of the forum demanding this or that user be banned. It eats up a lot of the admin's time too, you know, toys, pram, kindergarten.

On Tuesday, 13 October 2015 at 19:13:07 UTC, Walter Bright wrote:
> On 10/13/2015 6:36 AM, Daniel Kozak wrote:
>> lists.llvm.org/pipermail/llvm-dev/2015-October/091218.html
>>
>> Maybe we could have something similar in D community
>
> No. People who need to be told what decent behavior is won't pay attention to such a document.

This is the most compelling reason I've yet seen to use D over Rust or Go.

On Wednesday, 14 October 2015 at 13:21:28 UTC, logicchains wrote:
> On Tuesday, 13 October 2015 at 19:13:07 UTC, Walter Bright wrote:
>> On 10/13/2015 6:36 AM, Daniel Kozak wrote:
>>> lists.llvm.org/pipermail/llvm-dev/2015-October/091218.html
>>>
>>> Maybe we could have something similar in D community
>>
>> No. People who need to be told what decent behavior is won't pay attention to such a document.
>
> This is the most compelling reason I've yet seen to use D over Rust or Go.

Wow, damning with faint praise, why is this so important?  I agree with Walter, but surely you see something else to like more about D? :)

On Wednesday, 14 October 2015 at 16:07:01 UTC, Joakim wrote:
> On Wednesday, 14 October 2015 at 13:21:28 UTC, logicchains wrote:
>> On Tuesday, 13 October 2015 at 19:13:07 UTC, Walter Bright wrote:
>>> On 10/13/2015 6:36 AM, Daniel Kozak wrote:
>>>> lists.llvm.org/pipermail/llvm-dev/2015-October/091218.html
>>>>
>>>> Maybe we could have something similar in D community
>>>
>>> No. People who need to be told what decent behavior is won't pay attention to such a document.
>>
>> This is the most compelling reason I've yet seen to use D over Rust or Go.
>
> Wow, damning with faint praise, why is this so important?  I agree with Walter, but surely you see something else to like more about D? :)

I didn't mean it as damning with faint praise, I'm just personally really unfond of Django-style COCs (and I'm not the only one, given by the response on the Go mailing list to their COC's introduction).

Since you asked, D's compile time metaprogramming facilities are in my view the best thing about D. I think though that D's marketing lacks sufficient focus on that. As someone who's been paid to write Go but not D, I think the Go's biggest advantage in capturing developer mindshare is aesthetic; it's simple to grasp and subjectively "neat" in some sense. D on the other hand seems perpetually unfinished; ref counting isn't finished, gc-free exceptions aren't finished, "shared" isn't finished, safe/ref isn't finished (at least I recall reading somewhere on this list that there are still some bugs in the implementation that allow memory-unsafe behaviour), integrating Vibe.D's co-routines isn't (as far as I'm aware) finished, improving the GC isn't finished, typecons.Unique isn't finished. Even if it's not entirely logical, all these unfinished aspects can add up to produce a less positive aesthetic impression of the language compared to a language that comes across as more polished.

Plus, Go has a much simpler pitch: looks kinda like Python but is faster (due to being compiled if nothing else) and doesn't screw up parallelism. There are/were a lot of Python devs using Python where it probably wasn't appropriate (such as where I work), so it's an easy niche to fill.

On Thursday, 15 October 2015 at 06:36:32 UTC, logicchains wrote:
> Even if it's not entirely logical, all these unfinished aspects can add up to produce a less positive aesthetic impression of the language compared to a language that comes across as more polished.

If go is finished, why there's activity in its repository?