On Tuesday, 7 March 2017 at 03:04:05 UTC, Joakim wrote:
> https://z0ltan.wordpress.com/2017/02/21/goodbye-rust-and-hello-d/

This reenforces my estimation that the most persuasive feature of any language is the ability to get shit done. That's not a positive value judgement on that focus however. The argument probably should side on the ability to make safe/correct programs _while also_ creating them quickly.

>Very welcoming and helpful community that actually focuses
>on the technical side of things rather than getting sidetracked by social causes

I've seen this mentioned serval times now by people coming from Rust.

Rust users: Is the PC/politicking really that pervasive in their community?

On Tuesday, 7 March 2017 at 18:06:05 UTC, H. S. Teoh wrote:

> Like bachmeier, I have found D arrays (well, slices) to be exactly how I expect arrays to work. The "determinism issue" is really only a problem in exceptional cases where you probably should be using a custom type instead. Or in cases where you're worried about performance and therefore have to understand the nitty-gritty of exactly how slices work in all possible cases -- something that you already have to learn in the first place, if performance is a concern.  I love D arrays, warts and all.

Yep. If you want to give someone enough rope to get maximum performance, you have to give them enough rope to shoot themselves in the foot. Once you've moved into this territory, you've made a decision to throw away safety and convenience in the name of performance, and you better understand what you're doing.

On Tuesday, 7 March 2017 at 03:04:05 UTC, Joakim wrote:
> https://z0ltan.wordpress.com/2017/02/21/goodbye-rust-and-hello-d/

Very welcoming and helpful community that actually focuses on the technical side of things rather than getting sidetracked by social causes

What does he mean by sidetracked by social causes? I've seen this mentioned before. Is this a big thing in the Rust community?

On 03/07/2017 02:07 PM, Jack Stouffer wrote:
>
> I've seen this mentioned serval times now by people coming from Rust.
>
> Rust users: Is the PC/politicking really that pervasive in their community?

That surprised me too. My impression is that us D community folks chastise ourselves plenty for conversations constantly veering off the technical path! Was a pleasant surprise to read a new user say "nah, that's really not so bad in D". Who know? :)

On Tuesday, 7 March 2017 at 18:49:15 UTC, Guillaume Piolat wrote:
> On Tuesday, 7 March 2017 at 18:19:47 UTC, Ola Fosheim Grøstad wrote:
>> On Tuesday, 7 March 2017 at 17:59:28 UTC, Meta wrote:
>>> What exactly are we talking about here? The array stomping protection stuff?
>>
>> Lack of static guarantees on the underlying array buffer.
>
> Like with pointers, ownership doesn't have to be encoded in the type this way.

One can always work around misguided high level features with abstractions, but it doesn't scale well when features that lead to inconsistency is part of the core language. But in comparison to Go one does at least have the ability to abstract.

I still find it odd that people downplay the value of static guarantees in the core language. I am likewise puzzled by the dynamic choices in Golang, but to their credit they decided to focus on evolving the runtime.

I think Rust made a lot of the right choices, except they didn't go far enough to fully reap the benefits.

On 03/07/2017 12:52 PM, Ola Fosheim Grøstad wrote:
> On Tuesday, 7 March 2017 at 12:53:42 UTC, bachmeier wrote:
>> D's arrays are sane, consistent, and logically intuitive even
>> accounting for the "determinism issue". I've never understood why this
>> behavior is surprising - it's exactly what I'd expect. But then I
>> don't program in a way that it's relevant, so maybe I don't understand
>> how it matters.
>
> It is horrible. It is something you only would expect from a hacky
> scripting language. No ifs or buts.

I found it horribly scary at first myself, but what surprised me is that in all the years I've used D, I've never hit an problem resulting from that, not even once.

I admit I can't even begin to explain *why* it's never turned out to be a problem for me though. (And now, *that* is the part that bothers me a little.)

Filed, in my brain at least, under "Strange but True".

On 03/07/2017 11:33 AM, Soulsbane wrote:

> What does he mean by sidetracked by social causes? I've seen this
> mentioned before. Is this a big thing in the Rust community?

The author had opened the following thread:

  http://forum.dlang.org/post/dvmsnoxvdbmraisocpyt@forum.dlang.org

  "community seems infused with both the Feminism/SJW"

Ali

On Tuesday, 7 March 2017 at 20:44:26 UTC, Nick Sabalausky (Abscissa) wrote:
> I found it horribly scary at first myself, but what surprised me is that in all the years I've used D, I've never hit an problem resulting from that, not even once.

Just wait till you use a library that added a sentinel at the end of it temporarily...

On Tuesday, 7 March 2017 at 20:47:12 UTC, Ali Çehreli wrote:
> On 03/07/2017 11:33 AM, Soulsbane wrote:
>
>> What does he mean by sidetracked by social causes? I've seen this
>> mentioned before. Is this a big thing in the Rust community?
>
> The author had opened the following thread:
>
>   http://forum.dlang.org/post/dvmsnoxvdbmraisocpyt@forum.dlang.org
>
>   "community seems infused with both the Feminism/SJW"
>
> Ali

Ah, thanks for the link Ali!

On Tuesday, 7 March 2017 at 19:09:11 UTC, bachmeier wrote:

> Yep. If you want to give someone enough rope to get maximum performance, you have to give them enough rope to shoot themselves in the foot. Once you've moved into this territory, you've made a decision to throw away safety and convenience in the name of performance, and you better understand what you're doing.

I think ATS language is a contradiction here, it combines max performance (at some point it was #1 at the PL shootout, faster than C) with static guarantees Rust folks could only dream about. But the price you pay is amount of intellectual efforts (and time, a lot of time!) to write any non-trivial program.