On Sunday, 14 April 2019 at 19:28:54 UTC, Ola Fosheim Grøstad wrote:
> Compiling to C is a good option, but Walter has always been very much against it.

LLVM support is more important I think, which D already has which is great. The question is if the project is going to spend resources on maintaining the code generator in DMD. My opinion is that we could stop maintaining it and go full LLVM.

>> Rust goes nuts if you talk to any other language.
>
> Really? It interfaces with C doesn't it?

Rust supports C interoperability, not C++. There are some third party libraries for COM support. As soon you cast to a pointer which you often do when calling C interfaces, Rust looses track of the memory and you need to manage it yourself. What you can is to transmute back again after the call. Strange design decision as most are of const void* type or similar doing nothing to the memory.

On 4/14/19 2:09 PM, IGotD- wrote:
> That in mind it is absolute killer feature of D to have a C/C++ FFI that many other languages don't have.

I agree, and sadly it's not a popular enough opinion. A couple of years ago I called on the phone two of our most promising contributors who asked what they should help with, and pitched the core.stdcpp project. They both declined it. No hard feelings - we're grateful enough they chose to do other things within the D milieu -, but it goes to show that planning and "finding a guy" and telling people what to work on is easier said than done, at least for me. I'm glad Manu has taken point on that, wished he got wider help.

On Sunday, 14 April 2019 at 19:13:19 UTC, Nicholas Wilson wrote:
> On Sunday, 14 April 2019 at 15:16:09 UTC, Tourist wrote:
>> I'm going by what I see in the discussion and the links shown, not familiar with the codebase. YMMV.
>>
>> Walter refactors on timeless programming principles, more functional, more immutable, more encapsulation, more information hiding. You don't care for those.
>
> No, those are useful. I think that there are more impactful things that can be done.
>
>> You want to refactor file and directory organization
>
> Because we have ~120 source files in the same directory that do vastly different things and alphabetic sorting doesn't work when the files name need a 'd' prepended because otherwise the module name will collide with a keyword. Its a PITA to navigate.
>
>> fluff based on personal preference and fads.
>
> I'm certainly not the only one who wants that.
>
>> The greatest mistake W&A did was to get contributors incapable of doing original work.
>
> I suggest you do some original work (and some research) before you accuse others of being incapable of original work. FYI, I wrote dcompute, and most of the DMD/Druntime/Phobos PRs merged in the last 6 months went through me.
>
>> They bicker forever about reorganizations that do nothing.
>
> That you think the expected value of the reorganisations is zero shows you have no idea what you're talking about. I'd recommend dropping dropping that attitude until you do.

Argument by majority. Good Work.

I know od your work on dcompute. You should stick to that. Language work and community work is different.

The past years of pull requests:
https://github.com/pulls?utf8=%E2%9C%93&q=is%3Apr+user%3Adlang+closed%3A2015-01-01..2015-04-14+
https://github.com/pulls?utf8=%E2%9C%93&q=is%3Apr+user%3Adlang+closed%3A2016-01-01..2016-04-14+
https://github.com/pulls?utf8=%E2%9C%93&q=is%3Apr+user%3Adlang+closed%3A2017-01-01..2017-04-14+
https://github.com/pulls?utf8=%E2%9C%93&q=is%3Apr+user%3Adlang+closed%3A2018-01-01..2018-04-14+
https://github.com/pulls?utf8=%E2%9C%93&q=is%3Apr+user%3Adlang+closed%3A2019-01-01..2019-04-14+

1149 in 2015
851 in 2016
934 in 2017
1789 in 2018
933 in 2019

Not great. But you say 6 months previous from now:
https://github.com/pulls?utf8=%E2%9C%93&q=is%3Apr+user%3Adlang+closed%3A2014-10-14..2015-04-14+
https://github.com/pulls?utf8=%E2%9C%93&q=is%3Apr+user%3Adlang+closed%3A2015-10-14..2016-04-14+
https://github.com/pulls?utf8=%E2%9C%93&q=is%3Apr+user%3Adlang+closed%3A2016-10-14..2017-04-14+
https://github.com/pulls?utf8=%E2%9C%93&q=is%3Apr+user%3Adlang+closed%3A2017-10-14..2018-04-14+
https://github.com/pulls?utf8=%E2%9C%93&q=is%3Apr+user%3Adlang+closed%3A2018-10-14..2019-04-14+

1710 in 2014-2015
1271 in 2015-2016
1399 in 2016-2017
2699 in 2017-2018
1717 in 2018-2019

You are earning 3000 dollars from our donations to bearly match work that was done for free. Good Work.

On Sunday, 14 April 2019 at 20:48:44 UTC, Andrei Alexandrescu wrote:
> but it goes to show that planning and "finding a guy" and telling people what to work on is easier said than done, at least for me.

People have to move on this is life but projects should persist. How about making a list of projects on github with milestones and task lists and dedicate a forum section or a pinned thread to contributors where you regularly update which projects are progressing and where you need extra workers or funding also you could prioritize projects using polls from the community you can also make a buglist project and prioritize the most wanted bugs. I myself am a programmer I'm not experienced in management but it would make the job of a passing by contributor much easier if things are working in harmony and it would also make the users happy if they see the projects they voted for are in progress.

On Sunday, 14 April 2019 at 12:34:12 UTC, Chris wrote:
> On Sunday, 14 April 2019 at 09:42:01 UTC, Laeeth Isharc wrote:
> [...]
>
> As there is no way the two factions can communicate with each other, there is only one logical consequence. The faction with the broader focus (like myself) wanders off and what remains are the die hard users who are happy in their biotope, so D will remain a niche language and, of course, there's the danger that some people will break off, fork D and add to the general chaos (and you may read questions like this on Stackoverflow: "Which D should I use? There are N implementations. Dragon, Volt, BetterD, CleanD, SaneD...").

i think you are right.

On Sunday, 14 April 2019 at 22:57:48 UTC, Rose wrote:
> On Sunday, 14 April 2019 at 19:13:19 UTC, Nicholas Wilson wrote:
>> On Sunday, 14 April 2019 at 15:16:09 UTC, Tourist wrote:
>>> fluff based on personal preference and fads.
>> I'm certainly not the only one who wants that.
> Argument by majority. Good Work.

Lolwat? How else do you suggest I argue that it is not personal preference?

> I know od[sic] your work on dcompute. You should stick to that. Language work and community work is different.

The dcompute project, in terms of work, is like 90% language and compiler. I'm the one that started the DLF quarterly meetings, I'm the one that proposed the Dconf AGM,

> [stats]
>
> You are earning 3000 dollars from our donations to bearly match work that was done for free. Good Work.

Ahh, the fallacy that past trends will continue.

First of all, the number of PRs being submitted has changed over time, you should de-trend that, as has the number of reviewers, ditto. I was asked to take on that role because there was an ever-growing queue of stagnant PRs and a dearth of reviewers.

If you don't want to donate to the work I have done, thats fine, but don't go making fallacious insults. That kind of behaviour is not welcome here.

On Friday, 12 April 2019 at 07:35:05 UTC, Tofu Kaitlyn wrote:
> [...]
>
> I duno... what do yall think? Is D going to somehow explode in popularity in 5-10 years? Am I missing some part of the picture? Or am I right and if so what can be done about it?

Hello,

I'm using D since 2015. I will try to share my view of it as a simple "back-end web dev". For me, D has not failed at all. It improved a lot since my first hello world, but it still need a lot of work.

I remember using it before 2015, it was just a total mess. Today i'm happy to use it everyday at home and I try to push it at work (but it's not easy when everybody think Go or Rust are the best incoming programming languages...).

What I like about D:

- Easy and fast installation
- Fast compilation
- Beautiful syntax
- Base performance is great (dmd with gc) when you don't work on AAA games
- DUB works nicely for me

What I dont like about D:

[Esthetical part]
- Keyword naming. We really need to do something about small things like that. I don't like the fact that we have a "@" before a keyword and sometimes not.

- File naming. Why are we using filenameslikethat.d ? It's just ugly. Just go with something as simple as file_name.d. Again I understand that for you it's not a real problem and it's not top priority(and you are right), but trust me it just seems ugly when you come from another language.

[Others]
- Every time I try to read the source code of the std on github, I close the browser tab after 5 seconds. I'm trying to find the body of a simple function, and I need to find it and try to understand how it works in a +10000 lines source file.

I think the best improvement would be to separate tests and the real code. Sometimes I check the source code of the Go std, and honestly it's just 200x cleaner. It's so easy to read even for users like me. In D we have 50 lines of doc, 10 lines of code, 50 lines of tests. It's too much and people are doing the same mistake in their projects.

- Static lib. Always a linker error with coff/omf (I don't even know what it means) and a conversion is needed. I always give up and go with derelict.

- When I try to build a simple hello world with vibe, I have this fuc**** linker error that pop out on windows. This one is really unacceptable, I don't know how to fix this myself guys, but i'm sure one of you can. Just do it please. How are we supposed to spread Dlang when new users have to deal with things like that for a simple hello world ?



I still believe in this language and will continue to use it. But we really need to keep improving it (and I will try to help as well even if I don't have the required skills). My goal is to release a game with it on steam platform, and I think it's totally doable today.

Thanks for all the efforts you put in the language.

On Sunday, 14 April 2019 at 20:48:44 UTC, Andrei Alexandrescu wrote:
> On 4/14/19 2:09 PM, IGotD- wrote:
>> That in mind it is absolute killer feature of D to have a C/C++ FFI that many other languages don't have.
>
> I agree, and sadly it's not a popular enough opinion. A couple of years ago I called on the phone two of our most promising contributors who asked what they should help with, and pitched the core.stdcpp project. They both declined it. No hard feelings - we're grateful enough they chose to do other things within the D milieu -, but it goes to show that planning and "finding a guy" and telling people what to work on is easier said than done, at least for me. I'm glad Manu has taken point on that, wished he got wider help.

I have considered using D for a long time, and yet haven't. The reasons are maybe applicable to many. Ultimately it boils down to what you need to do to make a living. If you had tons of spare money you can probably afford to work on something you like or think is good; but if you need to earn then you have to go where the demand is.

The tooling (editor and debugger) is important too for anyone that wants to get the job done or suggest D to an organisation. Here is a proposal:

Freeze D development for 2 years and redirect all energy to tooling.

On 4/15/2019 3:09 AM, Dibyendu Majumdar wrote:
> Ultimately it boils down to what you need to do to make a living. If you had tons of spare money you can probably afford to work on something you like or think is good; but if you need to earn then you have to go where the demand is.
Laeeth Isharc just posted another list of job openings for D programmers at his company:

https://digitalmars.com/d/archives/digitalmars/D/Has_D_failed_unpopular_opinion_but_I_think_yes_325826.html#N325995

On Monday, 15 April 2019 at 10:09:40 UTC, Dibyendu Majumdar wrote:
> On Sunday, 14 April 2019 at 20:48:44 UTC, Andrei Alexandrescu wrote:
>> On 4/14/19 2:09 PM, IGotD- wrote:
>>> That in mind it is absolute killer feature of D to have a C/C++ FFI that many other languages don't have.
>>
>> I agree, and sadly it's not a popular enough opinion. A couple of years ago I called on the phone two of our most promising contributors who asked what they should help with, and pitched the core.stdcpp project. They both declined it. No hard feelings - we're grateful enough they chose to do other things within the D milieu -, but it goes to show that planning and "finding a guy" and telling people what to work on is easier said than done, at least for me. I'm glad Manu has taken point on that, wished he got wider help.
>
> I have considered using D for a long time, and yet haven't. The reasons are maybe applicable to many. Ultimately it boils down to what you need to do to make a living. If you had tons of spare money you can probably afford to work on something you like or think is good; but if you need to earn then you have to go where the demand is.
>
> The tooling (editor and debugger) is important too for anyone that wants to get the job done or suggest D to an organisation. Here is a proposal:
>
> Freeze D development for 2 years and redirect all energy to tooling.

Could you please tell what do you miss from the tooling? For me Intellij D plugin combined with language server works like a charme. Code completion, linter, formatting, everything works fine  Dub is working perfectly for all my use cases. I was never in need to really start the debugger but debugging is working to some extend in windows and is working fine on linux. I am even able to develop applications in windows and using WSL directly compile and debug them on linux.

Kind regards
Andre