On Monday, 22 April 2019 at 16:43:56 UTC, Seb wrote:

> Submitting a DIP with the proposed changes would be very welcomed. I'm happy to assist you with it. Just ping me on the DIP PR.

I have created a Github repo and added you as a collaborator. I will work on this as I have time.

On 4/22/19 1:34 PM, bachmeier wrote:
> On Monday, 22 April 2019 at 16:43:56 UTC, Seb wrote:
> 
>> Submitting a DIP with the proposed changes would be very welcomed. I'm happy to assist you with it. Just ping me on the DIP PR.
> 
> I have created a Github repo and added you as a collaborator. I will work on this as I have time.

Awesome you guys. Thank you!

On Friday, 12 April 2019 at 07:35:05 UTC, Tofu Kaitlyn wrote:
> Sorry for that title, but this is something I have been thinking about for a while...
>
> [...]

On Friday, 12 April 2019 at 07:35:05 UTC, Tofu K
This is a user testimony of D


https://medium.com/tripaneer-techblog/why-we-chose-the-d-language-and-vibe-d-3684131a71cd

On Thursday, 25 April 2019 at 00:28:41 UTC, Greatsam4sure wrote:
>
> On Friday, 12 April 2019 at 07:35:05 UTC, Tofu K
> This is a user testimony of D
>
>
> https://medium.com/tripaneer-techblog/why-we-chose-the-d-language-and-vibe-d-3684131a71cd

Sorry, but same could be said about other languages that have better tooling and community support. With the important difference that those other languages will not blow up in your face due to the maintainers fiddling around with new features. As regards maintenance, with D I used to lose a lot of time fixing my code from one compiler version to the next (what a waste of time) or trying to figure out what would be the best way to work around D's messy string handling and how it would affect both the performance and the flow / logic of the program. Again, Tripaneer seems to use D only for one specific purpose, so it's a niche application that says nothing about D's _general_ usefulness. You can write a great program for one particular purpose in Lisp too and go on about how great it is. That says nothing about how you will fare once you leave your bubble.

I started to use D for a very specific purpose, because back in the day D was the only language that offered what I needed (e.g. seamless C interop + DLL support, reasonably fast native code). We were a bit pressed for time, so D was the best option. However, once the whole thing grew bigger and we had to leave the bubble to cater for mobile for example, things no longer looked so bright, and DIY is not an option for everyone (due to lack of time / resources / expertise). It's as if D sold you a multi-purpose toolkit that only works at room temperature.

And of course, there's the paradox that the better you get at D, the more you know about programming, the more you know about programming, the more you notice that D is poorly implemented.

BTW: Maybe change the copyright notice to 2019. Now it says: "Copyright © 1999-2018 by the D Language Foundation" ;)

On Thursday, 25 April 2019 at 11:27:53 UTC, Chris wrote:
> On Thursday, 25 April 2019 at 00:28:41 UTC, Greatsam4sure wrote:
>> [...]
>
> Sorry, but same could be said about other languages that have better tooling and community support. With the important difference that those other languages will not blow up in your face due to the maintainers fiddling around with new features. As regards maintenance, with D I used to lose a lot of time fixing my code from one compiler version to the next (what a waste of time) or trying to figure out what would be the best way to work around D's messy string handling and how it would affect both the performance and the flow / logic of the program. Again, Tripaneer seems to use D only for one specific purpose, so it's a niche application that says nothing about D's _general_ usefulness. You can write a great program for one particular purpose in Lisp too and go on about how great it is. That says nothing about how you will fare once you leave your bubble.
>
> [...]

It would be great if you can provide constructive feedback with detailed points we can work on.
What issues do you have while updating dmd version? What is poorly implemented in your opinion?

If you help us, we can help you.

Kind regards
Andre

On Friday, 12 April 2019 at 13:43:46 UTC, Chris wrote:
> On Friday, 12 April 2019 at 13:19:01 UTC, bachmeier wrote:
>>
>>> PS For D zealots: add your favorite insult here, e.g. "entitled [...]"
>>
>> These statements would be better left on Slashdot...they don't help your case, they make it look like you're unreasonable.
>
> Ah well, damned if you do and damned if you don't. It's only that I've been called names here before and often the replies have been more on the esoteric side of things. But what did Joakim actually post that was so weird? I think D is getting to a stage where it's more of a religion than a tool. The signs are there:
>
> 1. criticism is considered weird / unreasonable (blasphemous?), sure we all know that Earth is a disc
>
> 2. topics are not addressed and the answers are often evasive and / or of a general philosophical or anecdotal nature
>
> 3. I've said it before: personal mistakes, wrong design decisions, are socialized and the whole community is "guilty" and has to "learn". I have a different take on leadership.
>
> 4. "At least we're not C++!"
>
> This is usually a sign of decay. But honestly, it's all been said over the years and nothing has changed...or changed for the worse.

This is generally what happens when an organization grows without proper structural control.

Essentially it becomes a cult. A cult is created when dissenting opinions are shut out. It reinforces the main opinions, attracts cult followers(generally known as bootlickers, which are people who view the world not in intellectual means but in terms of master/slave relationships).

There is a very real positive[meaning reinforced] feedback cycle that takes place because the negative[meaning opposing] feedback cycle has been removed.  One see's this in all cults and fanatics. Politics and religion are the biggest cult creators. Unfortunately, when one becomes part of a cult, it tends to destroy their mind and they then bring their cult mentalities to other areas.

Humans are insane, they cannot judge truth properly without proper experiences. Since they generally seek out only experiences that agree with their previous experiences they simply reinforce previously established mentalities. Since all mentalities are flawed, it is an unstable process that eventually diverges(although in practice, with correct social and parental guidance it can produce good results).

Essentially anyone that disagree's with the cult is bludgeoned in to leaving. No truly intellectual common ground can be had. Most of the cult members are bootlickers and eventually the best bootlickers rise to become the leaders from which the cult becomes a cancer with no real purpose but to grow and survive.

Ultimately it is the bootlickers, who generally are extremely ignorant since they do not know what real intelligence is. They simply reinforce the authorities views since they believe authorities must have true intelligence. By mimicking authorities they believe they will become an authority and become intelligent[again, for them authority and intelligence are identical, which is the flaw in their mind].


You can always tell a cult exists when people congregate and reinforce a singular opinion of leaders and some of them vie for "leadership" through what amounts to be a popularity contest. Anyone against that opinion is shunned, berated, exiled, etc. Now, I know what you are thinking... it sounds like I'm talking about pretty much all human organizations... yep... pretty much. It's the nature of how feedback works, doesn't matter if it's electrical, mechanical, social, mathematical... Hell, it is how we train neural nets. It seems to be how nature works itself. It's not the bootlickers fault, so to speak, but the leadership, who must understand what is going on to prevent the cultism from occurring in a negative way. The bootlickers cannot and will not change, so it is up to the leaders to guide them. Bootlickers are simply the child mind and they are the people who have no grown up. They are looking for guidance and acceptance.

There are a few bootlickers in the forums. What you are describing is D's decent in to a cult. Most other popular programming languages have well established cult's. At least it's programming and not politics, religion, or war.

On Thursday, 25 April 2019 at 11:27:53 UTC, Chris wrote:
> And of course, there's the paradox that the better you get at D, the more you know about programming, the more you know about programming, the more you notice that D is poorly implemented.
It's not a paradox, it's an axiom. I can say exactly the same about C++, Ruby, PHP and JavaScript.

On Thursday, 25 April 2019 at 11:27:53 UTC, Chris wrote:
> As regards maintenance, with D I used to lose a lot of time fixing my code from one compiler version to the next (what a waste of time)

If you think it was a waste of time, you could just have continued without updating. All releases are still there. http://downloads.dlang.org/releases/

Bastiaan.

On 4/25/2019 1:01 PM, Bastiaan Veelo wrote:
> If you think it was a waste of time, you could just have continued without updating. All releases are still there. http://downloads.dlang.org/releases/

True enough. With other products, I don't upgrade them until there's a good reason to. Saves a lot of effort.

On Thu, Apr 25, 2019 at 10:01:20PM -0700, Walter Bright via Digitalmars-d wrote:
> On 4/25/2019 1:01 PM, Bastiaan Veelo wrote:
> > If you think it was a waste of time, you could just have continued without updating. All releases are still there. http://downloads.dlang.org/releases/
> 
> True enough. With other products, I don't upgrade them until there's a good reason to. Saves a lot of effort.

More than once I've been bitten by software upgrades that either outright stop compiling or working with existing code, or break existing code in subtle but unacceptable ways that are unfixable because they are caused by fundamental changes that upstream will unlikely ever want to revert.

These days, I've come to prefer just embedding the exact version of the source code I depend on in my workspace, either checked in as part of the source tree or as a git submodule (that ensures exactly the correct version), and building everything manually.  That, or write my own code and rid myself of too many external dependencies that are likely to go sour without any warning when upstream decides to be "helpful" and break something that used to work fine.

Of course, that's speaking from the receiving end of things.  From an upstream standpoint, sometimes breaking changes are good, and necessary. IMNSHO D should not be so paranoid as to avoid changes at all.  A language that stops improving is a dead one.  As Andrei recently said, the trick is addition rather than replacement.  Keep the old backward compatible stuff around, possibly undocumented, or moved to some archival repo like UndeaD, but still there to keep existing code happy, and add new stuff like std.v2 to keep the good things flowing into the language.  As long as it's *possible* to replicate the older behaviour somehow without onerous effort, I think it's fair game.

And of course, when all else fails, older versions of dmd are still available on the download page.


T

-- 
Elegant or ugly code as well as fine or rude sentences have something in common: they don't depend on the language. -- Luca De Vitis