On Wednesday, 27 December 2017 at 16:46:18 UTC, Russel Winder wrote:
> On Wed, 2017-12-27 at 08:59 +0000, Dan Partelly via Digitalmars-d wrote:
>> 
> […]
>> I could not agree more with this. It is unfortunate D has dependencies on a garbage collector in language proper and in std.
>
> Given the current situation, D's best route to traction is to embrace GC and ignore all complaints other than "give us a better GC". (*)
>
> It is all about differentiation. Forget competing against C, C++, and
> Rust. D is the   C++ inspired language with GC that isn't Go.
>
>> […]
>
> (*) "Better C" is a specialist use case for Walter and the D backend.

Although I don't know D very well yet, it sounds like Russel hits the nail precisely on the head here. FWICT, folks have lately used scripting languages (ex. Python, Perl, Ruby) for larger and larger programs (and even JS+Node for local apps), but it seems to me like the pendulum is swinging back the other way as everyone wants optional types and also JIT implementations like [PyPy](http://pypy.org/).

I think the D wiki would benefit from an article on how D competes-with/differs-from Go.

Also, as a related aside, another language that D competes with is Vala. Now that D is part of GCC, I wonder if D could replace Vala.

On Wednesday, 27 December 2017 at 19:11:14 UTC, Laeeth Isharc wrote:
>'Competition is for losers', according to Peter Thiel.  It's completely the wrong mindset >to succeed in a free society.  What you're supposed to do is create a monopoly that you >earn and keep earning every day.  Economic quasi-rent, or pure profit,

I dont really know who Peter Thiel guys is but economic quasi-rent assumes that you already have something of value, which feels a real void. For example Gabe Newell and his Steam store.  In each one of us there is lazy rentier, it is only natural to cash in without doing any real effort(altough today;s rentiers are not like the nobles of the past, they do work 12h / day , but yeah .. ). But this destroys innovation, and in the long run is of little value to society since it doesn't produce more advances.

>D shouldn't compete against anything any more than it has tried to compete in the past.  >The way to success is to listen to people who like what you are doing anyway and would >like you to develop along

D doesnt exist in a void where it can form a natural monopoly.Also, in this world things do not sell themselves, no matter how good they are. Try what you suggest, no marketing, just excellent decisions and D will kill itself.


On Wednesday, 27 December 2017 at 19:11:14 UTC, Laeeth Isharc wrote:
> 
> It's open source!  It doesn't work like that.
>
> If you want people to work on something, write a proof of concept and talk about it.  T......  do the work in a community of highly intelligent, spirited, and independent-minded people?

Don't put the carriage before the horses. I understand enthusiasm but this is too much. You
cannot in good faith ask anyone which is interested in your cause to contribute his time in development or organize fund raising. Im not here to save the world , the baby seals , or D (if it needs saving), or whatever other crusade. Im here because Im curious about D, curious enough to want to know future direction and what the bright people around here think on it. I ask nobody yet to implement anything, neither do I try to shift its development in any way. I gather cursory information and for the rest ... I simply don't care.. yet.

That is all. Dont try to see more.

On Wednesday, 27 December 2017 at 19:13:15 UTC, John Gabriele wrote:
>
> Although I don't know D very well yet, it sounds like Russel hits the nail precisely on the head here. FWICT, folks have lately used scripting languages (ex. Python, Perl, Ruby) for larger and larger programs (and even JS+Node for local apps),

Resulting in terrible software. Take text editors written in JS+Electron. Slow bloated crap.Noticeable slow on a 2 core i5 MacBook Pro with 8Gb RAM. Press key , notice delay, see char on sceen. Really ???? I mean, REALLY ?  I had a 8086 which could edit text with no issue. Had a C64 with 64kb or RAM too which was up to the task of editing text.  Look at how slow Visual Studio became in time as more and more of it was written in managed languages and non-native code.  Android is pitifully slow , and not because of Linux.

Democratization of programming through JS/ PhP / Perl and other scripts was great for society because it empowered a lot of humans. But also created means to write bloated, inefficient, terribly slow software.  Nuclear weapons used where a surgeon's tool is mandated.

On Wednesday, 27 December 2017 at 20:53:46 UTC, Dan Partelly wrote:
> On Wednesday, 27 December 2017 at 19:13:15 UTC, John Gabriele wrote:
>>
>> Although I don't know D very well yet, it sounds like Russel hits the nail precisely on the head here. FWICT, folks have lately used scripting languages (ex. Python, Perl, Ruby) for larger and larger programs (and even JS+Node for local apps),
>
> Resulting in terrible software. Take text editors written in JS+Electron. Slow bloated crap. {snip}

Maybe I wasn't being very clear there. Also, my unedited comment on that was:

> Although I don't know D very well yet, it sounds like Russel hits the nail precisely on the head here. FWICT, folks have lately used scripting languages (ex. Python, Perl, Ruby) for larger and larger programs (and even JS+Node for local apps), but it seems to me like the pendulum is swinging back the other way as everyone wants optional types and also JIT implementations like [PyPy](http://pypy.org/).

with my point being that I think we're seeing many scripting language users now wanting the features that D already has (types, type inference, high performance and less resource usage (natively compiled)). I think it's a good time for D to attract those users.

On Wednesday, 27 December 2017 at 19:42:50 UTC, Dan Partelly wrote:
> Im not here to save the world , the baby seals , or D (if it needs saving), or whatever other crusade. Im here because Im curious about D, curious enough to want to know future direction and what the bright people around here think on it. I ask nobody yet to implement anything, neither do I try to shift its development in any way. I gather cursory information and for the rest ... I simply don't care.. yet.
>
> That is all. Dont try to see more.

Well, you're doing what most people do, when they hear about a new programming language - i.e. start comparing it to others.... there is a psychological basis for that phenomena - it's human.

But honestly, the best way to learn about a programming language, is to start using it.

So I ask you...what have you written in D lately?

On Wednesday, 27 December 2017 at 18:32:43 UTC, Dan Partelly wrote:
> On Wednesday, 27 December 2017 at 16:46:18 UTC, Russel Winder wrote:
>
>>
>> (*) "Better C" is a specialist use case for Walter and the D backend.
>
> Also, if betterC is a specialist use case for Walter only, why does Walter call it "a game changer for D" ?

I am pretty sure Walter put a question mark after the wording, which makes it a question, not a statement ;-)

-betterC (or as I prefer to call it, -slimD, since I also use C, which is the better C) is needed in D, because otherwise it would exclude a whole class of programmers, and problems that could not be solved. Its' not being forced on anyone. It's there to use if is suits your needs, or peaks your interest.

Is it a game changer? That remains an open question. But D certainly needs it, for those that want/need to use it (which is not just Walter).

But if you want to write 'System D', aka Dos (D operating system), then you will certainly need -slimD (aka -betterC) .. with all its enhancements to come.

On Thursday, 28 December 2017 at 01:09:34 UTC, codephantom wrote:
>
> But honestly, the best way to learn about a programming language, is to start using it.

Sure ,  **if** you decide it worth to be learned. And honestly, almost everybody knows that to get better at a task you must perform the task itself.

>
> So I ask you...what have you written in D lately?

Small snippets. I believe is the best way to start with a new language. Then you decide if you like it, and if it serves any purpose for you. Adopting a new language for anything serious is a big commitment.

My problem currently is that I have no freaking idea what niche D serves, since as I said I perceive multiple personalities in it. I like a lot in the language, but I do not like that I perceive it as a indecisive language.

This is one of the reasons I asked Walter in this thread what is the future of the language ?
Where does it to go ? No clear answer yet.

On Thursday, 28 December 2017 at 01:21:42 UTC, codephantom wrote:

>
> I am pretty sure Walter put a question mark after the wording, which makes it a question, not a statement ;-)
>
>

This is marketing. Many times in marketing questions are used to try to pass a certain perspective as a fact to the target population. You guys here are all pretty smart, so prolly you all seen it ;-)

On Thursday, 28 December 2017 at 02:21:09 UTC, Dan Partelly wrote:
>
> Small snippets. I believe is the best way to start with a new language. Then you decide if you like it, and if it serves any purpose for you. Adopting a new language for anything serious is a big commitment.

This is what I do to. Only discovered D a few months ago..have 300+ snippets and growing... It's the best way to learn about various features ...


> My problem currently is that I have no freaking idea what niche D serves, since as I said I perceive multiple personalities in it. I like a lot in the language, but I do not like that I perceive it as a indecisive language.

I see it as an alternative to C++ - i.e a powerful and flexible compiled language that I can actually understand, and enjoy using.

> This is one of the reasons I asked Walter in this thread what is the future of the language ?
> Where does it to go ? No clear answer yet.

That really depends on what people (not just Walter) do with it, and want to do with it. This varies alot, and will continue to vary alot, as we don't all think the same way, or want to solve the same problems.

Clearly though, D is targetted as a better C, an increasingly better C++, as well as better general use, multiparadigm, powerful, flexible, fast, compiled langauge..that is understandable and pleasant to work with.

It that's not enough to get people interested in it, then what is?

On Thursday, 28 December 2017 at 02:28:20 UTC, Dan Partelly wrote:
>
> This is marketing. Many times in marketing questions are used to try to pass a certain perspective as a fact to the target population. You guys here are all pretty smart, so prolly you all seen it ;-)

Yeah, true.. but Walter likes a debate too ;-)

So I agree, it was part marketing... but it was more .. what do you all think?