www.digitalmars.com         C & C++   DMDScript  

digitalmars.D - Why D is not a popular language?

reply Marcone <marcone email.om> writes:
I've been a programmer for 15 years. I program in several 
programming languages, only for Windows desktop with graphical 
interface. I have been programming in D for only 1 year and for 
me D is the best programming language. Why not invest more in 
Dlang? Better integration of D with Qt5 and Qt6 and the web would 
make Dlang rise in popularity.
Jan 08
next sibling parent aberba <karabutaworld gmail.com> writes:
On Friday, 8 January 2021 at 16:26:43 UTC, Marcone wrote:
 I've been a programmer for 15 years. I program in several 
 programming languages, only for Windows desktop with graphical 
 interface. I have been programming in D for only 1 year and for 
 me D is the best programming language. Why not invest more in 
 Dlang? Better integration of D with Qt5 and Qt6 and the web 
 would make Dlang rise in popularity.
https://aberba.com/2020/why-i-still-use-d/
Jan 08
prev sibling next sibling parent reply IGotD- <nise nise.com> writes:
On Friday, 8 January 2021 at 16:26:43 UTC, Marcone wrote:
 I've been a programmer for 15 years. I program in several 
 programming languages, only for Windows desktop with graphical 
 interface. I have been programming in D for only 1 year and for 
 me D is the best programming language. Why not invest more in 
 Dlang? Better integration of D with Qt5 and Qt6 and the web 
 would make Dlang rise in popularity.
I absolutely agree, that D should have better integration with the Qt library. I've used Qt together with Python (PySide) which works pretty well. There is nothing that stops D for being as usable with Qt as well. As D has very powerful metaprogramming, integrating the signals should be possible. C++ with Qt today requires a "pre-compiler" step as it has some additions. Question is how Qt can be integrated with D without this extra "pre-compiler".
Jan 08
parent reply matheus <matheus gmail.com> writes:
On Friday, 8 January 2021 at 16:31:34 UTC, IGotD- wrote:
 ...
Serious question: Is applications for desktop a thing? I usually write (For Hobby/Work) batch programs which is executed directly from terminal or WEB applications. In fact where I work most of the desktop applications were migrated to WEB. Matheus.
Jan 08
parent Guillaume Piolat <first.last gmail.com> writes:
On Friday, 8 January 2021 at 18:14:49 UTC, matheus wrote:
 Serious question: Is applications for desktop a thing?
It is a thing, for Anything that needs a GUI, and for some reason does not fit the web because it has some performance or data throughput requirement. games video apps graphics apps audio apps ... though of course Web-based software will rightfully eat into those applications that were native without reason.
Jan 08
prev sibling next sibling parent reply sighoya <sighoya gmail.com> writes:
On Friday, 8 January 2021 at 16:26:43 UTC, Marcone wrote:
Why D is not a popular language?
- historical reasons D1 vs D2, paradigm shift from oop to parametricity and value types - immature tooling - no killer app - no large company promoting D including offering jobs for - the best parts about D: meta programming is at the same time the worst part, it works pretty good in simple cases, soon as you building something complex, it didn't work as you intuitively assumed. Further, corner cases regarding feature compatibility and compiler bugs/limitations hinders productivity Double-edged sword is the broad spectrum of paradigms, D's pragmatism. On the one hand side it widens the fields for applicability of D, on the other side it leads to a community with mixed or better contrary goals impeding commitment to convention. However, all these things aren't either dramatic.
 Better integration of D with Qt5 and Qt6 and the web would make 
 Dlang rise in popularity.
Yes, but also better Gtk support would be good? Or maybe something new completely in D? The biggest part for me is more software ergonomic feeling when developing with D. We are not in the eighties anymore. I don't know surely if we can realize better tooling by utilizing and improving the DMD compiler, maybe we need, just as it is the case for Java, an D IDE compiler.
Jan 08
next sibling parent reply aberba <karabutaworld gmail.com> writes:
On Friday, 8 January 2021 at 18:13:52 UTC, sighoya wrote:
 On Friday, 8 January 2021 at 16:26:43 UTC, Marcone wrote:
Why D is not a popular language?
-
 Better integration of D with Qt5 and Qt6 and the web would 
 make Dlang rise in popularity.
Yes, but also better Gtk support would be good?
I'm not sure how gtkd can be made any better than it already is. I've got no complaints so far. Qt isn't even cool. Use GTK-d. Or maybe
 something new completely in D?
We have BeamU and DlangUI. They just don't have enough manpower to polish them up faster.
 The biggest part for me is more software ergonomic feeling when 
 developing with D. We are not in the eighties anymore. I don't 
 know surely if we can realize better tooling by utilizing and 
 improving the DMD compiler, maybe we need, just as it is the 
 case for Java, an D IDE compiler.
If you're on Windows, Visual Studio with Visual-d should be enough. There's also the popular VS code.
Jan 08
next sibling parent reply sighoya <sighoya gmail.com> writes:
On Friday, 8 January 2021 at 20:23:18 UTC, aberba wrote:
 I'm not sure how gtkd can be made any better than it already 
 is. I've got no complaints so far.
GTK 4 support?
 Qt isn't even cool. Use GTK-d.
Okay, I will remember.
 We have BeamU and DlangUI. They just don't have enough manpower 
 to polish them up faster.
Interesting, thanks. Cross-platform and pure D-Code in beamUI seems promising... But I've never really heard about people talking about here, why?
 If you're on Windows, Visual Studio with Visual-d should be 
 enough. There's also the popular VS code.
Yes, syntax checking and code completion is already quite good (even in eclipse), however we have, afaict, no semantic checking yet. We would either need to call DMD/LDC/GDC as compiler service (don't know to which extent this is possible, or we need a new compiler for the IDE). But heck, writing a new compiler incurs a lot of work and time to get in sync with semantic D.
Jan 08
parent reply aberba <karabutaworld gmail.com> writes:
On Friday, 8 January 2021 at 22:51:22 UTC, sighoya wrote:

 GTK 4 support?
Gtk4 just came out. Give it time. No Linux distribution or platform that I know of has support for gtk4 yet. So it's not going to be useful anytime soon. Give it time.
 We have BeamU and DlangUI. They just don't have enough 
 manpower to polish them up faster.
Interesting, thanks. Cross-platform and pure D-Code in beamUI seems promising... But I've never really heard about people talking about here, why?
I have no idea. Maybe not enough people in the community do stuff related to that. May be. The community is not exactly managed BTW. Starting 2021, I'm seeing even never (smaller) languages communities doing survey to factor those (or even few ideas) into language direction. We're doing none of those. We've got a great language though.
 If you're on Windows, Visual Studio with Visual-d should be 
 enough. There's also the popular VS code.
Yes, syntax checking and code completion is already quite good (even in eclipse), however we have, afaict, no semantic checking yet. We would either need to call DMD/LDC/GDC as compiler service (don't know to which extent this is possible, or we need a new compiler for the IDE). But heck, writing a new compiler incurs a lot of work and time to get in sync with semantic D.
D has many new features. Including UFCS and our kind of metaprogramming that makes such smart IDE features not an easy thing to do for D. It'll either take a while for DMD as a library (for tooling) to catch on OR a dedicated team of people to work on this full-tme. As to when that'll happen, I'm not sure. Those in charge don't even use such fancy tools themselves. So it's it quite unlikely to come from them.
Jan 09
next sibling parent SealabJaster <sealabjaster gmail.com> writes:
On Saturday, 9 January 2021 at 14:47:35 UTC, aberba wrote:
 Those in charge don't even use such fancy tools themselves. So 
 it's it quite unlikely to come from them.
I wonder to what extent the higher-ups and language designers have used/kept up with the newer features of languages such as There always seems to be that wall of one side being like "This is a large readability/productivity/whatever boost" and another being "I don't get how this is useful, just use a (buggy and or clunky and non-standard) library implementation and be done with it". I feel other languages are evolving both in language features *and* ecosystem *and* tooling and so on, all at the same time, whilst D mostly focuses on the language and there being only small, dedicated yet separate communities in D's ecosystem, leading to a feeling of (and maybe actual) stagnation in certain areas. Is there a well-known, public, high-level vision of D anymore? I've expressed this view many times now, but I don't seem to have noticed any real management for D as a whole - just isolated areas of it. I couldn't really tell you what D's future looks like, since the information either doesn't exist; is hidden behind closed doors, or just isn't in an obvious, easy-to-find location. For example I personally wouldn't call "X made a post in a thread with 50+ replies stating something interesting/important, alson the thread's title is also only barely related", as being an obvious location.
Jan 09
prev sibling parent oddp <oddp posteo.de> writes:
On 09.01.21 15:47, aberba via Digitalmars-d wrote:
 GTK 4 support?
Gtk4 just came out. Give it time.
We're getting there: https://github.com/gtkd-developers/GtkD/issues/315#issuecomment-757397817
Jan 10
prev sibling parent reply Jack <jckj33 gmail.com> writes:
On Friday, 8 January 2021 at 20:23:18 UTC, aberba wrote:
 On Friday, 8 January 2021 at 18:13:52 UTC, sighoya wrote:
 On Friday, 8 January 2021 at 16:26:43 UTC, Marcone wrote:
Why D is not a popular language?
-
 Better integration of D with Qt5 and Qt6 and the web would 
 make Dlang rise in popularity.
Yes, but also better Gtk support would be good?
I'm not sure how gtkd can be made any better than it already is. I've got no complaints so far. Qt isn't even cool. Use GTK-d. Or maybe
 something new completely in D?
We have BeamU and DlangUI. They just don't have enough manpower to polish them up faster.
Does DLangUI has a GUI designer? I've heard of this one before
Jan 08
next sibling parent evilrat <evilrat666 gmail.com> writes:
On Saturday, 9 January 2021 at 02:07:49 UTC, Jack wrote:
 On Friday, 8 January 2021 at 20:23:18 UTC, aberba wrote:
 On Friday, 8 January 2021 at 18:13:52 UTC, sighoya wrote:
 On Friday, 8 January 2021 at 16:26:43 UTC, Marcone wrote:
Why D is not a popular language?
-
 Better integration of D with Qt5 and Qt6 and the web would 
 make Dlang rise in popularity.
Yes, but also better Gtk support would be good?
I'm not sure how gtkd can be made any better than it already is. I've got no complaints so far. Qt isn't even cool. Use GTK-d. Or maybe
 something new completely in D?
We have BeamU and DlangUI. They just don't have enough manpower to polish them up faster.
Does DLangUI has a GUI designer? I've heard of this one before
Short answer: NO But look in examples, there is quick markup viewer in demo's which you can use to design your UI with markup on the fly (I think it was this one) https://github.com/buggins/dlangui/tree/master/examples/dmledit P.S. unfortunately DlangUI is currently abandoned, the project author doesn't have time and AFAIK nobody else is volunteered to pick up the maintenance.
Jan 08
prev sibling parent Imperatorn <johan_forsberg_86 hotmail.com> writes:
On Saturday, 9 January 2021 at 02:07:49 UTC, Jack wrote:
 On Friday, 8 January 2021 at 20:23:18 UTC, aberba wrote:
 On Friday, 8 January 2021 at 18:13:52 UTC, sighoya wrote:
 On Friday, 8 January 2021 at 16:26:43 UTC, Marcone wrote:
Why D is not a popular language?
-
 Better integration of D with Qt5 and Qt6 and the web would 
 make Dlang rise in popularity.
Yes, but also better Gtk support would be good?
I'm not sure how gtkd can be made any better than it already is. I've got no complaints so far. Qt isn't even cool. Use GTK-d. Or maybe
 something new completely in D?
We have BeamU and DlangUI. They just don't have enough manpower to polish them up faster.
Does DLangUI has a GUI designer? I've heard of this one before
You can use Glade though
Jan 09
prev sibling parent Jacob Carlborg <doob me.com> writes:
On 2021-01-08 19:13, sighoya wrote:

 Yes, but also better Gtk support would be good? Or maybe something new 
 completely in D?
Not something completely new, but there's DWT [1], a cross-platform GUI library completely implemented in D. It uses Win32 on Windows and GTK+ on Linux. [1] https://github.com/d-widget-toolkit/dwt -- /Jacob Carlborg
Jan 09
prev sibling next sibling parent reply Guillaume Piolat <first.last gmail.com> writes:
On Friday, 8 January 2021 at 16:26:43 UTC, Marcone wrote:
 I've been a programmer for 15 years. I program in several 
 programming languages, only for Windows desktop with graphical 
 interface. I have been programming in D for only 1 year and for 
 me D is the best programming language. Why not invest more in 
 Dlang? Better integration of D with Qt5 and Qt6 and the web 
 would make Dlang rise in popularity.
https://lmeyerov.github.io/projects/socioplt/papers/oopsla2013.pdf "We found that existing code, existing expertise, and open source libraries are the dominant drivers of adoption." D community typically focus on the language (and what goes into Phobos) rather than the indispensable 3rd-party libraries.
Jan 09
parent aberba <karabutaworld gmail.com> writes:
On Saturday, 9 January 2021 at 14:02:28 UTC, Guillaume Piolat 
wrote:
 On Friday, 8 January 2021 at 16:26:43 UTC, Marcone wrote:
 I've been a programmer for 15 years. I program in several 
 programming languages, only for Windows desktop with graphical 
 interface. I have been programming in D for only 1 year and 
 for me D is the best programming language. Why not invest more 
 in Dlang? Better integration of D with Qt5 and Qt6 and the web 
 would make Dlang rise in popularity.
https://lmeyerov.github.io/projects/socioplt/papers/oopsla2013.pdf "We found that existing code, existing expertise, and open source libraries are the dominant drivers of adoption."
My gut has always told me this. Glad it's backed by research.
 D community typically focus on the language (and what goes into 
 Phobos) rather than the indispensable 3rd-party libraries.
Jan 09
prev sibling next sibling parent reply zjh <fqbqrr 163.com> writes:
On Friday, 8 January 2021 at 16:26:43 UTC, Marcone wrote:


D's problems include:

1. The community is not well organized to form a resultant force. 
For example, list the most important things to do. then do it.

2. There is no benefit to GC in D. Better get rid of it.for 
example in the stdlib. or work on betterC. Structure should at 
least have inheritance. And can make use of associative arrays. 
With these two things, combined with metaprogramming, it's 
possible to beat today's new languages.

3, There are a lot of things in the forum that are not being used 
well. Collect and organize into posts, to gather good post 
together. The beginner is most important ,because they are the 
largest group.
4, the D expert's posts should make a seperate group.so that 
everyone can find it.
Jan 09
next sibling parent IGotD- <nise nise.com> writes:
On Saturday, 9 January 2021 at 14:55:33 UTC, zjh wrote:
 4, the D expert's posts should make a seperate group.so that 
 everyone can find it.
Searching topics for D isn't that bad. However, you should use Google to do it as the D forum is really a newsgroup which Google indexes. The search field in the dlang.org site is almost unusable. I recommend to use Google for searching dlang questions, then you usually find what you want pretty quickly.
Jan 09
prev sibling parent reply sighoya <sighoya gmail.com> writes:
On Saturday, 9 January 2021 at 14:55:33 UTC, zjh wrote:
 2. There is no benefit to GC in D. Better get rid of it.
I think this is part of the problem in our community, the disparity between proponents and opponents of GC utilization.
Jan 09
parent reply aberba <karabutaworld gmail.com> writes:
On Saturday, 9 January 2021 at 19:53:44 UTC, sighoya wrote:
 On Saturday, 9 January 2021 at 14:55:33 UTC, zjh wrote:
 2. There is no benefit to GC in D. Better get rid of it.
I think this is part of the problem in our community, the disparity between proponents and opponents of GC utilization.
GC isn't going away. That decision is also written on stone and sealed. D is always going to have a GC. The lack of communication/documentation on some of these things is definitely not helpful. D was designed with the better parts of Java and C++ (from the Origin of D)... except GC in D works very different from what comes mind when most people hear the term Garbage Collection. GC IN D IS NEVER GOING AWAY. LIVE WITH IT or write nogc code.
Jan 09
parent Ola Fosheim Grostad <ola.fosheim.grostad gmail.com> writes:
On Sunday, 10 January 2021 at 00:10:26 UTC, aberba wrote:
 of Java and C++ (from the Origin of D)... except GC in D works 
 very different from what comes mind when most people hear the 
 term Garbage Collection.
It works like any other GC from the 1960-1970s.
Jan 09
prev sibling next sibling parent reply Andre Pany <andre s-e-a-p.de> writes:
On Friday, 8 January 2021 at 16:26:43 UTC, Marcone wrote:
 I've been a programmer for 15 years. I program in several 
 programming languages, only for Windows desktop with graphical 
 interface. I have been programming in D for only 1 year and for 
 me D is the best programming language. Why not invest more in 
 Dlang? Better integration of D with Qt5 and Qt6 and the web 
 would make Dlang rise in popularity.
From a technical point of view, D is in a very good shape and even more fantastic additions are on the horizon for 2021: - named arguments - string interpolation - floating point to string conversion implemented in DRuntime - huge list of more features and fixes At this point of of time the community has the task to increase the popularity of D. Other languages have companies behind to spend money on advertising the language. For D, this advertisements needs to be done by the community. We need more technical evangelists in the community. This will increase the D community and therefore more developers will join which e.g. develops the better QT6 integration or any other missing part. Kind regards Andre
Jan 10
next sibling parent Imperatorn <johan_forsberg_86 hotmail.com> writes:
On Sunday, 10 January 2021 at 11:00:29 UTC, Andre Pany wrote:
 On Friday, 8 January 2021 at 16:26:43 UTC, Marcone wrote:
 [...]
From a technical point of view, D is in a very good shape and even more fantastic additions are on the horizon for 2021: [...]
Well said
Jan 10
prev sibling next sibling parent reply aberba <karabutaworld gmail.com> writes:
On Sunday, 10 January 2021 at 11:00:29 UTC, Andre Pany wrote:
 On Friday, 8 January 2021 at 16:26:43 From a technical point of 
 view, D is in a very good shape and even more fantastic 
 additions are on the horizon for 2021:

 - named arguments
 - string interpolation
 - floating point to string conversion implemented in DRuntime
 - huge list of more features and fixes

 At this point of of time the community has the task to increase 
 the popularity of D. Other languages have companies behind to 
 spend money on advertising the language. For D, this 
 advertisements needs to be done by the community.
 We need more technical evangelists in the community.
+1 How do you think that can happen? (Assuming the DLF isn't involved) How do we grow and aggregate content on D?
 This will increase the D community and therefore more 
 developers will join which e.g. develops the better QT6 
 integration or any other missing part.

 Kind regards
 Andre
Jan 10
next sibling parent reply Andre Pany <andre s-e-a-p.de> writes:
On Sunday, 10 January 2021 at 18:25:35 UTC, aberba wrote:
 On Sunday, 10 January 2021 at 11:00:29 UTC, Andre Pany wrote:
 On Friday, 8 January 2021 at 16:26:43 From a technical point 
 of view, D is in a very good shape and even more fantastic 
 additions are on the horizon for 2021:

 - named arguments
 - string interpolation
 - floating point to string conversion implemented in DRuntime
 - huge list of more features and fixes

 At this point of of time the community has the task to 
 increase the popularity of D. Other languages have companies 
 behind to spend money on advertising the language. For D, this 
 advertisements needs to be done by the community.
 We need more technical evangelists in the community.
+1 How do you think that can happen? (Assuming the DLF isn't involved) How do we grow and aggregate content on D?
 This will increase the D community and therefore more 
 developers will join which e.g. develops the better QT6 
 integration or any other missing part.

 Kind regards
 Andre
Some ideas: - (if you work for a software company) try to establish D. - if the company you work for already is using D, you could make this public (in alignment with the public relations department of your company). - create blogs/tutorials and post them on hn and r/programming. There is far too less D content on this channels. If you have done something, post it and speak about it. - try to speak about D as much as possible, everywhere Kind regards Andre
Jan 10
parent reply Aaron <aaron.ehsion gmail.com> writes:
On Sunday, 10 January 2021 at 18:41:40 UTC, Andre Pany wrote:
 On Sunday, 10 January 2021 at 18:25:35 UTC, aberba wrote:
 On Sunday, 10 January 2021 at 11:00:29 UTC, Andre Pany wrote:
 On Friday, 8 January 2021 at 16:26:43 From a technical point 
 of view, D is in a very good shape and even more fantastic 
 additions are on the horizon for 2021:

 - named arguments
 - string interpolation
 - floating point to string conversion implemented in DRuntime
 - huge list of more features and fixes

 At this point of of time the community has the task to 
 increase the popularity of D. Other languages have companies 
 behind to spend money on advertising the language. For D, 
 this advertisements needs to be done by the community.
 We need more technical evangelists in the community.
+1 How do you think that can happen? (Assuming the DLF isn't involved) How do we grow and aggregate content on D?
 This will increase the D community and therefore more 
 developers will join which e.g. develops the better QT6 
 integration or any other missing part.

 Kind regards
 Andre
Some ideas: - (if you work for a software company) try to establish D. - if the company you work for already is using D, you could make this public (in alignment with the public relations department of your company). - create blogs/tutorials and post them on hn and r/programming. There is far too less D content on this channels. If you have done something, post it and speak about it. - try to speak about D as much as possible, everywhere Kind regards Andre
This is part of the problem, the community is expected to do the heavy lifting in regards to building a community. It's not the only time I've heard this either. They would rather have someone "step up" and do the work for them than doing the work themselves. So why is there no Qt bindings? Ask yourself why you don't write the bindings yourself and you have your answer. No one is going to spend their free time doing what is essentially gruntwork. Especially for such a large fast moving project. An easy workaround is to just use something that already has a good be better for it as well. Anyways I would not recommend D to anyone and I won't be using it for any large projects going forward either. It is a nightmare to maintain. I had a C++ project I converted to D and have been actively developing for the last 5+ years. Ita a medoum sized project, about 100k lines. You are going to be dealing with problems you've never had to deal with any other programming language out there. It isn't for technical reasons either, it is just very bad project management. I dont know how many times I've encountered a problem, spent hours going down a rabbit hole just to find the root of the problem has been reported 10+ years ago and even though the community has come to the conclusion that half a solution is better than no solution, nothing ends up being done instead (by choice). I now maintain my own fork of LDC as I know others do. So I can't really say my project uses D anymore. I never thought I'd have to maintain a compiler when I first converted it to D, I have dreams that I convert the project back to C++. Anyways there's good reason why D isn't popular, you can argue what those reasons are. But from my first hand experience, that's why I won't be using D, and I'd never recommend tormenting my coworkers with the idea of it either.
Jan 10
parent reply John <john email.addr> writes:
On Sunday, 10 January 2021 at 19:33:01 UTC, Aaron wrote:
 [...]
Yup. Leadership, if you can call it that, wants to chase fads and implement new stuff instead of solidifying the base. Saddening, to say the least.
Jan 10
parent reply Andre Pany <andre s-e-a-p.de> writes:
On Sunday, 10 January 2021 at 20:13:46 UTC, John wrote:
 On Sunday, 10 January 2021 at 19:33:01 UTC, Aaron wrote:
 [...]
Yup. Leadership, if you can call it that, wants to chase fads and implement new stuff instead of solidifying the base. Saddening, to say the least.
This is not true, if you keep an eye on the druntime/dmd PR, you will see that a lot of them are about refactorings and code clean up. Also another example: https://forum.dlang.org/thread/rs0rt6$1qkg$1 digitalmars.com Kind regards Andre
Jan 10
parent reply Ola Fosheim Grostad <ola.fosheim.grostad gmail.com> writes:
On Sunday, 10 January 2021 at 20:33:45 UTC, Andre Pany wrote:
 On Sunday, 10 January 2021 at 20:13:46 UTC, John wrote:
 On Sunday, 10 January 2021 at 19:33:01 UTC, Aaron wrote:
 [...]
Yup. Leadership, if you can call it that, wants to chase fads and implement new stuff instead of solidifying the base. Saddening, to say the least.
This is not true, if you keep an eye on the druntime/dmd PR, you will see that a lot of them are about refactorings and code clean up.
Uhm, you just listed features D don't need: string interpolation, named arguments and string conversion in a runtime (stupid idea). The runtime should be slimmed down to a bare minimum. The type system should be fixed. Memory management and shared should be finalized. If that requires refactorning, great. Aaron certainly has a point.
Jan 10
parent reply Max Haughton <maxhaton gmail.com> writes:
On Sunday, 10 January 2021 at 22:23:59 UTC, Ola Fosheim Grostad 
wrote:
 On Sunday, 10 January 2021 at 20:33:45 UTC, Andre Pany wrote:
 On Sunday, 10 January 2021 at 20:13:46 UTC, John wrote:
 [...]
This is not true, if you keep an eye on the druntime/dmd PR, you will see that a lot of them are about refactorings and code clean up.
Uhm, you just listed features D don't need: string interpolation, named arguments and string conversion in a runtime (stupid idea). The runtime should be slimmed down to a bare minimum. The type system should be fixed. Memory management and shared should be finalized. If that requires refactorning, great. Aaron certainly has a point.
string interpolation and named arguments are pretty useful, not familiar with the final one.
Jan 10
next sibling parent Andre Pany <andre s-e-a-p.de> writes:
On Sunday, 10 January 2021 at 22:33:32 UTC, Max Haughton wrote:
 On Sunday, 10 January 2021 at 22:23:59 UTC, Ola Fosheim Grostad 
 wrote:
 On Sunday, 10 January 2021 at 20:33:45 UTC, Andre Pany wrote:
 On Sunday, 10 January 2021 at 20:13:46 UTC, John wrote:
 [...]
This is not true, if you keep an eye on the druntime/dmd PR, you will see that a lot of them are about refactorings and code clean up.
Uhm, you just listed features D don't need: string interpolation, named arguments and string conversion in a runtime (stupid idea). The runtime should be slimmed down to a bare minimum. The type system should be fixed. Memory management and shared should be finalized. If that requires refactorning, great. Aaron certainly has a point.
string interpolation and named arguments are pretty useful, not familiar with the final one.
Floating point to string conversion is currently solved by calling a C function. This leads to various consequences: ctfe, purity and it is even mentioned in the "string interpolation dip" as limitation. Kind regards Andre
Jan 10
prev sibling parent Ola Fosheim Grostad <ola.fosheim.grostad gmail.com> writes:
On Sunday, 10 January 2021 at 22:33:32 UTC, Max Haughton wrote:
 On Sunday, 10 January 2021 at 22:23:59 UTC, Ola Fosheim Grostad 
 wrote:
 string interpolation and named arguments are pretty useful, not 
 familiar with the final one.
Everything is useful if you need it, but you cannot both have success restructuring and adding features at the same time. Fixing core issues are critically important, other stuff can be left to experimental compiler branches. Dont merge in more stuff until the core is solid. I think this is a pretty basic development strategy.
Jan 10
prev sibling parent reply Q. Schroll <qs.il.paperinik gmail.com> writes:
On Sunday, 10 January 2021 at 18:25:35 UTC, aberba wrote:
 How do we grow and aggregate content on D?
Moving from https://forum.dlang.org/group/learn to stackoverflow (SO) would probably give D some more visibility. Just for the sake of code blocks and syntax highlighting, the move to stackoverflow would be a win. Learn could still be used for questions on how things are done in development. Like "How do I test my PR on my machine?" and alike. Things that are interesting for any D programmer should be more visible.
Jan 14
next sibling parent Imperatorn <johan_forsberg_86 hotmail.com> writes:
On Thursday, 14 January 2021 at 16:47:51 UTC, Q. Schroll wrote:
 On Sunday, 10 January 2021 at 18:25:35 UTC, aberba wrote:
 How do we grow and aggregate content on D?
Moving from https://forum.dlang.org/group/learn to stackoverflow (SO) would probably give D some more visibility. Just for the sake of code blocks and syntax highlighting, the move to stackoverflow would be a win. Learn could still be used for questions on how things are done in development. Like "How do I test my PR on my machine?" and alike. Things that are interesting for any D programmer should be more visible.
Fair point. Maybe there could be some synch between also 🤔
Jan 14
prev sibling next sibling parent reply Marcone <marcone email.com> writes:
On Thursday, 14 January 2021 at 16:47:51 UTC, Q. Schroll wrote:
 On Sunday, 10 January 2021 at 18:25:35 UTC, aberba wrote:
 How do we grow and aggregate content on D?
Moving from https://forum.dlang.org/group/learn to stackoverflow (SO) would probably give D some more visibility. Just for the sake of code blocks and syntax highlighting, the move to stackoverflow would be a win. Learn could still be used for questions on how things are done in development. Like "How do I test my PR on my machine?" and alike. Things that are interesting for any D programmer should be more visible.
"yahoo answers" is better than "stackoverflow". Dlang forum is very very better than "yahoo answers overflow".
Jan 14
next sibling parent Max Haughton <maxhaton gmail.com> writes:
On Thursday, 14 January 2021 at 19:06:06 UTC, Marcone wrote:

 "yahoo answers" is better than "stackoverflow".
No
Jan 14
prev sibling parent Imperatorn <johan_forsberg_86 hotmail.com> writes:
On Thursday, 14 January 2021 at 19:06:06 UTC, Marcone wrote:

 "yahoo answers" is better than "stackoverflow"
Not sure if joking or trying to create meme
Jan 14
prev sibling parent reply Nick Treleaven <nick geany.org> writes:
On Thursday, 14 January 2021 at 16:47:51 UTC, Q. Schroll wrote:
 Moving from https://forum.dlang.org/group/learn to 
 stackoverflow (SO) would probably give D some more visibility.
 Just for the sake of code blocks and syntax highlighting, the 
 move to stackoverflow would be a win.
Good idea, it would also be better for people googling dlang questions because they can quickly see the best answers that have been voted up. The question would often also be clearer and with a better title due to editing, so more likely to be picked up by Google. (Often I've searched the forum and failed to find the post I remember seeing despite using clear search terms).
Jan 15
parent reply Mike Parker <aldacron gmail.com> writes:
On Friday, 15 January 2021 at 13:21:34 UTC, Nick Treleaven wrote:
 On Thursday, 14 January 2021 at 16:47:51 UTC, Q. Schroll wrote:
 Moving from https://forum.dlang.org/group/learn to 
 stackoverflow (SO) would probably give D some more visibility.
 Just for the sake of code blocks and syntax highlighting, the 
 move to stackoverflow would be a win.
Good idea, it would also be better for people googling dlang questions because they can quickly see the best answers that have been voted up. The question would often also be clearer and with a better title due to editing, so more likely to be picked up by Google. (Often I've searched the forum and failed to find the post I remember seeing despite using clear search terms).
StackOverflow is ill-suited to serve as a community forum. But there's no need to "move". People already ask D questions there. IMO, they'd be much better off asking them here.
Jan 15
next sibling parent Ola Fosheim =?UTF-8?B?R3LDuHN0YWQ=?= <ola.fosheim.grostad gmail.com> writes:
On Friday, 15 January 2021 at 14:40:24 UTC, Mike Parker wrote:
 StackOverflow is ill-suited to serve as a community forum. But 
 there's no need to "move". People already ask D questions 
 there. IMO, they'd be much better off asking them here.
The answers in the learn forum are usually of very high quality indeed, but it would be even better with a topical index so that finding old answers would be easier. Doesn't have to be on SO, it could be embedded in dlang.org. I suspect many newbies are unaware of the existence of a learn forum...
Jan 15
prev sibling parent Q. Schroll <qs.il.paperinik gmail.com> writes:
On Friday, 15 January 2021 at 14:40:24 UTC, Mike Parker wrote:
 On Friday, 15 January 2021 at 13:21:34 UTC, Nick Treleaven 
 wrote:
 On Thursday, 14 January 2021 at 16:47:51 UTC, Q. Schroll wrote:
 Moving from https://forum.dlang.org/group/learn to 
 stackoverflow (SO) would probably give D some more visibility.
 Just for the sake of code blocks and syntax highlighting, the 
 move to stackoverflow would be a win.
Good idea, it would also be better for people googling dlang questions because they can quickly see the best answers that have been voted up. The question would often also be clearer and with a better title due to editing, so more likely to be picked up by Google. (Often I've searched the forum and failed to find the post I remember seeing despite using clear search terms).
StackOverflow is ill-suited to serve as a community forum. But there's no need to "move". People already ask D questions there. IMO, they'd be much better off asking them here.
As I said, Learn isn't really a community forum; it's where people, not necessarily beginners, ask questions about how to solve a problem with D. I never suggested moving General or Announce to SO. That wouldn't fit, that's for sure.
Jan 18
prev sibling parent reply Marcone <marcone email.com> writes:
The impression I have is that everything in D is difficult, old, 
obsolete and abandoned. I hope someday those responsible for D 
will invest in IDE, integration with Qt, and with Qt Designer to 
convert .ui to .d and a way to use packages that is as easy as in 
Python, because the dub is very bad.
Jan 12
next sibling parent reply evilrat <evilrat666 gmail.com> writes:
On Wednesday, 13 January 2021 at 00:51:35 UTC, Marcone wrote:
 The impression I have is that everything in D is difficult, 
 old, obsolete and abandoned. I hope someday those responsible 
 for D will invest in IDE, integration with Qt, and with Qt 
 Designer to convert .ui to .d and a way to use packages that is 
 as easy as in Python, because the dub is very bad.
Welcome to the wonderful world of Open Source! Ok no time for celebrations, since unlike corporate backed languages D is poor, all of those things will not happen just because someone asked, it may sound harsh but "if you need it - you make it, or pay someone to make it", that's what everyone says on that forum. And I mean they are not wrong. For example if thing X is exist in language A and thing Y is super easy to use that's because someone had to make it happen, and when nobody wants to make those nasty things the last argument is money, which you already probably guessed D lacks. And even with all that I disagree with that D is difficult and obsolete, for a teenager who will never code without IDE and simply doesn't care how computer/compilers/languages works? Maybe. For anyone who started with notepad as their IDE and crawled up on the hill of proficiency? No.
Jan 12
next sibling parent paulo pinto <pjmlp progtools.org> writes:
On Wednesday, 13 January 2021 at 06:11:14 UTC, evilrat wrote:
 On Wednesday, 13 January 2021 at 00:51:35 UTC, Marcone wrote:
...

 And even with all that I disagree with that D is difficult and 
 obsolete, for a teenager who will never code without IDE and 
 simply doesn't care how computer/compilers/languages works? 
 Maybe. For anyone who started with notepad as their IDE and 
 crawled up on the hill of proficiency? No.
Except that teenagers can also start businesses, and are the decision makers from tomorrow, so why bother cater to them anyway.
Jan 12
prev sibling parent reply Paul Backus <snarwin gmail.com> writes:
On Wednesday, 13 January 2021 at 06:11:14 UTC, evilrat wrote:
 On Wednesday, 13 January 2021 at 00:51:35 UTC, Marcone wrote:
 The impression I have is that everything in D is difficult, 
 old, obsolete and abandoned. I hope someday those responsible 
 for D will invest in IDE, integration with Qt, and with Qt 
 Designer to convert .ui to .d and a way to use packages that 
 is as easy as in Python, because the dub is very bad.
Welcome to the wonderful world of Open Source! Ok no time for celebrations, since unlike corporate backed languages D is poor, all of those things will not happen just because someone asked, it may sound harsh but "if you need it - you make it, or pay someone to make it", that's what everyone says on that forum. And I mean they are not wrong. For example if thing X is exist in language A and thing Y is super easy to use that's because someone had to make it happen, and when nobody wants to make those nasty things the last argument is money, which you already probably guessed D lacks.
This is all true, but the fact remains that some open source projects do a better job of working with contributors than others. D's leadership is almost legendarily bad at communicating openly and effectively with contributors, and the result is that despite having a community full of skilled and motivated developers, progress on important projects is often slow to nonexistent. The most recent example of this is Timon Gehr's tuple DIP, which is currently stalled due to "lack of enthusiasm from decision makers." [1] When someone like Timon shows up to volunteer his time and skills for your project, how arrogant and/or incompetent do you have to be to essentially slam the door in his face? People talk about "lack of volunteers," but the truth is that D *vastly* under-utilizes the volunteer resources it has already. That's the bottleneck in the current system, and that's what needs to be fixed if we want D to continue growing. [1] https://forum.dlang.org/post/rsrlkq$1nu0$1 digitalmars.com
Jan 12
next sibling parent drug <drug2004 bk.ru> writes:
On 1/13/21 10:21 AM, Paul Backus wrote:
 
 D *vastly* under-utilizes the volunteer resources it has already.
 
exactly
Jan 12
prev sibling next sibling parent reply evilrat <evilrat666 gmail.com> writes:
On Wednesday, 13 January 2021 at 07:21:26 UTC, Paul Backus wrote:
 D's leadership is almost legendarily bad at communicating 
 openly and effectively with contributors
 ...
Yes, there may be serious organizational/management problems, but the language itself is so cool. Being open source also means that it is possible that someone makes a fork with better management and vision and it might succeed, but on the other hand I hope this not happen any time soon, as D is barely living and such split might actually sink both projects. Heck, even I myself was thinking about making my own 'Better D' transpiler with new fancy sugar such as optional semicolon, fancy null conditional and null coalescing operators (aka ?. and ?? in Thanks to dmd as a library this might be even possible as a one man project.
Jan 13
next sibling parent Ola Fosheim =?UTF-8?B?R3LDuHN0YWQ=?= <ola.fosheim.grostad gmail.com> writes:
On Wednesday, 13 January 2021 at 08:13:18 UTC, evilrat wrote:
 Heck, even I myself was thinking about making my own 'Better D' 
 transpiler with new fancy sugar such as optional semicolon, 
 fancy null conditional and null coalescing operators (aka ?. 

 Thanks to dmd as a library this might be even possible as a one 
 man project.
I strongly encourage you to give this a try. The more experimental branches people create, the better. New features should not be added to the main branch without people having used it for some time IMHO. What I've done is to modify the compiler so that it accepts a new file-extension so that I can create my own modified parser as a duplicate, keeping the original parser for ".d" files. That way you can reuse all existing D-code and yet have full freedom to experiment. I haven't kept track of how much time I've spent on this, but maybe 3 days in total, starting from scratch? As you can see, I've made many small changes: https://github.com/OlaFosheimGrostad/dex . So, again, I would encourage you to try this. It takes some time to get familiar with the AST, but when you figure it out you have the ability to play with your ideas with little resistance. :) I haven't written enough tests for those changes though... so, I guess you would have to add at least 1 more day for that... :-)
Jan 13
prev sibling parent aberba <karabutaworld gmail.com> writes:
On Wednesday, 13 January 2021 at 08:13:18 UTC, evilrat wrote:
 On Wednesday, 13 January 2021 at 07:21:26 UTC, Paul Backus 
 wrote:
 D's leadership is almost legendarily bad at communicating 
 openly and effectively with contributors
 ...
Yes, there may be serious organizational/management problems, but the language itself is so cool. Being open source also means that it is possible that someone makes a fork with better management and vision and it might succeed, but on the other hand I hope this not happen any time soon, as D is barely living and such split might actually sink both projects.
That's what I fear and pray never happen. The Node.js project went through a similar experience before merging the fork back under a foundation. https://www.google.com/amp/s/www.theregister.com/AMP/2015/09/09/node_js_v400_reunites_with_io_js/ D is actually a foundation unlike Joyent which was rather a company before handing over to the Foundation created later to manage the project. I really hope we resolve this sooner.
Jan 13
prev sibling next sibling parent Imperatorn <johan_forsberg_86 hotmail.com> writes:
On Wednesday, 13 January 2021 at 07:21:26 UTC, Paul Backus wrote:
 On Wednesday, 13 January 2021 at 06:11:14 UTC, evilrat wrote:
 [...]
This is all true, but the fact remains that some open source projects do a better job of working with contributors than others. [...]
Maybe we need to give ppl more keys to the castle then, to improve the upstream experience. It's sad to hear stories like that w Timon.
Jan 13
prev sibling parent reply Mike Parker <aldacron gmail.com> writes:
On Wednesday, 13 January 2021 at 07:21:26 UTC, Paul Backus wrote:

 nonexistent. The most recent example of this is Timon Gehr's 
 tuple DIP, which is currently stalled due to "lack of 
 enthusiasm from decision makers."
Timon has not submitted his DIP yet. I don't see how anyone can blame the stalling of an unsubmitted DIP on the leadership.
Jan 13
parent Imperatorn <johan_forsberg_86 hotmail.com> writes:
On Wednesday, 13 January 2021 at 10:48:16 UTC, Mike Parker wrote:
 On Wednesday, 13 January 2021 at 07:21:26 UTC, Paul Backus 
 wrote:

 nonexistent. The most recent example of this is Timon Gehr's 
 tuple DIP, which is currently stalled due to "lack of 
 enthusiasm from decision makers."
Timon has not submitted his DIP yet. I don't see how anyone can blame the stalling of an unsubmitted DIP on the leadership.
Fair point. I think he means more like informal support though.
Jan 13
prev sibling parent Imperatorn <johan_forsberg_86 hotmail.com> writes:
On Wednesday, 13 January 2021 at 00:51:35 UTC, Marcone wrote:
 The impression I have is that everything in D is difficult, 
 old, obsolete and abandoned. I hope someday those responsible 
 for D will invest in IDE, integration with Qt, and with Qt 
 Designer to convert .ui to .d and a way to use packages that is 
 as easy as in Python, because the dub is very bad.
Also Marcone: "for me D is the best programming language" What's it gonna be? :D
Jan 13
prev sibling next sibling parent Imperatorn <johan_forsberg_86 hotmail.com> writes:
On Friday, 8 January 2021 at 16:26:43 UTC, Marcone wrote:
 I've been a programmer for 15 years. I program in several 
 programming languages, only for Windows desktop with graphical 
 interface. I have been programming in D for only 1 year and for 
 me D is the best programming language. Why not invest more in 
 Dlang? Better integration of D with Qt5 and Qt6 and the web 
 would make Dlang rise in popularity.
Yeah, D rocks, boulders even. One problem though is the upstream experience. Devs fork D and the changes doesn't get back in again. That experience needs to improve somehow
Jan 11
prev sibling parent Imperatorn <johan_forsberg_86 hotmail.com> writes:
On Friday, 8 January 2021 at 16:26:43 UTC, Marcone wrote:
 I've been a programmer for 15 years. I program in several 
 programming languages, only for Windows desktop with graphical 
 interface. I have been programming in D for only 1 year and for 
 me D is the best programming language. Why not invest more in 
 Dlang? Better integration of D with Qt5 and Qt6 and the web 
 would make Dlang rise in popularity.
Just a side note, D has actually been gaining a bit in popularity lately, breaking the trend a bit. According to github, members in Discord, Slack and Facebook
Jan 13