www.digitalmars.com         C & C++   DMDScript  

digitalmars.D - VisualD regressions are severe;

reply Manu <turkeyman gmail.com> writes:
So, since I've been off the wagon for a couple of years; VisualD, which
used to be ROCK SOLID seems to have suffered major regressions in almost
every aspect of its functionality.
I think it's largely related to dmd-as-a-lib now being the foundation for a
lot of tooling, and it's just criminally unstable...

Rainer used to maintain his own semantic analyser used for formatting,
auto-complete and suggestions, code navigation, and debugging; it worked
beautifully! But a couple years back, VisualD was switched to use DMD
frontend for those duties, and it barely works anymore.
The old bespoke code is still available, but it's so out of date with the
modern language that it's not usable anymore.

...to make this worse; Rainer has effectively checked-out too. We've lost
another one of our finest.

This is a general category of problem that's been an issue for a long time;
having unfunded one-man efforts maintain essential infrastructure. I wonder
if there are any opportunities available to do a lot better here. Does the
dlang foundation have any budget for critical infrastructure? And/or anyone
that would even consider working on the boring but essential stuff?

Is there actually anyone here who develops on Windows? I don't understand
how it could have regressed so far, unless it's just that nobody is using
it.

I now recognise a really major conundrum; I've recently returned to D to
start a company with a greenfields project. VisualD failing is essentially
terminal. I'm not sure what to do.
I don't have time available to try and pick up the project and work it
myself, but the current state is really pushing at the border of forcing me
to completely rewrite all my code in C++ on account of ecosystem
reliability.

Ideally, we really need to be properly funding development for critical
infrastructure... but I'm not sure we've ever had a sufficient budget to
maintain that sort of commitment.
Oct 15
next sibling parent reply "Richard (Rikki) Andrew Cattermole" <richard cattermole.co.nz> writes:
On 15/10/2024 10:07 PM, Manu wrote:

 ...to make this worse; Rainer has effectively checked-out too. We've 
 lost another one of our finest.


He implemented the negative annotation exportation switches for dmd at 
end of last year.

(dllimport and visibility)

Druntime/phobos are not currently offered as a shared library though.

That was a MAJOR contribution to dmd.


 I now recognise a really major conundrum; I've recently returned to D to 
 start a company with a greenfields project. VisualD failing is 
 essentially terminal. I'm not sure what to do. I don't have time 
 available to try and pick up the project and work it myself, but the 
 current state is really pushing at the border of forcing me to 
 completely rewrite all my code in C++ on account of ecosystem reliability.


Over the past year, VisualD has broken a couple of times.

The fix appears to update & recompile it.

Personally I only use VisualD for debugging, and Intellij for development.
Oct 15
parent reply Manu <turkeyman gmail.com> writes:
On Tue, 15 Oct 2024 at 19:16, Richard (Rikki) Andrew Cattermole via
Digitalmars-d <digitalmars-d puremagic.com> wrote:


 On 15/10/2024 10:07 PM, Manu wrote:

 ...to make this worse; Rainer has effectively checked-out too. We've
 lost another one of our finest.


 He implemented the negative annotation exportation switches for dmd at
 end of last year.

 (dllimport and visibility)

 Druntime/phobos are not currently offered as a shared library though.

 That was a MAJOR contribution to dmd.


Oh I know it... I mean, I have always said Rainer is one of the greatest,
most competent, most high impact, and most quiet and humble contributors
we've ever had... but I'm not sure what your point is. He seems to have
checked out... that's fine, he's welcome to do so, but it reinforces a
serious problem.
Key-person risk is a HUGE practical problem for D :/


 I now recognise a really major conundrum; I've recently returned to D to

 start a company with a greenfields project. VisualD failing is
 essentially terminal. I'm not sure what to do. I don't have time
 available to try and pick up the project and work it myself, but the
 current state is really pushing at the border of forcing me to
 completely rewrite all my code in C++ on account of ecosystem


 reliability.

 Over the past year, VisualD has broken a couple of times.

 The fix appears to update & recompile it.


To be clear, it's broken in almost every single semantic-related task that
it performs.
The way you describe here is because MS have changed their VS release model
so make major revision increments every couple of months rather than every
couple of years... and so yes, it needs to be regularly recompiled, but
that's not the cause of any functionality regression.

Personally I only use VisualD for debugging, and Intellij for development.


It's still the most competent debugger on earth.



 Oct 15



 next sibling parent  "Richard (Rikki) Andrew Cattermole" <richard cattermole.co.nz>  writes:


On 15/10/2024 10:42 PM, Manu wrote:

 On Tue, 15 Oct 2024 at 19:16, Richard (Rikki) Andrew Cattermole via 
 Digitalmars-d <digitalmars-d puremagic.com <mailto:digitalmars- 
 d puremagic.com>> wrote:
 
     On 15/10/2024 10:07 PM, Manu wrote:
      > ...to make this worse; Rainer has effectively checked-out too. We've
      > lost another one of our finest.
 
     He implemented the negative annotation exportation switches for dmd at
     end of last year.
 
     (dllimport and visibility)
 
     Druntime/phobos are not currently offered as a shared library though.
 
     That was a MAJOR contribution to dmd.
 
 
 Oh I know it... I mean, I have always said Rainer is one of the 
 greatest, most competent, most high impact, and most quiet and humble 
 contributors we've ever had... but I'm not sure what your point is. He 
 seems to have checked out... that's fine, he's welcome to do so, but it 
 reinforces a serious problem.
 Key-person risk is a HUGE practical problem for D :/


More like too busy. Which is not a good thing at all.


      > I now recognise a really major conundrum; I've recently returned
     to D to
      > start a company with a greenfields project. VisualD failing is
      > essentially terminal. I'm not sure what to do. I don't have time
      > available to try and pick up the project and work it myself, but the
      > current state is really pushing at the border of forcing me to
      > completely rewrite all my code in C++ on account of ecosystem
     reliability.
 
     Over the past year, VisualD has broken a couple of times.
 
     The fix appears to update & recompile it.
 
 
 To be clear, it's broken in almost every single semantic-related task 
 that it performs.
 The way you describe here is because MS have changed their VS release 
 model so make major revision increments every couple of months rather 
 than every couple of years... and so yes, it needs to be regularly 
 recompiled, but that's not the cause of any functionality regression.


But yes, that is basically what happened around a year ago.

Debugging didn't work. I had to disable the plugin it was so bad.



 Oct 15



prev sibling  parent  Guillaume Piolat <guillaume.piolat gmail.com>  writes:


On Tuesday, 15 October 2024 at 09:42:47 UTC, Manu wrote:

 To be clear, it's broken in almost every single 
 semantic-related task that
 it performs.
 The way you describe here is because MS have changed their VS 
 release model
 so make major revision increments every couple of months rather 
 than every
 couple of years... and so yes, it needs to be regularly 
 recompiled, but
 that's not the cause of any functionality regression.

 Personally I only use VisualD for debugging, and Intellij for 
 development.


I use VisualD a lot and I had sadly to disable semantic analysis.
You do loose completion, but still keep syntax, debugging, and 
search.
So it did become worse but what can we do to encourage things to 
become better?
DParser crashes are quite hard and time-consuming to isolate.



 Oct 17



prev sibling next sibling parent reply Max Samukha <maxsamukha gmail.com>  writes:


On Tuesday, 15 October 2024 at 09:07:12 UTC, Manu wrote:


 Is there actually anyone here who develops on Windows? I don't 
 understand how it could have regressed so far, unless it's just 
 that nobody is using it.


There seem to be near-zero users of D on Windows, and they are 
not using VisualD. I tried it recently but had to return to VS 
Code (which is, btw, currently unusable on Ubuntu 22.10, don't 
upgrade).



 Oct 15



  parent  Max Samukha <maxsamukha gmail.com>  writes:


On Tuesday, 15 October 2024 at 09:45:17 UTC, Max Samukha wrote:


 Ubuntu 22.10


24.10



 Oct 15



prev sibling next sibling parent reply Adam Wilson <flyboynw gmail.com>  writes:


On Tuesday, 15 October 2024 at 09:07:12 UTC, Manu wrote:

 This is a general category of problem that's been an issue for 
 a long time; having unfunded one-man efforts maintain essential 
 infrastructure. I wonder if there are any opportunities 
 available to do a lot better here. Does the dlang foundation 
 have any budget for critical infrastructure? And/or anyone that 
 would even consider working on the boring but essential stuff?


Got a spare 500k/USD/year? That's probably what it would take to 
staff the essentials.


 Is there actually anyone here who develops on Windows? I don't 
 understand how it could have regressed so far, unless it's just 
 that nobody is using it.


I work on Windows, but I've never used VisualD, just VSCode.



 Oct 15



  parent reply ryuukk_ <ryuukk.dev gmail.com>  writes:


On Tuesday, 15 October 2024 at 09:50:06 UTC, Adam Wilson wrote:

 On Tuesday, 15 October 2024 at 09:07:12 UTC, Manu wrote:

 This is a general category of problem that's been an issue for 
 a long time; having unfunded one-man efforts maintain 
 essential infrastructure. I wonder if there are any 
 opportunities available to do a lot better here. Does the 
 dlang foundation have any budget for critical infrastructure? 
 And/or anyone that would even consider working on the boring 
 but essential stuff?


 Got a spare 500k/USD/year? That's probably what it would take 
 to staff the essentials.


Typical american mindset, "we only hire people from the Bay Area 
or Washington, and they cost us 500k USD a year, deal with it", 
"oh we out of funds, call the VCs we need to hire 100 more people"

The solution is to stop scare people and to start catching up 
with other languages, all while cleaning up the old mess, D is 
failing behind, not in tooling since it never was any good, but 
in everything else, including in areas to make it appalling to 
talents


What other betterC languages out there doing?

Odin for example:

https://github.com/DanielGavin/ols


Supports all of the language features, including generics, ALL OF 
THEM


One man project



 Is there actually anyone here who develops on Windows? I don't 
 understand how it could have regressed so far, unless it's 
 just that nobody is using it.


 I work on Windows, but I've never used VisualD, just VSCode.


serve-d takes MINUTES to compile, this is not acceptable, you 
can't get any work done like that

Even the vscode experience sucks, LSP still doesn't understand 
what a template is or what is a mixin, wich is what phobos is 
made of, templates


Thankfully, my own lsp does, at least for simple cases, and 
compiles in less than 0.5 seconds (except for DCD)

https://github.com/ryuukk/dls


This is an area D foundation keeps ignoring (on top of ignoring 
DIPs for years, *cough* tuples)

So much work to do, so many years wasted discussing meaningless 
topics


We get things done because we care

If we don't care, we become careless and as a result it start to 
show on papers; and then people start complaining about it and 
end up quitting


Phobos 3 for example, alreay year in the works, and what's in 
phobos 3 today? you guessed right, more templates! yay!

Focus people, focus, i'm not smart, but i listen to what's going 
on in the world, and the world is begging for people to go back 
to C, let's make D an option

I do my part
https://github.com/ryuukk/dls

What about you, reader? send your PRs to DMD and tooling, be no 
scared, i dare you



 Oct 15



 next sibling parent  "Richard (Rikki) Andrew Cattermole" <richard cattermole.co.nz>  writes:


On 15/10/2024 11:20 PM, ryuukk_ wrote:

     Got a spare 500k/USD/year? That's probably what it would take to
     staff the essentials.
 
 Typical american mindset, "we only hire people from the Bay Area or 
 Washington, and they cost us 500k USD a year, deal with it", "oh we out 
 of funds, call the VCs we need to hire 100 more people"


You need to reread this. This is a baseless accusation, when Adam was 
talking about multiple roles.



 Oct 15



prev sibling next sibling parent reply Max Samukha <maxsamukha gmail.com>  writes:


On Tuesday, 15 October 2024 at 10:20:54 UTC, ryuukk_ wrote:

 the world is begging for people to go back to C,


A small number of loud gamedevs is not the world. Also, some of 
them first butcher every "complex and unnecessary" feature like 
namespaces or type-based overloading and then have to make 
excuses as in https://www.youtube.com/watch?v=a2B_izyma58.



 Oct 15



  parent  ryuukk_ <ryuukk.dev gmail.com>  writes:


On Tuesday, 15 October 2024 at 10:47:33 UTC, Max Samukha wrote:

 On Tuesday, 15 October 2024 at 10:20:54 UTC, ryuukk_ wrote:

 the world is begging for people to go back to C,


 A small number of loud gamedevs is not the world. Also, some of 
 them first butcher every "complex and unnecessary" feature like 
 namespaces or type-based overloading and then have to make 
 excuses as in https://www.youtube.com/watch?v=a2B_izyma58.


they are valid "excuses", have you read the endless move 
constructor thread? this kind of things causes bikeshedding, 
eating away time and effort that would be more useful elsewhere

impactful change fro example: tuple and tagged union

It's not just gamedevs, it's also the embedded world and AI/ML

Indians are embracing C as well, perhaps because of a shift in 
the industry? or the above, perhaps both ;)



 Oct 15



prev sibling next sibling parent  Mike Parker <aldacron gmail.com>  writes:


On Tuesday, 15 October 2024 at 10:20:54 UTC, ryuukk_ wrote:


 This is an area D foundation keeps ignoring (on top of ignoring 
 DIPs for years, *cough* tuples)


We aren't ignoring any DIPs. I've told you this before. The tuple 
DIP hasn't yet been submitted.



 Oct 15



prev sibling next sibling parent reply Adam Wilson <flyboynw gmail.com>  writes:


On Tuesday, 15 October 2024 at 10:20:54 UTC, ryuukk_ wrote:

 On Tuesday, 15 October 2024 at 09:50:06 UTC, Adam Wilson wrote:

 Got a spare 500k/USD/year? That's probably what it would take 
 to staff the essentials.


 Typical american mindset, "we only hire people from the Bay 
 Area or Washington, and they cost us 500k USD a year, deal with 
 it", "oh we out of funds, call the VCs we need to hire 100 more 
 people"


There about half-a-dozen people who work on D part-time, 
uncompensated, on projects that are marked as critical to the 
DPL/DLF. This does not include myself. 500k is a bare-bones 
budget to support them.

Or are you suggesting that we should all give up our day jobs and 
work on D full-time and unpaid so that you can get your free 
goodies faster? Meanwhile people like Razvan are left to what ... 
starve? That seems a bit self-defeating as I've never met anybody 
capable of writing code from a coffin. Humans need to eat. Food 
is acquired through the use of something called "money". Any 
donations to projects that do not produce money are strictly on 
an "as available" basis.

Was that "typical American" jab was an attempt to make me look 
like a heartless capitalist pig? Maybe try not suggesting 
starvation as the alternative? Not a good look mate.


 Phobos 3 for example, alreay year in the works, and what's in 
 phobos 3 today? you guessed right, more templates! yay!


Was anybody under the impression that Phobos 3 was going to be 
void of templates. If so I apologize for the misleading language. 
We are hoping to reduce the template usage in a few specific 
areas, but that's about it. Templates are an immensely useful 
language feature and we're not going to deny ourselves the use of 
them.



 Oct 17



 next sibling parent reply ryuukk_ <ryuukk.dev gmail.com>  writes:


On Friday, 18 October 2024 at 01:06:07 UTC, Adam Wilson wrote:

 On Tuesday, 15 October 2024 at 10:20:54 UTC, ryuukk_ wrote:

 On Tuesday, 15 October 2024 at 09:50:06 UTC, Adam Wilson wrote:

 Got a spare 500k/USD/year? That's probably what it would take 
 to staff the essentials.


 Typical american mindset, "we only hire people from the Bay 
 Area or Washington, and they cost us 500k USD a year, deal 
 with it", "oh we out of funds, call the VCs we need to hire 
 100 more people"


 There about half-a-dozen people who work on D part-time, 
 uncompensated, on projects that are marked as critical to the 
 DPL/DLF. This does not include myself. 500k is a bare-bones 
 budget to support them.

 Or are you suggesting that we should all give up our day jobs 
 and work on D full-time and unpaid so that you can get your 
 free goodies faster? Meanwhile people like Razvan are left to 
 what ... starve? That seems a bit self-defeating as I've never 
 met anybody capable of writing code from a coffin. Humans need 
 to eat. Food is acquired through the use of something called 
 "money". Any donations to projects that do not produce money 
 are strictly on an "as available" basis.

 Was that "typical American" jab was an attempt to make me look 
 like a heartless capitalist pig? Maybe try not suggesting 
 starvation as the alternative? Not a good look mate.


 Phobos 3 for example, alreay year in the works, and what's in 
 phobos 3 today? you guessed right, more templates! yay!


 Was anybody under the impression that Phobos 3 was going to be 
 void of templates. If so I apologize for the misleading 
 language. We are hoping to reduce the template usage in a few 
 specific areas, but that's about it. Templates are an immensely 
 useful language feature and we're not going to deny ourselves 
 the use of them.


D needs better marketer then

Weka it now worth 1.6B

https://www.weka.io/company/weka-newsroom/press-releases/weka-nets-140m-in-series-e-funding-at-1-6b-valuation/

They said they love D and it helped them greatly

Perhaps it's time to ask them for help and fund a tooling team?

Same for Symmetry, DConf is nice, but does that help bring more 
people in? what's the engagement of the DConf livestream looks 
like?

Weka browse the forum, time to help us, or we'll sink, you just 
said it, D foundation is low on money! Weka?! help us!



 Oct 18



 next sibling parent reply ryuukk_ <ryuukk.dev gmail.com>  writes:


On Friday, 18 October 2024 at 07:20:38 UTC, ryuukk_ wrote:

 On Friday, 18 October 2024 at 01:06:07 UTC, Adam Wilson wrote:

 On Tuesday, 15 October 2024 at 10:20:54 UTC, ryuukk_ wrote:

 On Tuesday, 15 October 2024 at 09:50:06 UTC, Adam Wilson 
 wrote:

 Got a spare 500k/USD/year? That's probably what it would 
 take to staff the essentials.


 Typical american mindset, "we only hire people from the Bay 
 Area or Washington, and they cost us 500k USD a year, deal 
 with it", "oh we out of funds, call the VCs we need to hire 
 100 more people"


 There about half-a-dozen people who work on D part-time, 
 uncompensated, on projects that are marked as critical to the 
 DPL/DLF. This does not include myself. 500k is a bare-bones 
 budget to support them.





https://ziglang.org/news/300k-from-mitchellh/


They are 200k short, too bad for them!



 Oct 18



  parent  ryuukk_ <ryuukk.dev gmail.com>  writes:


 I would like to use this opportunity to thank Bun and 
 TigerBeetle for supporting financially not only ZSF but also 
 ZigTools, the community initiative that supports the 
 development of ZLS, among other things.


ZLS is their language server, TigerBeetle is a successfull 
database company, written in Zig


Weka, don't you want free PR for your company? open source is 
trendy these days



 Oct 18



prev sibling next sibling parent  Sergey <kornburn yandex.ru>  writes:


On Friday, 18 October 2024 at 07:20:38 UTC, ryuukk_ wrote:

 Weka browse the forum, time to help us, or we'll sink, you just 
 said it, D foundation is low on money! Weka?! help us!


Weka managers: Tell developers to start rewriting all D in 
Zig/Rust
:D



 Oct 18



prev sibling  parent reply Paulo Pinto <pjmlp progtools.org>  writes:


On Friday, 18 October 2024 at 07:20:38 UTC, ryuukk_ wrote:

 On Friday, 18 October 2024 at 01:06:07 UTC, Adam Wilson wrote:

 [...]


 D needs better marketer then

 Weka it now worth 1.6B

 https://www.weka.io/company/weka-newsroom/press-releases/weka-nets-140m-in-series-e-funding-at-1-6b-valuation/

 They said they love D and it helped them greatly

 Perhaps it's time to ask them for help and fund a tooling team?

 Same for Symmetry, DConf is nice, but does that help bring more 
 people in? what's the engagement of the DConf livestream looks 
 like?

 Weka browse the forum, time to help us, or we'll sink, you just 
 said it, D foundation is low on money! Weka?! help us!


Wasn't Symmetry acquired? What are they using now?

Also looking at the Weka careers page, I am starting to see 
something else for filesystems programming.

https://www.weka.io/company/careers/?gh_jid=4234786007#career-position



 Oct 18



 next sibling parent reply "Richard (Rikki) Andrew Cattermole" <richard cattermole.co.nz>  writes:


On 18/10/2024 10:29 PM, Paulo Pinto wrote:

 On Friday, 18 October 2024 at 07:20:38 UTC, ryuukk_ wrote:

 On Friday, 18 October 2024 at 01:06:07 UTC, Adam Wilson wrote:

 [...]


 D needs better marketer then

 Weka it now worth 1.6B

 https://www.weka.io/company/weka-newsroom/press-releases/weka- 
 nets-140m-in-series-e-funding-at-1-6b-valuation/

 They said they love D and it helped them greatly

 Perhaps it's time to ask them for help and fund a tooling team?

 Same for Symmetry, DConf is nice, but does that help bring more people 
 in? what's the engagement of the DConf livestream looks like?

 Weka browse the forum, time to help us, or we'll sink, you just said 
 it, D foundation is low on money! Weka?! help us!


 
 Wasn't Symmetry acquired? What are they using now?


That sounds like Sociomantic which is from a long time ago now.



 Oct 18



  parent  Paulo Pinto <pjmlp progtools.org>  writes:


On Friday, 18 October 2024 at 09:40:25 UTC, Richard (Rikki) 
Andrew Cattermole wrote:

 On 18/10/2024 10:29 PM, Paulo Pinto wrote:

 On Friday, 18 October 2024 at 07:20:38 UTC, ryuukk_ wrote:

 [...]


 
 Wasn't Symmetry acquired? What are they using now?


 That sounds like Sociomantic which is from a long time ago now.


Ah, that is right, thanks for the correction.



 Oct 18



prev sibling  parent reply Sergey <kornburn yandex.ru>  writes:


On Friday, 18 October 2024 at 09:29:20 UTC, Paulo Pinto wrote:

 Also looking at the Weka careers page, I am starting to see 
 something else for filesystems programming.

 https://www.weka.io/company/careers/?gh_jid=4234786007#career-position


As D Gods said, it is the standard practice for D shops.
Nobody knows D, but if you know C++/Rust - you can switch to D 
quite easily.
That's why for hiring process nobody is mentioning D



 Oct 18



  parent  Paulo Pinto <pjmlp progtools.org>  writes:


On Friday, 18 October 2024 at 09:53:03 UTC, Sergey wrote:

 On Friday, 18 October 2024 at 09:29:20 UTC, Paulo Pinto wrote:

 Also looking at the Weka careers page, I am starting to see 
 something else for filesystems programming.

 https://www.weka.io/company/careers/?gh_jid=4234786007#career-position


 As D Gods said, it is the standard practice for D shops.
 Nobody knows D, but if you know C++/Rust - you can switch to D 
 quite easily.
 That's why for hiring process nobody is mentioning D


Maybe yes, maybe no, who knows.

What matters is what other people see, when looking for where D 
matters.



 Oct 18



prev sibling next sibling parent  ryuukk_ <ryuukk.dev gmail.com>  writes:


On Friday, 18 October 2024 at 01:06:07 UTC, Adam Wilson wrote:

 Phobos 3 for example, alreay year in the works, and what's in 
 phobos 3 today? you guessed right, more templates! yay!


 Was anybody under the impression that Phobos 3 was going to be 
 void of templates. If so I apologize for the misleading 
 language. We are hoping to reduce the template usage in a few 
 specific areas, but that's about it. Templates are an immensely 
 useful language feature and we're not going to deny ourselves 
 the use of them.


What was the motive to get started with Phobos 3?

What's in Phobos 3 today? see the point i try to make?

I'm not trying to sound mean or disrespectful, i apologies if i 
make it sound that way, but i have to mention what i see, i'd 
move away if i wasn't invested in D with my projects, and i 
believe in this language, but again, i don't want to be the only 
one who believes in it, if i am wrong, i'd hope for other people 
to let me know too, so i can do better the next time



 Oct 18



prev sibling next sibling parent  ryuukk_ <ryuukk.dev gmail.com>  writes:


On Friday, 18 October 2024 at 01:06:07 UTC, Adam Wilson wrote:

 There about half-a-dozen people who work on D part-time, 
 uncompensated, on projects that are marked as critical to the 
 DPL/DLF. This does not include myself. 500k is a bare-bones 
 budget to support them.

 Or are you suggesting that we should all give up our day jobs 
 and work on D full-time and unpaid so that you can get your 
 free goodies faster? Meanwhile people like Razvan are left to 
 what ... starve? That seems a bit self-defeating as I've never 
 met anybody capable of writing code from a coffin. Humans need 
 to eat. Food is acquired through the use of something called 
 "money". Any donations to projects that do not produce money 
 are strictly on an "as available" basis.

 Was that "typical American" jab was an attempt to make me look 
 like a heartless capitalist pig? Maybe try not suggesting 
 starvation as the alternative? Not a good look mate.


Let's make it a talk for DConf 2025, there is budget for that!



 Oct 18



prev sibling  parent  user1234 <user1234 12.de>  writes:


On Friday, 18 October 2024 at 01:06:07 UTC, Adam Wilson wrote:

 On Tuesday, 15 October 2024 at 10:20:54 UTC, ryuukk_ wrote:

 On Tuesday, 15 October 2024 at 09:50:06 UTC, Adam Wilson wrote:

 Got a spare 500k/USD/year? That's probably what it would take 
 to staff the essentials.


 Typical american mindset, "we only hire people from the Bay 
 Area or Washington, and they cost us 500k USD a year, deal 
 with it", "oh we out of funds, call the VCs we need to hire 
 100 more people"


 There about half-a-dozen people who work on D part-time, 
 uncompensated, on projects that are marked as critical to the 
 DPL/DLF. This does not include myself. 500k is a bare-bones 
 budget to support them.

 Or are you suggesting that we should all give up our day jobs 
 and work on D full-time and unpaid so that you can get your 
 free goodies faster? [...]


In my opinion the problem is not money, the problem would rather 
be the lack of people involved.

Example: the d-lang community org on GH (DCD, D-scanner, i.e most 
of the projects initially created by Hackerpilot, sure there's 
also D-YAML...) has the same problem as VD now. At the beginning, 
let's say during the first three years, that worked fine but at 
some point initial maintainers started being less focused but at 
the same time what was happening was a lack of contributors 
renewal. "No next-gen", to summarize.

By analogy: VD is maintained by one developer, this developer 
does something else, "bam", the project stalls.

Open-sources projects in the D ecosystem barly manage to create 
emulation. Beyond the compilers and the standard library it's 
most of the time single-man projects.



 Oct 18



prev sibling  parent reply Elias <desisma heidel.beer>  writes:


On Tuesday, 15 October 2024 at 10:20:54 UTC, ryuukk_ wrote:

 and what's in you guessed right, more templates! yay!


Good old template paradox…
I’m starting to love it ;)


“The meta programming capabilites of D are unmatched and are 
essential…”
“I really, really like D's templating and metaprogramming.”

But at the same time:
“Wait… You’re actually using templates!? No, don’t. They are bad 
for you. Oh no, more templates!”



 Oct 19



  parent  monkyyy <crazymonkyyy gmail.com>  writes:


On Sunday, 20 October 2024 at 00:14:25 UTC, Elias wrote:

 On Tuesday, 15 October 2024 at 10:20:54 UTC, ryuukk_ wrote:

 and what's in you guessed right, more templates! yay!


 Good old template paradox…
 I’m starting to love it ;)


 “The meta programming capabilites of D are unmatched and are 
 essential…”
 “I really, really like D's templating and metaprogramming.”

 But at the same time:
 “Wait… You’re actually using templates!? No, don’t. They are 
 bad for you. Oh no, more templates!”


Just use more templates and also embrace compiler bugs, paradox 
solved



 Oct 20



prev sibling next sibling parent  Kagamin <spam here.lot>  writes:


On Tuesday, 15 October 2024 at 09:07:12 UTC, Manu wrote:

 Is there actually anyone here who develops on Windows?


I work on windows, but I use suckless approach, IDE isn't 
necessary for that. If I wanted to write a bloatware, I would 




 Oct 15



prev sibling next sibling parent reply Meta <jared771 gmail.com>  writes:


On Tuesday, 15 October 2024 at 09:07:12 UTC, Manu wrote:

 So, since I've been off the wagon for a couple of years; 
 VisualD, which
 used to be ROCK SOLID seems to have suffered major regressions 
 in almost
 every aspect of its functionality.
 I think it's largely related to dmd-as-a-lib now being the 
 foundation for a
 lot of tooling, and it's just criminally unstable...

 Rainer used to maintain his own semantic analyser used for 
 formatting,
 auto-complete and suggestions, code navigation, and debugging; 
 it worked
 beautifully! But a couple years back, VisualD was switched to 
 use DMD
 frontend for those duties, and it barely works anymore.
 The old bespoke code is still available, but it's so out of 
 date with the
 modern language that it's not usable anymore.

 ...to make this worse; Rainer has effectively checked-out too. 
 We've lost another one of our finest.

 This is a general category of problem that's been an issue for 
 a long time; having unfunded one-man efforts maintain essential 
 infrastructure. I wonder if there are any opportunities 
 available to do a lot better here. Does the dlang foundation 
 have any budget for critical infrastructure? And/or anyone that 
 would even consider working on the boring but essential stuff?

 Is there actually anyone here who develops on Windows? I don't 
 understand how it could have regressed so far, unless it's just 
 that nobody is using it.

 I now recognise a really major conundrum; I've recently 
 returned to D to
 start a company with a greenfields project. VisualD failing is 
 essentially
 terminal. I'm not sure what to do.
 I don't have time available to try and pick up the project and 
 work it
 myself, but the current state is really pushing at the border 
 of forcing me
 to completely rewrite all my code in C++ on account of ecosystem
 reliability.

 Ideally, we really need to be properly funding development for 
 critical infrastructure... but I'm not sure we've ever had a 
 sufficient budget to maintain that sort of commitment.


What specific problems are you hitting? I don't write much D 
anymore (or do really *any* hobby programming, really), but I did 
recently dust off one of my old projects and get it compiling 
again with the latest dmd. I'm on Windows 11 and using Visual 
Studio with Visual D, and besides it not recognizing some stuff 
like the shortened method syntax, I haven't really noticed any 
issues.



 Oct 15



  parent reply Manu <turkeyman gmail.com>  writes:


On Wed, 16 Oct 2024, 01:51 Meta via Digitalmars-d, <
digitalmars-d puremagic.com> wrote:


 On Tuesday, 15 October 2024 at 09:07:12 UTC, Manu wrote:

 So, since I've been off the wagon for a couple of years;
 VisualD, which
 used to be ROCK SOLID seems to have suffered major regressions
 in almost
 every aspect of its functionality.
 I think it's largely related to dmd-as-a-lib now being the
 foundation for a
 lot of tooling, and it's just criminally unstable...

 Rainer used to maintain his own semantic analyser used for
 formatting,
 auto-complete and suggestions, code navigation, and debugging;
 it worked
 beautifully! But a couple years back, VisualD was switched to
 use DMD
 frontend for those duties, and it barely works anymore.
 The old bespoke code is still available, but it's so out of
 date with the
 modern language that it's not usable anymore.

 ...to make this worse; Rainer has effectively checked-out too.
 We've lost another one of our finest.

 This is a general category of problem that's been an issue for
 a long time; having unfunded one-man efforts maintain essential
 infrastructure. I wonder if there are any opportunities
 available to do a lot better here. Does the dlang foundation
 have any budget for critical infrastructure? And/or anyone that
 would even consider working on the boring but essential stuff?

 Is there actually anyone here who develops on Windows? I don't
 understand how it could have regressed so far, unless it's just
 that nobody is using it.

 I now recognise a really major conundrum; I've recently
 returned to D to
 start a company with a greenfields project. VisualD failing is
 essentially
 terminal. I'm not sure what to do.
 I don't have time available to try and pick up the project and
 work it
 myself, but the current state is really pushing at the border
 of forcing me
 to completely rewrite all my code in C++ on account of ecosystem
 reliability.

 Ideally, we really need to be properly funding development for
 critical infrastructure... but I'm not sure we've ever had a
 sufficient budget to maintain that sort of commitment.


 What specific problems are you hitting? I don't write much D
 anymore (or do really *any* hobby programming, really), but I did
 recently dust off one of my old projects and get it compiling
 again with the latest dmd. I'm on Windows 11 and using Visual
 Studio with Visual D, and besides it not recognizing some stuff
 like the shortened method syntax, I haven't really noticed any
 issues.


Syntax highlighting takes 10s of seconds, sometimes minutes, and may never
complete.
Go-to definition for most symbols don't work.
Press '.' after an aggregate or an enum or whatever, the list that pops up
rarely if ever shows any appropriate members or completion suggestions.
Ctrl-space (complete what I'm typing) doesn't work anymore.

I think it's all related to the new DMD-based semantic analysis bailing out
in various situations that I don't understand.

The debugger has significant problems to; important debug features that do
essential stuff like show the string for a custom string type have been
locking up under unpredictable circumstances.

It's all just very unstable and unreliable in general. Again I don't mean
to criticise Rainer; it's been rock solid for the past decade, but the
enormous scale of regressions since I was last actively using it really
demonstrates we have extreme key-person risk with this stuff... and this is
deal-breaker stuff. If Visual D doesn't work well, I will migrate my
company to C++; there is no practical alternative.

I've spent 15+ years trying to see D move beyond a hobby/curiosity/toy, but
this stuff shows we're not in good shape even after such a long time :/






 Oct 15



 next sibling parent reply monkyyy <crazymonkyyy gmail.com>  writes:


On Tuesday, 15 October 2024 at 16:25:37 UTC, Manu wrote:

 
 I've spent 15+ years trying to see D move beyond a 
 hobby/curiosity/toy, but this stuff shows we're not in good 
 shape even after such a long time :/


Have you been a d-doomer long or is this new?



 Oct 15



  parent reply Manu <turkeyman gmail.com>  writes:


On Wed, 16 Oct 2024 at 06:26, monkyyy via Digitalmars-d <
digitalmars-d puremagic.com> wrote:


 On Tuesday, 15 October 2024 at 16:25:37 UTC, Manu wrote:

 I've spent 15+ years trying to see D move beyond a
 hobby/curiosity/toy, but this stuff shows we're not in good
 shape even after such a long time :/


 Have you been a d-doomer long or is this new?



"Doom"? I mean, you need to be realistic, and honest. Point to the
industrial users of D; they're still the same ones as 10 years ago. I
haven't identified new ones. Almost all the people at dconf are the same
people.
I started a green-fields project in D recently; I thought this is great, it
doesn't need to bind with legacy, no C++ troubles, no awkward shim's or
build systems... my colleagues were enthusiastic about this direction, but
I had a conversation yesterday that started "are you sure this is the right
choice?"... because they have demonstrated to themselves in a very short
amount of time all the classic problems. They're not zealots; they're just
sensible and professional engineers, and trying to make sensible decisions
with respect to a business making technology commitments.

There's a certain amount of optimism that just can't sustain 16 years of
energy... and it's not like casual effort, I've spent at least 12 years
solidly working hard to get things into a workable shape. If that all goes
backward by a country mile when I look away for a hot minute; that's very
disheartening indeed.
So I guess the answer to your question is; yes, I approach D now with a
very different mood. It's changed from one of optimism to one of despair,
and I'm definitely in a last-hoorah mindset. There's some important
progress going on right now (which should have happened over a decade ago,
but I'll grant better late than never!), so that's encouraging, but
ancillary matters like extensive VisualD regressions are just something I
didn't expect, and/or have time or energy to deal with.

Threads like this are as old as the hills, and if we're not going to start
to take them seriously, then when? Maybe it's not that we're not taking
them seriously per-se, but nobody knows what to do. Rainer has been doing
god's work, but he's still just one guy, and he's doing other stuff right
now. He also has no obligation to make his time/energy available; and
of-course, you can say that about every person and every aspect of an open
source project like this, but to commit to that position is to send a clear
message that "D is not for industry". We've gotta do better than that
SOMEHOW; or else it's just as I say, this is a small community huddled
around a curiosity, a hobby; where just a couple of businesses with juuuust
the right set of contextual parameters have been able to make a sustained
commitment, but there's no broad path for growth... we've had way more than
enough time now to prove otherwise, but we haven't.
This isn't 'doom', it's just being realistic. As the champion of several
promising industrial efforts, over 3-4 companies (one with billion dollar
budgets), I still haven't been able to seal the deal. And now my own tiny
startup where I control literally everything... it still doesn't seem to be
possible to land. I guess that's on me; I'm a complete and comprehensive
failure! ...but I don't think that's entirely fair, because there's just no
comparable success stories that invalidate my experience. Find me some case
studies where a motivated individual tried to introduce D to their
organisation and were successful?

The ecosystem needs to recognise the key risk projects and strategically
direct resources to them. Look at it from the perspective of a business
considering a technology commitment; it's a time-bomb (as I've apparently
discovered at the most un-timely moment possible)... what do we do to
convince ourselves otherwise?



 Oct 15



 next sibling parent  Juraj <junk vec4.xyz>  writes:


Hi,
I fully understand your frustration.
IMHO what happened is VS Code just took the oxygen out of other 
tooling projects, as "no one" is using VisualStudio anymore.

other stuff, everyone just moved on.
The core of D tooling is DCD and I must say I am impressed how 
well it works once it is correctly set up and it handles Code 
Complete and Go To Definition pretty consistently (for a language 
with templates and mixins).

The debugger is a thought one, as slow as VS debugger is, it is 
just a good user experience, but it feels like most of D 
development is not done under Windows.
I heard RemedyBG is quite competent but I have zero experience 
with it.

On Tuesday, 15 October 2024 at 23:57:51 UTC, Manu wrote:

 The ecosystem needs to recognise the key risk projects and 
 strategically direct resources to them. Look at it from the 
 perspective of a business considering a technology commitment; 
 it's a time-bomb (as I've apparently discovered at the most 
 un-timely moment possible)... what do we do to convince 
 ourselves otherwise?


The fragmentation in D community use of editors and IDEs is very 
real, but I would expect VS Code being the better (best) out of 
the box experience. And D LSP 
[serve-d](https://github.com/Pure-D/serve-d)  being the bedrock 
for most IDE plugins today.

Years ago I would state that good VisualStudio integration is 
essential for a serious language, today I am not so sure.
Of course a broken one is even worst than none.


Juraj



 Oct 16



prev sibling  parent  monkyyy <crazymonkyyy gmail.com>  writes:


On Tuesday, 15 October 2024 at 23:57:51 UTC, Manu wrote:

 On Wed, 16 Oct 2024 at 06:26, monkyyy via Digitalmars-d < 
 digitalmars-d puremagic.com> wrote:


 On Tuesday, 15 October 2024 at 16:25:37 UTC, Manu wrote:

 I've spent 15+ years trying to see D move beyond a 
 hobby/curiosity/toy, but this stuff shows we're not in good 
 shape even after such a long time :/


 Have you been a d-doomer long or is this new?



 "Doom"? I mean, you need to be realistic, and honest.


Im counting myself among the doomer crowd.


 They're not zealots;
 important debug features


Im for the "printf debugger" team, not the big ide team. But ive 
also have some experiments in debugging(wasm+raylib, truly was 
awful not having basic strings handling once it got even 
slightly), so I made an in-"engine" debugger; hacky and used a 
compiler bug but it helped.

Im also looking into making a better writeln for opend.

Maybe I could point you at some options if your compromising on 
ide vs printf debate?



 Oct 16



prev sibling next sibling parent reply Meta <jared771 gmail.com>  writes:


On Tuesday, 15 October 2024 at 16:25:37 UTC, Manu wrote:

 On Wed, 16 Oct 2024, 01:51 Meta via Digitalmars-d, < 
 digitalmars-d puremagic.com> wrote:


 On Tuesday, 15 October 2024 at 09:07:12 UTC, Manu wrote:

 So, since I've been off the wagon for a couple of years;
 VisualD, which
 used to be ROCK SOLID seems to have suffered major 
 regressions
 in almost
 every aspect of its functionality.
 I think it's largely related to dmd-as-a-lib now being the
 foundation for a
 lot of tooling, and it's just criminally unstable...

 Rainer used to maintain his own semantic analyser used for
 formatting,
 auto-complete and suggestions, code navigation, and 
 debugging;
 it worked
 beautifully! But a couple years back, VisualD was switched to
 use DMD
 frontend for those duties, and it barely works anymore.
 The old bespoke code is still available, but it's so out of
 date with the
 modern language that it's not usable anymore.

 ...to make this worse; Rainer has effectively checked-out 
 too. We've lost another one of our finest.

 This is a general category of problem that's been an issue 
 for a long time; having unfunded one-man efforts maintain 
 essential infrastructure. I wonder if there are any 
 opportunities available to do a lot better here. Does the 
 dlang foundation have any budget for critical 
 infrastructure? And/or anyone that would even consider 
 working on the boring but essential stuff?

 Is there actually anyone here who develops on Windows? I 
 don't understand how it could have regressed so far, unless 
 it's just that nobody is using it.

 I now recognise a really major conundrum; I've recently
 returned to D to
 start a company with a greenfields project. VisualD failing 
 is
 essentially
 terminal. I'm not sure what to do.
 I don't have time available to try and pick up the project 
 and
 work it
 myself, but the current state is really pushing at the border
 of forcing me
 to completely rewrite all my code in C++ on account of 
 ecosystem
 reliability.

 Ideally, we really need to be properly funding development 
 for critical infrastructure... but I'm not sure we've ever 
 had a sufficient budget to maintain that sort of commitment.


 What specific problems are you hitting? I don't write much D 
 anymore (or do really *any* hobby programming, really), but I 
 did recently dust off one of my old projects and get it 
 compiling again with the latest dmd. I'm on Windows 11 and 
 using Visual Studio with Visual D, and besides it not 
 recognizing some stuff like the shortened method syntax, I 
 haven't really noticed any issues.


 Syntax highlighting takes 10s of seconds, sometimes minutes, 
 and may never
 complete.
 Go-to definition for most symbols don't work.
 Press '.' after an aggregate or an enum or whatever, the list 
 that pops up
 rarely if ever shows any appropriate members or completion 
 suggestions.
 Ctrl-space (complete what I'm typing) doesn't work anymore.


That's very weird. I'm using a recent version of Visual D from a 
few months ago, and I'm not seeing *any* of the problems you've 
mentioned. Granted I haven't needed to use the debugger in 
awhile, but syntax highlighting, go to definition, the symbol 
list and autocompletion... They all work. Is it possible the 
problem is due to some Visual D/Visual Studio settings? Like I 
have the setting checked to generate the full map file for me 
project, plus the JSON file that Visual D uses for a lot of 
stuff. Other than that, I dunno.

To be honest though, were I to start a new project for a startup, 
I would go with Java without question. D is fine for hobby 
programming and smaller scale stuff, but modern Java is Good 
Enough™️ - they're even getting pattern matching now with enhanced 
switch statements. I honestly don't know why anyone uses anything 
other than Java for serious projects, unless you have hard real 
time requirements or are working with microcontrollers (which you 
are, IIRC, so). The one downside is that there are *so* many 
shitty Java programmers (many whom I have worked with), that 
picking a more niche language automatically weeds out.



 Oct 16



  parent  Anonymous <anonymous example.com>  writes:


On Wednesday, 16 October 2024 at 21:14:29 UTC, Meta wrote:

 The one downside is that there are *so* many shitty Java 
 programmers (many whom I have worked with), that picking a more 
 niche language automatically weeds out.


Clojure or Scala fixes that.



 Oct 16



prev sibling  parent  ShadoLight <ettienne.gilbert gmail.com>  writes:


On Tuesday, 15 October 2024 at 16:25:37 UTC, Manu wrote:

 On Wed, 16 Oct 2024, 01:51 Meta via Digitalmars-d, < 
 digitalmars-d puremagic.com> wrote:


 On Tuesday, 15 October 2024 at 09:07:12 UTC, Manu wrote:




..

 Syntax highlighting takes 10s of seconds, sometimes minutes, 
 and may never
 complete.
 Go-to definition for most symbols don't work.
 Press '.' after an aggregate or an enum or whatever, the list 
 that pops up
 rarely if ever shows any appropriate members or completion 
 suggestions.
 Ctrl-space (complete what I'm typing) doesn't work anymore.

 I think it's all related to the new DMD-based semantic analysis 
 bailing out in various situations that I don't understand.

 The debugger has significant problems to; important debug 
 features that do essential stuff like show the string for a 
 custom string type have been locking up under unpredictable 
 circumstances.


In my experience VisualD often have issues with the latest (or 
even previous version) of VS. I think it was a constant issue for 
Rainer to accommodate the breaking changes from each new version 
of VS. Which version of VS are you using?

Personally I will not even bother to try it with VS 2022.

But I imagine that, since you are targeting micro-controllers, 
you probably don't need the latest and greatest VS - if you can 
tolerate it you can try with VS 2017 (or even 2015) - I still 
have a few issues but most of the stuff you report actually 
worked for me on VS 2015. I've changed some time ago to VS 2017, 
but since then didn't have the time to use VisualD extensively 
... but at least the basics seem to work.



 Oct 17



prev sibling next sibling parent reply RazvanN <razvan.nitu1305 gmail.com>  writes:


On Tuesday, 15 October 2024 at 09:07:12 UTC, Manu wrote:

 So, since I've been off the wagon for a couple of years; 
 VisualD, which
 used to be ROCK SOLID seems to have suffered major regressions 
 in almost
 every aspect of its functionality.
 I think it's largely related to dmd-as-a-lib now being the 
 foundation for a
 lot of tooling, and it's just criminally unstable...

 Rainer used to maintain his own semantic analyser used for 
 formatting,
 auto-complete and suggestions, code navigation, and debugging; 
 it worked
 beautifully! But a couple years back, VisualD was switched to 
 use DMD
 frontend for those duties, and it barely works anymore.
 The old bespoke code is still available, but it's so out of 
 date with the
 modern language that it's not usable anymore.

 ...to make this worse; Rainer has effectively checked-out too. 
 We've lost another one of our finest.

 This is a general category of problem that's been an issue for 
 a long time; having unfunded one-man efforts maintain essential 
 infrastructure. I wonder if there are any opportunities 
 available to do a lot better here. Does the dlang foundation 
 have any budget for critical infrastructure? And/or anyone that 
 would even consider working on the boring but essential stuff?

 Is there actually anyone here who develops on Windows? I don't 
 understand how it could have regressed so far, unless it's just 
 that nobody is using it.

 I now recognise a really major conundrum; I've recently 
 returned to D to
 start a company with a greenfields project. VisualD failing is 
 essentially
 terminal. I'm not sure what to do.
 I don't have time available to try and pick up the project and 
 work it
 myself, but the current state is really pushing at the border 
 of forcing me
 to completely rewrite all my code in C++ on account of ecosystem
 reliability.

 Ideally, we really need to be properly funding development for 
 critical infrastructure... but I'm not sure we've ever had a 
 sufficient budget to maintain that sort of commitment.


I don't think it's quite fair to say that the problem here was 
the switch
to dmdlib. The problem seems to be the fact that the project is 
not being
actively maintained. The same issues would have appeared even if 
the switch
wouldn't have been done due to the evolution of the language and 
the lack
of time to update visualD to the new dmd behaviors.

However, the broader problem that you are pinpointing: not having 
critical
infrastructure that is being managed by the foundation is indeed 
a problem.
But the reason for this is the lack of resources that can be 
directed to
what we consider to be important. I'd wish we could fix all of 
these, but
unfortunately, we just don't have the manpower.

I am trying to advance the state of dmdlib, butI don't think that 
there's anyone
else working on this.

RazvanN



 Oct 16



 next sibling parent  Eduard Staniloiu <edi33416 gmail.com>  writes:


On Wednesday, 16 October 2024 at 10:01:37 UTC, RazvanN wrote:

 On Tuesday, 15 October 2024 at 09:07:12 UTC, Manu wrote:


Hi Manu,

I fully agree that there is an infrastructure problem.

Why are you trying to use VisualD? I've never used it so I don't 
have experience with it, but reading between the lines (and I 
might be wrong here), does it offer a better debugging experience?

I've used code-d with VSCode and that works pretty well.

Given that you and your colleagues have tried to adopt D recently 
in the industry, could you give a list of the problems that 
you've hit? Maybe they can serve as a roadmap of pressing things 
that hinder adoption.



 Oct 16



prev sibling  parent reply Lance Bachmeier <no spam.net>  writes:


On Wednesday, 16 October 2024 at 10:01:37 UTC, RazvanN wrote:


 I'd wish we could fix all of these, but unfortunately, we just 
 don't have the manpower.


Well, add some more bureacracy to the process, I'm sure that will 
fix it. We don't have enough people doing stuff but at least we 
have the prettiest codebases that follow all the rules.



 Oct 18



 next sibling parent reply Adam Wilson <flyboynw gmail.com>  writes:


On Saturday, 19 October 2024 at 03:21:43 UTC, Lance Bachmeier 
wrote:

 On Wednesday, 16 October 2024 at 10:01:37 UTC, RazvanN wrote:


 I'd wish we could fix all of these, but unfortunately, we just 
 don't have the manpower.


 Well, add some more bureacracy to the process, I'm sure that 
 will fix it. We don't have enough people doing stuff but at 
 least we have the prettiest codebases that follow all the rules.


Well, some of us have itchy merge buttons and a willingness to 
bend a few rules to scratch that itch. Although my experience so 
far isn't that people mind the prettifying much, it's been more 
problematic when you suggest changes that are a bit more involved.



 Oct 18



  parent reply monkyyy <crazymonkyyy gmail.com>  writes:


On Saturday, 19 October 2024 at 06:58:45 UTC, Adam Wilson wrote:

 some of us have itchy merge buttons


https://issues.dlang.org/show_bug.cgi?id=20801
https://issues.dlang.org/show_bug.cgi?id=24307
https://gist.github.com/crazymonkyyy/432cf88729cd230fb0754b8d5b958653
https://gist.github.com/crazymonkyyy/6de535001d70a9bc45c0ea20913a38c1



 Oct 19



  parent reply Adam Wilson <flyboynw gmail.com>  writes:


On Saturday, 19 October 2024 at 17:08:54 UTC, monkyyy wrote:

 On Saturday, 19 October 2024 at 06:58:45 UTC, Adam Wilson wrote:

 some of us have itchy merge buttons


 https://issues.dlang.org/show_bug.cgi?id=20801
 https://issues.dlang.org/show_bug.cgi?id=24307
 https://gist.github.com/crazymonkyyy/432cf88729cd230fb0754b8d5b958653
 https://gist.github.com/crazymonkyyy/6de535001d70a9bc45c0ea20913a38c1


Gist's don't merge.



 Oct 23



  parent reply monkyyy <crazymonkyyy gmail.com>  writes:


On Thursday, 24 October 2024 at 06:09:45 UTC, Adam Wilson wrote:

 On Saturday, 19 October 2024 at 17:08:54 UTC, monkyyy wrote:

 On Saturday, 19 October 2024 at 06:58:45 UTC, Adam Wilson 
 wrote:

 some of us have itchy merge buttons


 https://issues.dlang.org/show_bug.cgi?id=20801
 https://issues.dlang.org/show_bug.cgi?id=24307
 https://gist.github.com/crazymonkyyy/432cf88729cd230fb0754b8d5b958653
 https://gist.github.com/crazymonkyyy/6de535001d70a9bc45c0ea20913a38c1


 Gist's don't merge.


Would you prefer an soac appication? Posts on phoboes v3 forum? 
Planning posts on this forum?

Im unsure how to *ever* get something merged into phoboes, lets 
start with my very first thing I believe belongs in phoboes, a 
circular buffer; I was told to wait on a feature that someone 
wanted over a decade ago... that isn't relevant to this spefic 
case even if I agreed the feature had merit(I dont, and that 
opinion I only streghting over time as more effort is wasted on 
it without shipping); waiting a decade to *start* is not eager.

At least be plausible and claim you have high standards(also no, 
Ive seen what phoboes is); you have a political problem not a man 
power shortage.



 Oct 26



  parent reply Elias (0xEAB) <desisma heidel.beer>  writes:


On Sunday, 27 October 2024 at 00:57:02 UTC, monkyyy wrote:

 On Thursday, 24 October 2024 at 06:09:45 UTC, Adam Wilson wrote:

 Gist's don't merge.


 Would you prefer an soac appication?


I guess, a PR would be something one could potentially merge.



 Oct 26



  parent reply monkyyy <crazymonkyyy gmail.com>  writes:


On Sunday, 27 October 2024 at 01:03:52 UTC, Elias (0xEAB) wrote:

 On Sunday, 27 October 2024 at 00:57:02 UTC, monkyyy wrote:

 On Thursday, 24 October 2024 at 06:09:45 UTC, Adam Wilson 
 wrote:

 Gist's don't merge.


 Would you prefer an soac appication?


 I guess, a PR would be something one could potentially merge.


If I translated my "weak" concept to the official style guide and 
added giant wall of text "documenting" it(being verbose and 
unclear via overt unnecessary formality to match phoboes style) 
would odds of success would you expect?



 Oct 26



 next sibling parent reply Elias (0xEAB) <desisma heidel.beer>  writes:


On Sunday, 27 October 2024 at 01:25:58 UTC, monkyyy wrote:

 would odds of success would you expect?


Higher ones than in the case where there is no PR to merge in the 
first place.



 Oct 26



  parent  monkyyy <crazymonkyyy gmail.com>  writes:


On Sunday, 27 October 2024 at 01:57:28 UTC, Elias (0xEAB) wrote:

 On Sunday, 27 October 2024 at 01:25:58 UTC, monkyyy wrote:

 would odds of success would you expect?


 Higher ones than in the case where there is no PR to merge in 
 the first place.


Invalid format please pick a number between 0-100



 Oct 26



prev sibling  parent reply Adam Wilson <flyboynw gmail.com>  writes:


On Sunday, 27 October 2024 at 01:25:58 UTC, monkyyy wrote:

 If I translated my "weak" concept to the official style guide 
 and added giant wall of text "documenting" it(being verbose and 
 unclear via overt unnecessary formality to match phoboes style) 
 would odds of success would you expect?


If you convert them into mergeable PR's, and if you are willing 
address change requests, and if you are willing to add the 
necessary documentation, then I don't see why merging them would 
be a problem.

And dropping the passive aggressive snark would definitely help 
your odds.



 Oct 26



  parent reply monkyyy <crazymonkyyy gmail.com>  writes:


On Sunday, 27 October 2024 at 02:27:23 UTC, Adam Wilson wrote:

 On Sunday, 27 October 2024 at 01:25:58 UTC, monkyyy wrote:

 If I translated my "weak" concept to the official style guide 
 and added giant wall of text "documenting" it(being verbose 
 and unclear via overt unnecessary formality to match phoboes 
 style) would odds of success would you expect?


 If you convert them into mergeable PR's, and if you are willing 
 address change requests, and if you are willing to add the 
 necessary documentation, then I don't see why merging them 
 would be a problem.


Also not a number and thats a question for the perma-optimists; 
and *for you* I really really intent to drive the point home; 
waiting a decade on a theatrical feature isn't eager. I was here 
for a few of them and made my opinion known at the time.


 some of us have itchy merge buttons




 And dropping the passive aggressive snark would definitely help 
 your odds.


I was unaware I was passive or snarky. Im livid, perhaps Im not 
communicating that well enough. I offered to help again and again 
and again; with wise restraint to want clear communication before 
starting on anything. Call me cynical, but I listened to others 
stories of being burned by the politics and didnt want to dive in 
head first.

You have no success integrating new people, and low success 
keeping people around. That's a political problem. Not a money 
problem; nor something you can dodge with someone who wants to 
pick a fight.

here let me write that in Phoboes-style for ya
```d
/**
Weakens an alias, so it must be compatible with T or it returns a 
valid T instead,(T.init), this can be used to make over loadable 
function arguments with default values. WeakAlias is for 
compiletime, WeakValue for runtime.
**/

auto ref WeakAlias(T,alias A)(){
	static if(is(T:typeof(A))){
		return A;
	} else {
		return T.init;
}}
auto ref WeakValue(T,S)(auto ref S a){
	static if(is(T:S)){
		return a;
	} else {
		return T.init;
}}

unittest{
	void foo(T)(T i=WeakValue!T(1)){}
	struct myint{int i;}
	foo(myint(2));
	//void bar(T)(T i=1){}
	//bar(myint(2));// fails, "cannot implicitly convert"
}
```



 Oct 26



  parent reply Elias (0xEAB) <desisma heidel.beer>  writes:


On Sunday, 27 October 2024 at 03:24:19 UTC, monkyyy wrote:

 here let me write that in Phoboes-style for ya


Why don’t you send a PR to the Phobos repo? :)
Most development is happening over there (and on Bugzilla), not 
this newsgroup.



 Oct 27



  parent reply monkyyy <crazymonkyyy gmail.com>  writes:


On Sunday, 27 October 2024 at 23:40:36 UTC, Elias (0xEAB) wrote:

 On Sunday, 27 October 2024 at 03:24:19 UTC, monkyyy wrote:

 here let me write that in Phoboes-style for ya


 Why don’t you send a PR to the Phobos repo? :)
 Most development is happening over there (and on Bugzilla), not 
 this newsgroup.


a) I care little for linus`s tool; I write very different code, 
and consider that whole back end to be an implementation detail 
for a free publishing service with bad ux
b) https://xkcd.com/1205/ *but worse* given probabilistic 
success, learning the core devs workflow, finding the right file, 
learning style of the commits message, and hell mimicking their 
god awful hyper formal dialict is painful
c) no previous reciprocation



 Oct 28



  parent reply Adam Wilson <flyboynw gmail.com>  writes:


On Monday, 28 October 2024 at 11:11:35 UTC, monkyyy wrote:

 On Sunday, 27 October 2024 at 23:40:36 UTC, Elias (0xEAB) wrote:

 On Sunday, 27 October 2024 at 03:24:19 UTC, monkyyy wrote:

 here let me write that in Phoboes-style for ya


 Why don’t you send a PR to the Phobos repo? :)
 Most development is happening over there (and on Bugzilla), 
 not this newsgroup.


 a) I care little for linus`s tool; I write very different code, 
 and consider that whole back end to be an implementation detail 
 for a free publishing service with bad ux
 b) https://xkcd.com/1205/ *but worse* given probabilistic 
 success, learning the core devs workflow, finding the right 
 file, learning style of the commits message, and hell mimicking 
 their god awful hyper formal dialict is painful
 c) no previous reciprocation


Well, then I don't know how to help you. Everybody else has to 
follow the procedures and use the tools. And I certainly can't 
ask somebody else to take your code and do the rest of the work 
for you. If you won't do that then I'm not sure what else I can 
say.



 Oct 29



  parent reply monkyyy <crazymonkyyy gmail.com>  writes:


On Wednesday, 30 October 2024 at 02:29:30 UTC, Adam Wilson wrote:

 
 Well, then I don't know how to help you.


I believe data structures the be important for a generic 
programming, and that std's are the natural fit for where they 
belong. I could easily fill a 3 page rant on the topic; but this 
shouldn't be that controversial.

I believe allocators are stupid, period. And its been too long to 
pretend its going to happen.

*You* could start merging data structures *from anyone*, into 
phoboes v3. If you wont accept this work from *anyone* due to the 
allocator drama, why claim to be eager about merging code?



 Oct 30



  parent reply Elias (0xEAB) <desisma heidel.beer>  writes:


On Wednesday, 30 October 2024 at 10:07:19 UTC, monkyyy wrote:

 I could easily fill a 3 page rant on the topic; but this 
 shouldn't be that controversial.


How about filling 3 pages of the PR queue with submissions for 
your ideas instead? :P



 I believe data structures the be important for a generic 
 programming, and that std's are the natural fit for where they 
 belong.


How would you like to get them into the standard library if not 
by sending them through the regular PR process? (Even for someone 
else to pick them up, they’d otherwise have to stumble over your 
links here in the newsgroup instead of just having a look at the 
open PRs.)



 Oct 30



  parent  monkyyy <crazymonkyyy gmail.com>  writes:


On Wednesday, 30 October 2024 at 14:10:42 UTC, Elias (0xEAB) 
wrote:

 
 How would you like to get them into the standard library if not 
 by sending them through the regular PR process?


I think your misunderstanding all three of my points

1) I dont use git expect as a requirement, such as github; yet I 
use github still if lazily

2) there could be a higher success rate for long term integration 
of first time contributors

3) there could be a reciprocation feedback loop; i.e. if my bugs 
got fixed I would spend more time on bug reports instead of 
workarounds and defensive syntax testing

These are conditional, not absolutes.


 (Even for someone else to pick them up, they’d otherwise have 
 to stumble over your links here in the newsgroup instead of 
 just having a look at the open PRs.)


Am I suppose to care if they are seen then discarded?

I believe 4 things would drastically improve the std which are 
trivial to write

1) merging data structures litterally ever
2) simplifying the clunky meta-programming templates
3) renaming passes/lazy clarifying of some horribly formal docs
4) inclusion of "trivial" one liner algorithms that are 
composites of other range functions

These suggestions are met with:

1) allocator are super important/ship on dub
2) templates are not important for v3
~~3) (havnt pushed on this one much, but renaming one of the 
`std.write`s would be a breaking change)~~
4) "to travail"



 Oct 30



prev sibling  parent reply Elias <desisma heidel.beer>  writes:


On Saturday, 19 October 2024 at 03:21:43 UTC, Lance Bachmeier 
wrote:

 Well, add some more bureacracy to the process, I'm sure that 
 will fix it. We don't have enough people doing stuff but at 
 least we have the prettiest codebases that follow all the rules.


And more eventloop libraries. We need more eventloop libraries. 
Not actual applications, just barebones stuff that wouldn’t pass 
muster when eventually used in real-world apps. It only needs to 
support this one unusual feature, the lack of which in other 
libraries is someone’s pet peeve.

(Oh, and image decoders. I feel like we don’t have enough 
libraries to throw our JPGs and PNGs in yet. Especially on the 
front of TGA we’re lacking.)

There sure must be something wrong with the plenty of existing 
libs, hence more and more are created. There surely is a reason. 



/s



 Oct 19



  parent  Adam Wilson <flyboynw gmail.com>  writes:


On Sunday, 20 October 2024 at 00:32:05 UTC, Elias wrote:

 On Saturday, 19 October 2024 at 03:21:43 UTC, Lance Bachmeier 
 wrote:

 Well, add some more bureacracy to the process, I'm sure that 
 will fix it. We don't have enough people doing stuff but at 
 least we have the prettiest codebases that follow all the 
 rules.


 And more eventloop libraries. We need more eventloop libraries. 
 Not actual applications, just barebones stuff that wouldn’t 
 pass muster when eventually used in real-world apps. It only 
 needs to support this one unusual feature, the lack of which in 
 other libraries is someone’s pet peeve.

 (Oh, and image decoders. I feel like we don’t have enough 
 libraries to throw our JPGs and PNGs in yet. Especially on the 
 front of TGA we’re lacking.)

 There sure must be something wrong with the plenty of existing 
 libs, hence more and more are created. There surely is a 


 just a myth!”

 /s


I suspect that in many ways this phenomena is a result of of how 
much easier it is to build things like that in D. Don't like what 
anybody else built, why not build your own, it'll take less time 
than pretty much every other language out there (especially Rust) 
and you'll own it and can do whatever you want with it.

That said, it's also a sign, particularly on the data formats 
front that it's time for the standard library to step in start 
offering ... you know ... standards. Langauges with large 
standard libraries don't seem to suffer this fate as much, 
because in 99% of all cases, you would be better off used the 
standard implementation and getting about your actual work.



 Oct 20



prev sibling next sibling parent  GrimMaple <grimmaple95 gmail.com>  writes:


On Tuesday, 15 October 2024 at 09:07:12 UTC, Manu wrote:

 ...


Last time I checked, most of Visual D issues were caused by the 
new Visual Studio versions, and it worked somewhat fine on Visual 
Studio 2017 (or 2019, you can try that too). As far as it goes, I 
would deeply recommend using VS Code instead, as it generally 
gives a better experience (visuald sucks, and I only used it a 
few times for the debugger)

Also, I encourage you to check out OpenD if you want to see D 
moving... _somewhere_, I guess



 Oct 18



prev sibling  parent reply Chris Piker <chris hoopjump.com>  writes:


On Tuesday, 15 October 2024 at 09:07:12 UTC, Manu wrote:


 Ideally, we really need to be properly funding development for 
 critical infrastructure... but I'm not sure we've ever had a 
 sufficient budget to maintain that sort of commitment.


Which of the multiple donation destinations does the D foundation 
prefer these days for donations?  I see about four options and am 
not sure which is the preferred destination at this point.

Thanks,



 Oct 20



  parent reply Mike Parker <aldacron gmail.com>  writes:


On Sunday, 20 October 2024 at 21:20:14 UTC, Chris Piker wrote:

 On Tuesday, 15 October 2024 at 09:07:12 UTC, Manu wrote:



 Which of the multiple donation destinations does the D 
 foundation prefer these days for donations?  I see about four 
 options and am not sure which is the preferred destination at 
 this point.


Whatever is best for you is the right answer. That's why we have 
different options. Some people can't or don't want to use PayPal, 
some people prefer OpenCollective in support of their mission, 
some people prefer GitHub because they are already supporting 
other projects there, etc.



 Oct 21



  parent reply Chris Piker <chris hoopjump.com>  writes:


On Monday, 21 October 2024 at 14:12:53 UTC, Mike Parker wrote:


 Whatever is best for you is the right answer.


Okay I've selected one.  Since I'm about to start using Visual 
Studio for a new project, Visual-D is something I'd like to 
support.  Thanks for putting visual-d upkeep on the front burner.



 Oct 21



  parent reply Manu <turkeyman gmail.com>  writes:


On Tue, 22 Oct 2024 at 03:26, Chris Piker via Digitalmars-d <
digitalmars-d puremagic.com> wrote:


 On Monday, 21 October 2024 at 14:12:53 UTC, Mike Parker wrote:


 Whatever is best for you is the right answer.


 Okay I've selected one.  Since I'm about to start using Visual
 Studio for a new project, Visual-D is something I'd like to
 support.  Thanks for putting visual-d upkeep on the front burner.


The problem is that there is only one maintainer. He's not interested or
motivated by funding in the past.
In order for it to have a healthier existence, it needs more than one
maintainer, and if another could be motivated with funding then that's
something, but funding needs to be specifically directed to that person.
There's no dlang foundation effort to try and find/fund maintainers for
these essential projects.

The interesting catch that I reckon we can see with Visual Studio, is that
it's usually industry professionals that are using it, and as such they are
less likely to have bandwidth away from their work to work on that. It's
not an ecosystem that lends to hobbyists so much; the overlap in the venn
diagram between VS users and dlang hobbyists is small.
Ideally, a D company that uses VS should direct one of their staff to have
some hours dedicated to tooling... but there aren't any such companies as
I'm aware?



 Oct 23



  parent reply Adam Wilson <flyboynw gmail.com>  writes:


On Thursday, 24 October 2024 at 03:58:43 UTC, Manu wrote:

 On Tue, 22 Oct 2024 at 03:26, Chris Piker via Digitalmars-d < 
 digitalmars-d puremagic.com> wrote:

 The problem is that there is only one maintainer. He's not 
 interested or
 motivated by funding in the past.
 In order for it to have a healthier existence, it needs more 
 than one
 maintainer, and if another could be motivated with funding then 
 that's
 something, but funding needs to be specifically directed to 
 that person.
 There's no dlang foundation effort to try and find/fund 
 maintainers for
 these essential projects.

 The interesting catch that I reckon we can see with Visual 
 Studio, is that
 it's usually industry professionals that are using it, and as 
 such they are
 less likely to have bandwidth away from their work to work on 
 that. It's
 not an ecosystem that lends to hobbyists so much; the overlap 
 in the venn
 diagram between VS users and dlang hobbyists is small.
 Ideally, a D company that uses VS should direct one of their 
 staff to have
 some hours dedicated to tooling... but there aren't any such 
 companies as
 I'm aware?


You are correct. If you're using all-up Visual Studio, you're 
paying for a yearly license, and these things aren't cheap (I 
have one, AMA). So if you're using one, you're probably very 
invested in the MSFT ecosystem. So I have, because at work I use 
SQL Server, Azure, and .NET Framework, in addition to .NET Core 
and Linux.



Realistically, the vast majority of our users who are going to 
interested in LSP's and Debuggers are going to be on VS Code. I 
know there are many Emacs and VIM users as well, but for the 
specific purposes you're talking about the high-value impact is 
going to be working on VS Code support.

Is what it is. 🤷‍♂️



 Oct 23



 next sibling parent  Paulo Pinto <pjmlp progtools.org>  writes:


On Thursday, 24 October 2024 at 06:17:21 UTC, Adam Wilson wrote:

 On Thursday, 24 October 2024 at 03:58:43 UTC, Manu wrote:

 On Tue, 22 Oct 2024 at 03:26, Chris Piker via Digitalmars-d < 
 digitalmars-d puremagic.com> wrote:

 The problem is that there is only one maintainer. He's not 
 interested or
 motivated by funding in the past.
 In order for it to have a healthier existence, it needs more 
 than one
 maintainer, and if another could be motivated with funding 
 then that's
 something, but funding needs to be specifically directed to 
 that person.
 There's no dlang foundation effort to try and find/fund 
 maintainers for
 these essential projects.

 The interesting catch that I reckon we can see with Visual 
 Studio, is that
 it's usually industry professionals that are using it, and as 
 such they are
 less likely to have bandwidth away from their work to work on 
 that. It's
 not an ecosystem that lends to hobbyists so much; the overlap 
 in the venn
 diagram between VS users and dlang hobbyists is small.
 Ideally, a D company that uses VS should direct one of their 
 staff to have
 some hours dedicated to tooling... but there aren't any such 
 companies as
 I'm aware?


 You are correct. If you're using all-up Visual Studio, you're 
 paying for a yearly license, and these things aren't cheap (I 
 have one, AMA). So if you're using one, you're probably very 
 invested in the MSFT ecosystem. So I have, because at work I 
 use SQL Server, Azure, and .NET Framework, in addition to .NET 
 Core and Linux.



 Realistically, the vast majority of our users who are going to 
 interested in LSP's and Debuggers are going to be on VS Code. I 
 know there are many Emacs and VIM users as well, but for the 
 specific purposes you're talking about the high-value impact is 
 going to be working on VS Code support.

 Is what it is. 🤷‍♂️


Additionally, with the pressure of other language ecosystems, 
game industry adoption for tooling and scripting, and now 

steadly improved to the point D isn't really worth looking into, 
for companies already invested into .NET ecosystem.

Sure metaprogramming is much easier in D than the pain of dealing 
with Code Generators, but that is about it, AOT is now part of 
the picture, lots of improvements for low level systems like 
programming, SIMD support, and wide adoption among key players in 
the games industry, e.g. Devil May Cry for PS5 uses a CAPCOM 
in-house engine built on a fork from .NET Core.

And regardless of Walter's opinion on C++/CLI, it is pretty much 
alive among .NET developers that need to interop with C and C++ 
code on Windows, it even handles C++20 (minus modules).



 Oct 24



prev sibling  parent reply Chris Piker <chris hoopjump.com>  writes:


On Thursday, 24 October 2024 at 06:17:21 UTC, Adam Wilson wrote:


 You are correct. If you're using all-up Visual Studio, you're 
 paying for a yearly license, and these things aren't cheap (I 
 have one, AMA).


Ah, didn't know that. Visual Studio is only $65/year for academic 
institutions.  Had no idea it was so expensive in the outside 
world.  I figured MS subsidized the cost to get more developers 
working on Windows.



 Oct 24



  parent  Manu <turkeyman gmail.com>  writes:


On Fri, 25 Oct 2024 at 06:21, Chris Piker via Digitalmars-d <
digitalmars-d puremagic.com> wrote:


 On Thursday, 24 October 2024 at 06:17:21 UTC, Adam Wilson wrote:


 You are correct. If you're using all-up Visual Studio, you're
 paying for a yearly license, and these things aren't cheap (I
 have one, AMA).


 Ah, didn't know that. Visual Studio is only $65/year for academic
 institutions.  Had no idea it was so expensive in the outside
 world.  I figured MS subsidized the cost to get more developers
 working on Windows.


There is Visual Studio Community Edition these days, which is free.
It used to be fairly feature reduced, but today it's basically
full-featured; I use Community Edition, and there's nothing at all missing
compared to the Enterprise version that I care about.
So, for all intents and purposes, VS is free (for non-commercial use).



 Oct 24