On Mon, 22 Oct 2018 03:49:44 -0400, Nick Sabalausky (Abscissa) wrote:
> So I'm honestly *shocked* to hear this. I NEVER would've guessed. I'm pretty sold on rolling-release at this point, but I'm intrigued enough that I'm gonna have to give the latest Ubuntu a try, at least in a VM.

The latest Ubuntu has Unity 7 as an option and defaults to GNOME, so you'd have to compile a Unity 8 prerelease from source.

On Sunday, 21 October 2018 at 17:15:03 UTC, Russel Winder wrote:
> On Sun, 2018-10-21 at 08:42 +0000, Paolo Invernizzi via Digitalmars-d wrote:
>> […]
>> 
>> Linux is not only the desktop, and Qt simply dominates in industrial, medical and automation sector, that's where the money is.
>> 
>> Qt is pushing strongly on the embedded marked, bare metal or linux (kernel) based...
>
> Which means that D not having a good play in the Qt space is a big
> barrier to adoption. This means it ought to be a strategic goal to have
> a (and I think I mean one here) D binding to Qt and QML.

It's easy to go and proclaim a strategic goal such as this. What actually matters is execution. And that requires some serious developer time that someone (ideally a whole team) needs to invest. I don't see this happening without some equally serious financial backing.

I, for one, won't have any time to work on such a thing in the foreseeable future.

Does anyone have experience with bounty programs? I envision one where bounties get posted with an open pool for each one. That is, if you agree that one task should be done, you can put additional money into its pool and make that particular bounty more attractive. It won't help attract professional developers but it might nudge some hobbyists to earn some money while taking on an interesting challenge. This might work for all the small things around D that lots of D users don't like and want improved.

On Mon, 2018-10-22 at 16:23 +0000, Gregor Mückl via Digitalmars-d wrote:
> 
[…]
> It's easy to go and proclaim a strategic goal such as this. What actually matters is execution. And that requires some serious developer time that someone (ideally a whole team) needs to invest. I don't see this happening without some equally serious financial backing.

Agreed. And sadly a common facet of the D situation: all words and no action – including from me.

> I, for one, won't have any time to work on such a thing in the foreseeable future.

In principle I could, but only as part of a team. This is not a one person project.

> Does anyone have experience with bounty programs? I envision one where bounties get posted with an open pool for each one. That is, if you agree that one task should be done, you can put additional money into its pool and make that particular bounty more attractive. It won't help attract professional developers but it might nudge some hobbyists to earn some money while taking on an interesting challenge. This might work for all the small things around D that lots of D users don't like and want improved.

I am not in a position to receive income, so I work for love or large payments to charities I nominate. Currently I am a bit overly obsessed with DVB and DAB+ using GStreamer, so mostly programming for love – and hence very fickle about doing stuff that requires commitment. The QtGStreamer has just been declared unmaintained :-(

-- 
Russel.
===========================================
Dr Russel Winder      t: +44 20 7585 2200
41 Buckmaster Road    m: +44 7770 465 077
London SW11 1EN, UK   w: www.russel.org.uk

On 2018-10-21 22:31, Patrick Schluter wrote:

> I like it and I'm looking forward that it gets beyond swt 3.4.
> I ported my Java GUI SWT program to D and it was a breeze to do. I didn't even require to change the structure of the app and the class hierarchy. There was only the file and string handling that I had to change, in fact make so much more readable and efficient.
> There were some difficulties because of compiler issues in version 2.7x, but those were resolved and everything went smooth after that.

That's great to hear :)

-- 
/Jacob Carlborg

On 2018-10-22 12:06, Russel Winder wrote:
> Jacob,
> 
> GitHub is currently making a total mess for me of our conversation on
> Issue 42, I see stuff then it goes away. Apologies if I have made a
> mess of that conversation for you.

Yeah, I noticed that. GitHub had/still having some major issues [1].

> Swing is, but JavaFX is now OpenJFX and a separate think to OpenJDK.

JavaFX is shipped with Java 8 at least, the one I was running (I'm a bit behind).
> Indeed. But compared to Qt (and maybe GTK+ and wxWidgets) it is very
> much a niche framework.

Why do you think that, besides from being written in Java?

> Or it has many native GUIs whereas Windows and macOS offer no choice?

Haha, I guess you can look at it that way.

> I still think getting a Qt binding for D à la PyQt, PySide2, Rust-Qt,
> i.e. automated with minimal manual tweeks, would be a Very Good Thing™
> for the D pitch in the desktop GUI applications arena.

Sure, different toolkits for different needs.

[1] https://status.github.com/messages

-- 
/Jacob Carlborg

A couple of notes on Qt.
Qt has for a while uses GTK on gnome and other GTK based desktops as the
native widget set.
Qt does do Mac native but not using the same code path as you would use on
other platforms, You are going to need to a reasonable amount of
version(mac)...else
Practically every commercial application I have purchased that runs on
Linux in the last 5 years is Qt Based (or rolls their own UI from scratch),
usually they control the theme, colour scheme rather than the system one.

IMO Qt is currently the least bad option when it comes to making cross platform productivity applications. Having stable bindings would be a huge help for D.


>
>