tried to get d as the language of choice (daring) for a card projrct. it took almost 4 month with my ordering, threatening and ...

things died because of

1.) no gui lib with builder
2.) any other language had usable libs, extensions and standard algos
3.) competing standard libraries
4.) no foreseeable releases (such as 1.0)

most horrifying was the lack of an integrated editor debugger thingy.

as a user i think it is becoming at least deterring


richard


ps

even so i forced people for a long time, we are doing it noe in c# (kind   of kills me)

Ad 1):
There's DFL for Windows with Entice as a great 'form designer':
http://www.dprogramming.com/dfl.php
http://www.dprogramming.com/entice.php

Alex

On Wed, 2006-10-18 at 23:15 +0200, Richard Koch wrote:
> tried to get d as the language of choice (daring) for a card projrct. it took almost 4 month with my ordering, threatening and ...
> 
> things died because of
> 
> 1.) no gui lib with builder
> 2.) any other language had usable libs, extensions and standard algos
> 3.) competing standard libraries
> 4.) no foreseeable releases (such as 1.0)
> 
> most horrifying was the lack of an integrated editor debugger thingy.
> 
> as a user i think it is becoming at least deterring
> 
> 
> richard
> 
> 
> ps
> 
> even so i forced people for a long time, we are doing it noe in c# (kind
>    of kills me)

Richard Koch wrote:
> tried to get d as the language of choice (daring) for a card projrct. it took almost 4 month with my ordering, threatening and ...
> 
> things died because of
> 
> 1.) no gui lib with builder
> 2.) any other language had usable libs, extensions and standard algos
> 3.) competing standard libraries
> 4.) no foreseeable releases (such as 1.0)
> 
> most horrifying was the lack of an integrated editor debugger thingy.
> 
> as a user i think it is becoming at least deterring
> 
> 
> richard
> 
> 
> ps
> 
> even so i forced people for a long time, we are doing it noe in c# (kind   of kills me)

For 3), what's so bad about competition that stops you from doing work with D?

For 4) True enough, although it feels like that D 1.0 is right around the corner.

clayasaurus wrote:
> Richard Koch wrote:
>> tried to get d as the language of choice (daring) for a card projrct. it took almost 4 month with my ordering, threatening and ...
>>
>> things died because of
>>
>> 1.) no gui lib with builder
>> 2.) any other language had usable libs, extensions and standard algos
>> 3.) competing standard libraries
>> 4.) no foreseeable releases (such as 1.0)
>>
>> most horrifying was the lack of an integrated editor debugger thingy.
>>
>> as a user i think it is becoming at least deterring
>>
>>
>> richard
>>
>>
>> ps
>>
>> even so i forced people for a long time, we are doing it noe in c# (kind   of kills me)
> 
> For 3), what's so bad about competition that stops you from doing work with D?
> 
> For 4) True enough, although it feels like that D 1.0 is right around the corner.

I think you're right.

There are always going to be things that can be improved.  A big 1.0 total-feature-freeze-no-new-incompatibilities-ever release is never going to happen so that shouldn't be the goal.

If you look at the evolution of C and C++ there were certainly additions after "1.0" that caused a few incompatibilities.  Like the transition from KnR C to ANSI C or the addition of bool/true/false keywords to C++.   The way those were handled was generally with compiler flags in the transition periods.

D can use that strategy.

And future D's can still introduce incompatibilities.

1.0 just says, "this release will be maintained and stay accessible with the current feature set for a good long while".  It doesn't say progress will stop or that all future developments will be backwards compatible.  It just says that you can count on having 1.0 features for a while, with the only changes being bugfixes.

Well, that's my feeling anyway.  And I definitely didn't feel ready to say that a few months back.

--bb

Richard Koch wrote:
> 4.) no foreseeable releases (such as 1.0)

well in other languages you can have this just as well.
Until now each release of python has had something really useful.
So, there is always that little thing in your head that says, mmm, let's
wait until that or that feature is available, it would be just so much
better ;-)

Maybe Walter should release version 2.0 tomorrow, the first few weeks
everybody would be ? (fill in ...)
But after a few weeks, everybody would be like: whooaaw D has reach
version 2.0 and we're working towards a version 3.0 :->

roel

rm wrote:
> Richard Koch wrote:
>> 4.) no foreseeable releases (such as 1.0)
> 
> well in other languages you can have this just as well.
> Until now each release of python has had something really useful.
> So, there is always that little thing in your head that says, mmm, let's
> wait until that or that feature is available, it would be just so much
> better ;-)
> 
> Maybe Walter should release version 2.0 tomorrow, the first few weeks
> everybody would be ? (fill in ...)
> But after a few weeks, everybody would be like: whooaaw D has reach
> version 2.0 and we're working towards a version 3.0 :->
> 
> roel

Or maybe he should just quietly slide the decimal point over on the next release.   Next message from Walter:

   "DMD 1.730 release"

:-)

--bb

Bill Baxter wrote:
> rm wrote:
>> Richard Koch wrote:
>>> 4.) no foreseeable releases (such as 1.0)
>>
>> well in other languages you can have this just as well.
>> Until now each release of python has had something really useful.
>> So, there is always that little thing in your head that says, mmm, let's
>> wait until that or that feature is available, it would be just so much
>> better ;-)
>>
>> Maybe Walter should release version 2.0 tomorrow, the first few weeks
>> everybody would be ? (fill in ...)
>> But after a few weeks, everybody would be like: whooaaw D has reach
>> version 2.0 and we're working towards a version 3.0 :->
>>
>> roel
> 
> Or maybe he should just quietly slide the decimal point over on the next release.   Next message from Walter:
> 
>    "DMD 1.730 release"
> 
> :-)
> 
> --bb

oh yes, that would be unintentional I guess :-)

but regrettably, in std.compiler I do find the following:

    uint D_major = 0;
    uint D_minor = 134;

greetz,
roel

Richard Koch skrev:
> most horrifying was the lack of an integrated editor debugger thingy.
> 
> as a user i think it is becoming at least deterring
> 
What is wrong with Emacs and gdb?

Why not try out Walters own debugger tips at:
http://www.digitalmars.com/d/windbg.html
I am quite sure that if it is good enough for Walter, it is good enough for you and me.

Or Xcode and integrated gdb. Code completion does not work, but the rest is fully functional.


Sure there is no 'Digital Mars Visual D Studio 2005', but from my point of view that is a Good Thing(tm).


// Fredrik Olsson

Fredrik Olsson wrote:
> Richard Koch skrev:
>> most horrifying was the lack of an integrated editor debugger thingy.
>>
>> as a user i think it is becoming at least deterring
>>
> What is wrong with Emacs and gdb?
> 
> Why not try out Walters own debugger tips at:
> http://www.digitalmars.com/d/windbg.html
> I am quite sure that if it is good enough for Walter, it is good enough for you and me.

Absolutely false. In the context he is speaking of, managers are looking for tools that will enable RAD. D is certainly not capable of calling itself a RAD language. (For the language novice) With C#, most any competent programmer who has never seen the language before can sit down at the IDE and bang out an app in a day, with little hassle.

Kyle Furlong skrev:
> Fredrik Olsson wrote:
>> Richard Koch skrev:
>>> most horrifying was the lack of an integrated editor debugger thingy.
>>>
>>> as a user i think it is becoming at least deterring
>>>
>> What is wrong with Emacs and gdb?
>>
>> Why not try out Walters own debugger tips at:
>> http://www.digitalmars.com/d/windbg.html
>> I am quite sure that if it is good enough for Walter, it is good enough for you and me.
> 
> Absolutely false. In the context he is speaking of, managers are looking for tools that will enable RAD. D is certainly not capable of calling itself a RAD language. (For the language novice) With C#, most any competent programmer who has never seen the language before can sit down at the IDE and bang out an app in a day, with little hassle.

And who says D is a RAD language? The website in it's very first paragraph says: "D is a systems programming language". With the exception of the news archives I can not find a single hit on RAD on www.digitalmars.com/d.

So obviously if what someone wants is a RAD tool for writing UI-apps, then D is not the right tool, nor does the author claim so. It does not mean that D is any less good at solving the problem domain it do targets.


And I see that as a strength of D, not being tightly coupled with an IDE. C# is tightly coupled with Visual Studio, and is pretty useless without it. You can make it work, but well then it is no longer easy and "trouble free". A language do not need that property o be successful, and excel in it's field.

Java works pretty much every where, lots of IDE:s available, none required. Needing and IDE to get started is not an requirement for success or adoption; Perl, PHP, Python, C/C++, and countless others do just fine.

In fact the majority of languages get IDE-support because they are popular, they do not get popular because they have IDE-support.


// Fredrik Olsson