clayasaurus wrote:
> Sweeney has expressed interest in a new language so I don't think it would be too far fetched. No harm in trying, at least.

Hmm. Maybe you're right!

Georg Wrede wrote:
> Walter Bright wrote:
>>> Many of these features are new to the entire extended language family, or have been implemented in radically different ways than previously in the extended language family. Are mixins going to cause a problem with lambdas? Is liberal use of slices going to make DBC unmaintainable? Who knows! Who's going to find out? The early adopters.
<snip>
>> But on the other hand, I don't think this is a reason to be hesitant to use D. If those more unusual combinations cause a problem, just don't use them that way. It isn't necessary to remove the features from the language!
> 
> The C++ guys certainly shouldn't complain, they already do it every day.
> 
> And not using the entire feature set in each source file might be a good idea with any language. :-)

But the C++ guys do complain. For every thread on C++ at gamedev, there is no end to the complaints of experienced developers (I'm not counting fanboys). Unexpected interactions between features and undefined behavior is one source of frustration, what makes C++ complex.

If game developers are to move to a different language than C++, this has to be better, not the same. In this light, C# may look more favorably to some (indie at least, who don't need to port to all platforms). More lightweight, more tools, libraries, company support. I think if you want to attract game developers, you will have to compete with C# not C++. I'm willing to bet most sane C++ developers are already convinced they want to switch language if migration costs permit it.

I don't know if not using features is a valid option, because it's based on the premise that it's understood already how exactly they will interact with other features, and that 3rd party libraries will not use them. Even if this is so, it does increase complexity of development.

Is this viable in a project with, say, half a million lines of code?

Pragma wrote:

[snip]

>> We only have to become the _preferred_ language in _one_ area! After that it'll automatically follow that we gain other areas and a huge general interest. ("If C is good enough for Unix itself, it has to be a good language. -- If D is good enouhg for some of the best games shops, it has to be a good language.")
> 
> Agreed.  While I had a false start myself moving toward webapps for D, I think we're much closer than we were.  But add to that the current turbulence and quiet revolution we've seen in web design techniques, and things seem farther away than ever.

[snip]

Are you talking about Ajax here?  Frameworks like Rails / Django?  Just trying to clarify your point here.  Assuming that Ajax is what you're referring to, the need for a kick-ass server side remains, and it's where DSP could shine. Kris has said that small mods to Mango could allow it to handle the different demands of Ajax calls more effectively, as Jetty's new version does.  Couple that with a really easy server-side framework that wraps up a bunch of WS complexity and you're on your way.  Oooh, I'm getting OT again... shocking.

I'd like to put this delicately, but I'm not sure how.  In my mind, there's a strength of development that exists in game devs that's not there in web devs (maybe I'm just talking about myself ;) ).  However, I would guess the web devs are out there in larger numbers.  Perhaps this crowd would benefit from D's improvements and be the market Georg talks about.

BA

clayasaurus wrote:
> Georg Wrede wrote:
> 
>>
>>
 >
>> <snip> Not being an expert on this, does using several cores present some fundamental needs of change to a "regular" threading code?
>>
> 
> I'm not an expert either, but I think it is pretty much having threading built-in to the language itself.

I think the best use of those are 'vectorized' or data-parallel operations.  D is suited for this but not in its current incarnation.

-DavidM

Brad Anderson wrote:
> Pragma wrote:
> 
> [snip]
> 
>>> We only have to become the _preferred_ language in _one_ area! After
>>> that it'll automatically follow that we gain other areas and a huge
>>> general interest. ("If C is good enough for Unix itself, it has to be
>>> a good language. -- If D is good enouhg for some of the best games
>>> shops, it has to be a good language.")
>> Agreed.  While I had a false start myself moving toward webapps for D, I
>> think we're much closer than we were.  But add to that the current
>> turbulence and quiet revolution we've seen in web design techniques, and
>> things seem farther away than ever.  
> 
> [snip]
> 
> Are you talking about Ajax here?  Frameworks like Rails / Django?  Just trying
> to clarify your point here.  Assuming that Ajax is what you're referring to,
> the need for a kick-ass server side remains, and it's where DSP could shine.
> Kris has said that small mods to Mango could allow it to handle the different
> demands of Ajax calls more effectively, as Jetty's new version does.  Couple
> that with a really easy server-side framework that wraps up a bunch of WS
> complexity and you're on your way.  Oooh, I'm getting OT again... shocking.

As a matter of fact, yes I am. ;)

I've spent the better part of the past few months working on AJAX-based software (on the job at that!).  Rants about browser readyness* for deep Javascript apps aside, I've noticed that a grammar like DSP (as originally concieved) is becomming increasingly irrelevant for web-applications; look at Ruby on Rails.  Something that talks AJAX and WebSerivces as it's native tounge is really the way forward, and I am 0% complete on a D based solution there.  Granted, Mango+DDL is still how this will happen, but that's all just the underpinnings.

(* okay, one rant: IE's JS debugging support is deplorable enough to satisfy several anti-patterns *and* violate parts the Geneva Convention at the same time.)

> 
> I'd like to put this delicately, but I'm not sure how.  In my mind, there's a
> strength of development that exists in game devs that's not there in web devs
> (maybe I'm just talking about myself ;) ).  However, I would guess the web
> devs are out there in larger numbers.  Perhaps this crowd would benefit from
> D's improvements and be the market Georg talks about.

Well put, and you're not just talking about yourself.  Yea, some of us walk that line, but for the most part everyone has a focus and that tends to fall mostly within one problem domain or another.  There's no shame in that.

We can't all be John Carmack, Bjarne Stoustrup, or even Walter.  The rest of us have to settle for using their software instead.

-- 
- EricAnderton at yahoo

Lutger wrote:
> 
> If game developers are to move to a different language than C++, this has to be better, not the same. In this light, C# may look more favorably to some (indie at least, who don't need to port to all platforms). More lightweight, more tools, libraries, company support. I think if you want to attract game developers, you will have to compete with C# not C++.

For Microsoft platforms I very much agree.  Though that leaves Sony, Nintendo, and Apple as potential target vendors where C# is unlikely to be supported.  Cellphones and such as a possibility as well, though there D would be competing with Java.

> I'm willing to bet most sane C++ developers are already convinced they want to switch language if migration costs permit it.

Another big issue is support tools.  Without solid debug tool support, D will be a hard sell to most companies.

> I don't know if not using features is a valid option, because it's based on the premise that it's understood already how exactly they will interact with other features, and that 3rd party libraries will not use them. Even if this is so, it does increase complexity of development.
> 
> Is this viable in a project with, say, half a million lines of code?

It's not like this is a new problem.  Until just a few years ago, most popular C++ compilers were pretty finicky.  In fact I could time on a stopwatch how long it would take me to crash VC6, and VC6 is still used for a lot of projects.


Sean

clayasaurus wrote:

> Georg Wrede wrote:
>> 
>> 
>> clayasaurus wrote:
>>> Georg Wrede wrote:
>> Otoh, I guess we should not aspire to convert established major players at this point. The reason being, if I were the chairman over there I'd be scared to switch to an unknown language, if for nothing else, at least all our C++ libraries would become pretty hard to use.
>> 
> 
> Sweeney has expressed interest in a new language so I don't think it would be too far fetched. No harm in trying, at least.
> 

Do the search "sweeney site:digitalmars.com" at Google, and you'll find many posts from him in the oooooold D newsgroup, about 5 years ago.

-- 
Lars Ivar Igesund
blog at http://larsivi.net
DSource & #D: larsivi

Lutger wrote:
> Georg Wrede wrote:
> 
>> Walter Bright wrote:
>>> But on the other hand, I don't think this is a reason to be hesitant to use D. If those more unusual combinations cause a problem, just don't use them that way. It isn't necessary to remove the features from the language!
>>
>> The C++ guys certainly shouldn't complain, they already do it every day.
>>
>> And not using the entire feature set in each source file might be a good idea with any language. :-)
> 
> But the C++ guys do complain. For every thread on C++ at gamedev, there is no end to the complaints of experienced developers (I'm not counting fanboys). Unexpected interactions between features and undefined behavior is one source of frustration, what makes C++ complex.

Ah, my point exactly: they shouldn't complain about a few isolated "feature clash problems" in D. "Don't complain about eating bean stew when what you've eaten for years is just pig fodder." Except D is more like ice cream, IMHO. ;-)

> If game developers are to move to a different language than C++, this has to be better, not the same. 

I'd like to see the person who says D is not *much* better in this respect!

> In this light, C# may look more favorably to some (indie at least, who don't need to port to all platforms). 

You can't be serious! Maybe for an indie writing a new tic-tac-toe, but for Serious Development, I sure hope not.

> More lightweight, more tools, libraries, company support. I think if you want to attract game developers, you will have to compete with C# not C++. 

That's another thing. Tools and libraries we lack, no denying.

> I don't know if not using features is a valid option, because it's based on the premise that it's understood already how exactly they will interact with other features, and that 3rd party libraries will not use them. Even if this is so, it does increase complexity of development.

The C++ guys should be used to it.

> Is this viable in a project with, say, half a million lines of code?

I'd say not. But to start the next project in a new language (after suitably trying it out with a couple of prototyping projects), then yes.

Excellent thread! Ive said before that D *SHOULD* be the future of game development, and for some smart people it will be. Ive worked on several commercial game projects and it occurs to me that a large bulk of our engines were basically macros, templates, and patterns to implement many of the idioms D currently contains (and sometimes java); basically, to transform C++ into a 'better' language. And we still use C++ almost exclusively for performance reasons. This is why D should fill this space quite well. I tend to have a more laid back approach to all of this however, I believe we have a lot of work to do regarding the language infrastructure, and once that is in place the popularity will take care of itself. Build it and they will come.

Also I noticed one of the things sweeny mentioned in the DNG was persistent objects.... something we may be heading towards with DDL. So the future is pretty bright and I feel that once we have some of these world class features like reflection etc, many people will take a serious look at D. And on that note ive noticed some people will probably *never* be convinced that a language is better for them than c++, this has something to do with human nature that I dont fully understand. This is a big reason I stopped worrying about D popularity.

also as a big fan of QT, Id like to request a S&S mechanism; or some sort of messaging pattern in the language. I think this would take D "over the top!"

Brad Anderson wrote:
> 
>>>We only have to become the _preferred_ language in _one_ area! After
>>>that it'll automatically follow that we gain other areas and a huge
>>>general interest.
> 
> I'd like to put this delicately, but I'm not sure how.  In my mind, there's a
> strength of development that exists in game devs that's not there in web devs
> (maybe I'm just talking about myself ;) ).  However, I would guess the web
> devs are out there in larger numbers.  Perhaps this crowd would benefit from
> D's improvements and be the market Georg talks about.

Hmm. I think I see. :-)

At any rate, there are more web developers than games developers, so I don't have a problem with it.

My original point, after all, was that we do need to find a proper target and focus some effort in actually capturing it.

If it's games developers, web devs, whatever -- the point is that we serve D's interest best by giving that group our best. (Simply because we can't give our best to everybody, yet.)

The key here is _focus_. I's all the more important at a stage where the  number and depth of our libraries is small, general documentation is next to nothing, and the little we do have in documentation way is... (Not that I'd start throwing rocks in a glass house, I've been asked to help with documentation, and I've yet to pick that up. But I aim to.)

Any web developers here? What would _you_ like, wish, need, if you were just starting a web project in D?? -- Hmm. Actually, this is _the_ wrong forum to ask that. Anybody know somebody who's contemplating and who's not yet a D convertee? What's he thinking? What does he need? Or wish to get him up to speed?