View mode: basic / threaded / horizontal-split · Log in · Help
December 12, 2005
Re: A safer/better C++?
I am waiting for the game industry to discover D. Many studios have libraries
and macros or custom preprocessors with built in ice makers and all sorts of
bells and whistles just to construct idioms within C++ to make it something akin
to D. I agree that there is some interesting psychology regarding languages. Its
as if the value of designing a new language and that of using Duct Tape v.2005
is completely out of whack. The game industry is going to see a monumental
increase in size and complexity of data before we know it. Some development
tools are going to crush under their own weight. My current position is D
*should* be the future of the game industry. Its just not worth the cost benifit
whatchamacallit to continue doing things the way they are now vs redesigning the
language which creates the foundation to the tools and engines. Which reminds me
of a thought I was having while porting the D front end to D. C/C++ was
originally as much a language spec as a compiler the same way D is as much a
language spec and a compiler. The complexity of the language is somewhat
relative to the complexity of the language the compiler is written in. C was
built using assembler, the C++ preprocessor using c? and D using C++. This is a
fundamental advance that should have occured a long time ago. The fact that the
software industry in general has not begun down this path before has as much to
do with collective psychology as money, management, legacy whatever. I had many
of the same ideas as dave when I first found D a couple years ago, D is cool but
lets just tweak C++ to hell and back. On a side note I think a lot of this has
to do with lack of compiler experience and this is where walter has become our
hero. So for whatever reason here we are today in 2005 with this situation. I
very much think if people in the game industry understood D they would fall in
love with it. One thing to mention however, most every game studio holds their
internal procedures, libraries, tools, and cup cake recipies as guarded as the
1985 bears with their playbook. It would be interesting to see how many
developers would even want to advertise their use of something like D. Not that
I would have any problem with it but its interesting community. I would like to
write an article for game developer magazine but im too busy. im planning a
studio around a brand new engine and suite of tools built in D.
December 12, 2005
Re: A safer/better C++?
"Walter Bright" <newshound@digitalmars.com> wrote in message 
news:dniddb$1dk0$1@digitaldaemon.com...
>
> "Dave" <Dave_member@pathlink.com> wrote in message
> news:dni7a9$viu$1@digitaldaemon.com...
>> After reading some of the threads, and seeing mention of D in there
> basically
>> just get ignored, it becomes apparent that one could come up with the
> better
>> mousetrap and still the very people who should take notice won't even
>> acknowledge the existance of something that quite probably is not only
> better
>> but readily available.
>
> There is some curious psychology going on with this. I've met many well
> known C++ experts who won't say anything positive about D in public, but 
> in
> private to me they are quite enthusiastic about it. It's almost as if they
> fear censure or loss of face?
>
> One thing D does is defy the conventional wisdom about C++. There are two
> that you mention; some others are:
> 3) C++'s non-safety is necessary to get its power and flexibility
> 4) The solution to C++ programming language problem X is to educate the
> programmers better
>
>
>> So it seems D is stuck in the proverbial 'catch 22' situation here, where
> the
>> real merits of the language only become apparent after learning how to 
>> use
> it,
>> yet those who need to try it won't because of their (understandable)
>> assumptions...
>>
>> How can D break out of this situation?
>
> The biggest thing is to write articles about D and get them published.
> Present papers on D at conferences. Continue to bring up D in contexts 
> where
> it is appropriate. Dave - an article about performance? Kris - an article
> about Mango? Don - an article about D templates? Thomas - an article about
> your automated testing system? David - an article about implementing GDC?
> Everyone else, too, who has made important contributions to D and this 
> n.g.
> You guys all know your stuff and these would be important, interesting
> articles.

What about a place where all this expertise can be shared, where people can 
publish articles, tips and code? I'm thinking of an MSDN for D, a D 
Developers Network if you like. Community blogs are also pretty effective. 
Artima (http://www.artima.com/) might be a good model to follow.

I'm not sure getting into CUJ is the best way - it's primarily a C++ 
magazine, and while there are occasional series on other languages, how many 
articles about D will its editors take?

>
> Me, I'll be presenting at Amazon's developer conference in January and in
> SDWest in March. But D can't succeed if it's just me writing articles and
> presenting.
>
> Would it help if I offered a bounty for articles? <g>
>

Rhetorical, surely.
December 12, 2005
Re: A safer/better C++?
> How about a programming challenge? Develop a programming problem that 
> requires
> the contestant to program against a D language API.

There's always the language shoot-out, but I've recently noticed that 
intel's C compiler took over the first spot!?

L.
December 12, 2005
Re: A safer/better C++?
"Lionello Lunesu" <lio@remove.lunesu.com> wrote in message 
news:dnjo96$17tr$1@digitaldaemon.com...
>> How about a programming challenge? Develop a programming problem that 
>> requires
>> the contestant to program against a D language API.
>
> There's always the language shoot-out, but I've recently noticed that 
> intel's C compiler took over the first spot!?
>
> L.

Yes, and SmartEiffel just took over the lead from both...

There are some problems with a couple of tests that run well on my system, 
but don't on theirs.

The tests are "Reverse Complement" and "Cheap-Concurrency". More is 
discussed here:

http://www.digitalmars.com/drn-bin/wwwnews?digitalmars.D.learn/2287
(Bascially, Reverse Complement runs (and should) in half a second on my 
machine, but takes 10 secs. on their equally matched machine)

Anybody running Debian unstable they could use to check these problems out?

Another problem with "Cheap-Concurrency" is that the number of threads that 
phobos supports on Linux is artificially low (hence the loop that re-creates 
threads).

Also, we're missing two tests and I don't have the time to figure them out: 
Regex DNA and Chameneos. Anyone want to take a stab at those? Make sure to 
follow the "rules" for each test; the FAQ for the site is here: 
http://shootout.alioth.debian.org/faq.php

Thanks,

- Dave
December 12, 2005
Re: A safer/better C++?
Dave wrote:

> Also, we're missing two tests and I don't have the time to figure them out: 
> Regex DNA and Chameneos. Anyone want to take a stab at those? Make sure to 
> follow the "rules" for each test; the FAQ for the site is here: 

I implemented the Regex DNA a while back. Really straight-forward 
looking at other solutions :). I never submitted it because my program 
was painfully slow. I guess either the D regex implementation is 
extremely slow or I'm not using it correctly.

/Oskar
December 12, 2005
Re: A safer/better C++?
In article <dnjmmu$14mp$1@digitaldaemon.com>, John C says...
>
>What about a place where all this expertise can be shared, where people can 
>publish articles, tips and code? I'm thinking of an MSDN for D, a D 
>Developers Network if you like. Community blogs are also pretty effective. 
>Artima (http://www.artima.com/) might be a good model to follow.
>

:)

Glad I'm not the only one thinking about this.

http://www.digitalmars.com/drn-bin/wwwnews?digitalmars.D.announce/2082

I've never seen artima, but I'll check it out.

>I'm not sure getting into CUJ is the best way - it's primarily a C++ 
>magazine, and while there are occasional series on other languages, how many 
>articles about D will its editors take?

You hit the nail right on the head.

I think D will always make great "guest appaearances" there but utlimately, its
not our scene.  D has its own mentality, momentum and community apart from C++
now, and it will continue to grow that way.  Better to build things up much like
the Python, PHP and web standards folks have rather than seek exposure in a
community that has too much invested in how things are done without D.

- EricAnderton at yahoo
December 12, 2005
Re: A safer/better C++?
In article <dnhvoi$dja$1@digitaldaemon.com>, Walter Bright says...
>
>Here's a lively debate over in comp.lang.c++.moderated people might be
>interested in:
>
>http://groups.google.com/group/comp.lang.c++.moderated/browse_thread/thread/60117e9c1cd1c510/c92f7fd0dc9fedd1?lnk=st&q=safer+better+c%2B%2B&rnum=1&hl=en
>#c92f7fd0dc9fedd1
>
>

I don't know whether or not this would have any bearing on that group or not,
but maybe you could mention your experiences implementing DMD script since this
is publically available, non-trivial code that they could look over?

On my machines, the D version is 33% (AMD) to 50% (P4) faster than the C++
version running the included sieve.ds benchmark, and I believe you said that it
was implemented with a 10% reduction in lines of code along with other
advantages.

BTW - Does the D version also build faster?

Thanks,

- Dave
December 12, 2005
Re: A safer/better C++?
"pragma" <pragma_member@pathlink.com> wrote in message 
news:dnk5q0$1ofo$1@digitaldaemon.com...
> In article <dnjmmu$14mp$1@digitaldaemon.com>, John C says...
>>
>>What about a place where all this expertise can be shared, where people 
>>can
>>publish articles, tips and code? I'm thinking of an MSDN for D, a D
>>Developers Network if you like. Community blogs are also pretty effective.
>>Artima (http://www.artima.com/) might be a good model to follow.
>>
>
> :)
>
> Glad I'm not the only one thinking about this.
>
> http://www.digitalmars.com/drn-bin/wwwnews?digitalmars.D.announce/2082

So I see. If I had some spare cash for a web host and domain, I'd get it off 
the ground.

John.

>
> I've never seen artima, but I'll check it out.
>
>>I'm not sure getting into CUJ is the best way - it's primarily a C++
>>magazine, and while there are occasional series on other languages, how 
>>many
>>articles about D will its editors take?
>
> You hit the nail right on the head.
>
> I think D will always make great "guest appaearances" there but 
> utlimately, its
> not our scene.  D has its own mentality, momentum and community apart from 
> C++
> now, and it will continue to grow that way.  Better to build things up 
> much like
> the Python, PHP and web standards folks have rather than seek exposure in 
> a
> community that has too much invested in how things are done without D.
>
> - EricAnderton at yahoo
December 12, 2005
Re: A safer/better C++?
Walter Bright wrote:

> There is some curious psychology going on with this. I've met many well
> known C++ experts who won't say anything positive about D in public, but in
> private to me they are quite enthusiastic about it. It's almost as if they
> fear censure or loss of face?
> 

Very interesting that you mention this... such "peer pressure" is 
noticeable in the scientific community as well.  Every so often some 
bright but hapless scientist presents research that counters prior 
evidence on a politically loaded issue: for example, anything that even 
hints of influencing opinion on ethics or morality or metaphysics; or 
anything that might remotely influence popularly accepted industry -- 
think pharmacology or immunology; or anything contrary to politically 
popular themes -- think "green house" gas emission sources.

No matter that his "unusual" results be right or true, he is derided 
severely by his peers who support the popular opinion, whatever that may 
be at the time (and oh how that changes).  Many of these, though few 
they be, lose jobs, face, and position in the scientific community 
merely for presenting objectively relevant material.  The scientific 
community is not always the objective giant that the general public is 
made to believe it to be.

Every get tired of reading the "scientists say..." quotes?

It's as if we are meant to believe there's always a wonderful consensus 
on such matters.

It's the same with D verses C++.  It's all about staying in the groove. 
  People ignore it for two major reasons: (1) fear of moving away from 
what's familiar and (2) fear of moving away from what's popular. They 
don't want to lose face in front of their peers.

Very few people have the guts to shake off the grip of "false", old ways 
in order to boldly proclaim the new and the true... all alone.

Cheers, Walter, for being willing to take lone way yourself.

-JJR
December 12, 2005
Re: A safer/better C++?
pragma wrote:
> In article <dnjmmu$14mp$1@digitaldaemon.com>, John C says...
>> What about a place where all this expertise can be shared, where people can 
>> publish articles, tips and code? I'm thinking of an MSDN for D, a D 
>> Developers Network if you like. Community blogs are also pretty effective. 
>> Artima (http://www.artima.com/) might be a good model to follow.
>>
> 
> :)
> 
> Glad I'm not the only one thinking about this.
> 
> http://www.digitalmars.com/drn-bin/wwwnews?digitalmars.D.announce/2082
> 
> I've never seen artima, but I'll check it out.
> 

I think that's the site that Matthew Wilson often referred to.  I think 
 it's fairly respected for it's high-quality technical articles/blogs 
on software development topics.

-JJR
1 2 3 4 5 6 7
Top | Discussion index | About this forum | D home