The internetnews.com artical just went out on the ACM TechNews mailing list

Quote:

When Is a D Better Than C? When It's a Language
InternetNews.com (01/05/07) Patrizio, Andy

A newly developed free programming language aims to address the shortcomings of C/C++, though some question the likelihood of the language's success since it is not being released in conjunction with any other product or platform. The language, D, was developed by Walter Bright, who created the first C++ compiler, Zortech++. He has developed a compiler and standard libraries for both Windows and Linux for D. The Phobus standard library and the compiler front-end are open source, and a D compiler is included for the widely-used open source C compiler, GCC. D is able to produce compiled code without the need of a virtual machine. The language is also somewhat backwards compatible with C: It can be interfaced with any C API without the need for a call interface; but it adds Java and Microsoft C# functions such as garbage collection, an inline assembler, and Java-like single inheritance. In production since 2001, D has received considerable contribution from the Slashdot/open source developer community. Bright designed the language with the experience of C++ programmers in mind, not the sale of a product. He says, "The idea is to make programming in D the most productive possible. Quicker to learn, quicker to write code in, quicker to debug, and quicker to maintain." However, Forrester analyst Jeff Hammond questions the programming language's market potential. He says, "To make a technology viable, the technology has to be more than just good. You have to build a business model around it. What's the business model here?"

BCS wrote:
> 
> The internetnews.com artical just went out on the ACM TechNews mailing list
> 
> Quote:
> 
> When Is a D Better Than C? When It's a Language
> InternetNews.com (01/05/07) Patrizio, Andy
> 
> A newly developed free programming language aims to address the shortcomings of C/C++, though some question the likelihood of the language's success since it is not being released in conjunction with any other product or platform. The language, D, was developed by Walter Bright, who created the first C++ compiler, Zortech++. He has developed a compiler and standard libraries for both Windows and Linux for D. The Phobus standard library and the compiler front-end are open source, and a D compiler is included for the widely-used open source C compiler, GCC. D is able to produce compiled code without the need of a virtual machine. The language is also somewhat backwards compatible with C: It can be interfaced with any C API without the need for a call interface; but it adds Java and Microsoft C# functions such as garbage collection, an inline assembler, and Java-like single inheritance. In production since 2001, D has received considerable contribution from the Slashdot/open source developer community. Bright designed the language with the experience of C++ programmers in mind, not the sale of a product. He says, "The idea is to make programming in D the most productive possible. Quicker to learn, quicker to write code in, quicker to debug, and quicker to maintain." However, Forrester analyst Jeff Hammond questions the programming language's market potential. He says, "To make a technology viable, the technology has to be more than just good. You have to build a business model around it. What's the business model here?"
> 

Once again, Forrester shows its value as a contrary indicator...

There are quite a few successful languages out there that *don't* (or didn't) have a "business model" wrapped around it. Perl, Ruby, Python, C, C++ to name just a few. Those languages showed how they were useful to programmers and _then_ business models developed around the language, not the other way around.

As for the other languages that started with a business model... Well, the business model for Java ended up changing (set-top devices to server development), not Java itself. C# took a few years itself to become really popular, and that was after massive spending and influence by MS, and also basically because the alternative (VB.Net) wasn't palatable by many.

Dave wrote:
> BCS wrote:
>>
>> The internetnews.com artical just went out on the ACM TechNews mailing list
>> ...
>> However, Forrester analyst Jeff Hammond questions the programming language's market potential. He says, "To make a technology viable, the technology has to be more than just good. You have to build a business model around it. What's the business model here?"
>>
> 
> Once again, Forrester shows its value as a contrary indicator...
> 
> There are quite a few successful languages out there that *don't* (or didn't) have a "business model" wrapped around it. Perl, Ruby, Python, C, C++ to name just a few. Those languages showed how they were useful to programmers and _then_ business models developed around the language, not the other way around.
> 
> As for the other languages that started with a business model... Well, the business model for Java ended up changing (set-top devices to server development), not Java itself. C# took a few years itself to become really popular, and that was after massive spending and influence by MS, and also basically because the alternative (VB.Net) wasn't palatable by many.

To me it's a pretty encouraging sign that D is on the move if such analysts are even commenting (or being asked to comment) on the language and its potential.

--bb

On Mon, 08 Jan 2007 21:07:44 +0000, BCS wrote:

> When Is a D Better Than C? When It's a Language
> InternetNews.com (01/05/07) Patrizio, Andy
> 
> A newly developed free programming language aims to address the shortcomings of C/C++, though some question the likelihood of the language's success since it is not being released in conjunction with any other product or platform. The language, D, was developed by Walter Bright, who created the first C++ compiler, Zortech++. He has developed a compiler and standard libraries for both Windows and Linux for D. The Phobus standard library and the compiler front-end are open source, and a D compiler is included for the widely-used open source C compiler, GCC. D is able to produce compiled code without the need of a virtual machine. The language is also somewhat backwards compatible with C: It can be interfaced with any C API without the need for a call interface; but it adds Java and Microsoft C# functions such as garbage collection, an inline assembler, and Java-like single inheritance. In production since 2001, D has received considerable contribution from the Slashdot/open source developer community. Bright designed the language with the experience of C++ programmers in mind, not the sale of a product. He says, "The idea is to make programming in D the most productive possible. Quicker to learn, quicker to write code in, quicker to debug, and quicker to maintain." However, Forrester analyst Jeff Hammond questions the programming language's market potential. He says, "To make a technology viable, the technology has to be more than just good. You have to build a business model around it. What's the business model here?"


This write-up was really poorly done, in my opinion.  I don't think it's an accurate representation of D, in general.

Also, "Phobus" is spelled wrong.  Usually being petty about spelling is not my way... but if this is somehow an official quotation from the ACM mailing list, I worry about the reception it will get.

And "it adds Java and Microsoft C# functions such as garbage collection"?
What??  That just sounds wrong: since when do those languages
represent the origination of D's gc? The whole announcement sounds wrong in
much the same way.

-JJR

Dave wrote:
> 
> As for the other languages that started with a business model... Well, the business model for Java ended up changing (set-top devices to server development), not Java itself. C# took a few years itself to become really popular, and that was after massive spending and influence by MS, and also basically because the alternative (VB.Net) wasn't palatable by many.

And for the first few years, the market was totally confused about whether .NET was a library/VM or whether it was a server object architecture.  While tying a business model to a language may garner a lot of initial attention if the business model can be popularized, it doesn't provide for any long-term guarantees.  Look at Java applets, for example.  They are a horribly failed business model, and Java has succeeded almost despite this.

Sean

John Reimer wrote:
> And "it adds Java and Microsoft C# functions such as garbage collection"?
> What??  That just sounds wrong: since when do those languages
> represent the origination of D's gc?

Or the origins of aany GC.  Lisp had garbage collection in the 60's.

BA

BCS wrote:
> 
> The internetnews.com artical just went out on the ACM TechNews mailing list

Awesome.  D's on a roll lately.

> What's the business model here?"

"Build it, and they will come."

Seriously, it's what the big guys did. :)

If a technology is good enough for everyone to use, and is what they've been looking for, then they'll likely use it - provided they can find out about it.  Thankfully, it's still easy to be heard on the Internet without having to shell out a megabuck, provided you have something halfway provocative or worthwhile to say.  Being able to say "It's c++, only readable and with a GC" or "Java without a VM" is more than enough to get people to read, even if they're only interested in debunking the claim.  After all, whiz-bang advertising is only good for getting attention for something people would tend to ignore in the first place.

Also, a language is just a tool and not the kind you can charge* for anyway.  Look at MS and Sun.  They put gobs of cash into their respective VM-based platforms and they still can't capitalize on them directly; they sell developer suites and training instead.

Walter has done nothing different - he built a better tool for next to nothing from which other, profitable, opportunities will arise.  The same could be said for any one of us with a D library or utility that plans on using that to provide a profitable service of some kind. Ultimately, the only major difference is one of scale - it's still the same model.

(* encumbering developer software, or even just proprietary language specifications, with expensive and/or restrictive licenses, tends to turn people off real quick.)

-- 
- EricAnderton at yahoo

On Mon, 08 Jan 2007 17:12:16 -0500, Brad Anderson wrote:

> John Reimer wrote:
>> And "it adds Java and Microsoft C# functions such as garbage collection"?
>> What??  That just sounds wrong: since when do those languages
>> represent the origination of D's gc?
> 
> Or the origins of aany GC.  Lisp had garbage collection in the 60's.
> 
> BA


Yes, that's what I meant also.  gc's have been around for a very long time.  Smalltalk is another example.


-JJR

> However, Forrester analyst Jeff Hammond questions

Let's put it this way: how many languages would kill to be even puked-upon by "a Forrester Analyst"???? And we're there already.

I couldn't care less about his opinion, it's the fact that such a guy gets quoted (or should I say caught) studying D, that's what really is significant.

Woo, the juggernaut has commenced the initial firing-up sequence. Fire and brimstone seem increasingly inevitable, the devotees have damp palms, the pilgrims are getting palpitations. We truly are approaching the day of the Benevolent Dragon, with a capital D!

D-men!