A friend of mine sent this article:

http://www.theregister.co.uk/2004/07/08/dtrace_user_take/

It's about a utility called DTrace, providing realtime kernel stats/monitoring for Sun Solaris.

The interesting part of the article is that the author states that the DTrace utility was written in D - and then also claims that the authors of DTrace take credit for inventing it!

I wonder which is the fallacy: is this really written in D?  Did the creators invent a different language and call it D?  Did the creators arrogantly and falsely claim they created D?  Or did the author of the article make a mistake?

Interesting...

Greg Vanore wrote:
> A friend of mine sent this article:
> 
> http://www.theregister.co.uk/2004/07/08/dtrace_user_take/
> 

DTrace author's blog: http://blogs.sun.com/roller/page/bmc (includes a link to his Usenix paper on DTrace)

DTrace home page: http://www.sun.com/bigadmin/content/dtrace

--Steve

Stephen Waits wrote:
> DTrace author's blog: http://blogs.sun.com/roller/page/bmc (includes a link to his Usenix paper on DTrace)

Further examination of the DTrace paper (http://www.sun.com/bigadmin/content/dtrace/dtrace_usenix.pdf) reveals that it is indeed another language named "D".

Somebody want to mail them and let them know that the name "D" is taken?  :)  Along those same lines, has Walter done anything to legally reserve rights on that name???


From section 5 (of above linked paper):

5 D Language
DTrace users can specify arbitrary predicates and actions
using the high-level D programming language. D
is a C-like language that supports all ANSI C operators
and allows access to the kernel’s native types and
global variables. D includes support for several kinds
of user-defined variables, including global, clause-local,
and thread-local variables and associative arrays. D programs
are compiled into DIF by a compiler implemented
in the DTrace library; the DIF is then bundled into an
in-memory object file representation and sent to the inkernel
DTrace framework for validation and probe enabling.
The dtrace(1M) command provides a generic
front-end to the D compiler and DTrace, but other layered
tools can be built on top of the compiler library as
well, such as the new implementation of lockstat(1M)
described earlier.
5.1 Program Structure
...

--Steve

Disappointing.  It would have been neat if they had used the real D. ;)

I looked at the syntax, and it's certainly far from our D.  Although they chose to keep printf as well.

"Stephen Waits" <steve@waits.net> wrote in message news:cckg4v$5h$1@digitaldaemon.com...
> Stephen Waits wrote:
> > DTrace author's blog: http://blogs.sun.com/roller/page/bmc (includes a link to his Usenix paper on DTrace)
>
> Further examination of the DTrace paper (http://www.sun.com/bigadmin/content/dtrace/dtrace_usenix.pdf) reveals that it is indeed another language named "D".
>
> Somebody want to mail them and let them know that the name "D" is taken?
>   :)  Along those same lines, has Walter done anything to legally
> reserve rights on that name???
>
>
>  From section 5 (of above linked paper):
>
> 5 D Language
> DTrace users can specify arbitrary predicates and actions
> using the high-level D programming language. D
> is a C-like language that supports all ANSI C operators
> and allows access to the kernel’s native types and
> global variables. D includes support for several kinds
> of user-defined variables, including global, clause-local,
> and thread-local variables and associative arrays. D programs
> are compiled into DIF by a compiler implemented
> in the DTrace library; the DIF is then bundled into an
> in-memory object file representation and sent to the inkernel
> DTrace framework for validation and probe enabling.
> The dtrace(1M) command provides a generic
> front-end to the D compiler and DTrace, but other layered
> tools can be built on top of the compiler library as
> well, such as the new implementation of lockstat(1M)
> described earlier.
> 5.1 Program Structure
> ...
>
> --Steve

D is the name of a Sun internal language. It was invented named after Walter's D.

"Greg Vanore" <Greg_member@pathlink.com> wrote in message news:cckf6h$307j$1@digitaldaemon.com...
> A friend of mine sent this article:
>
> http://www.theregister.co.uk/2004/07/08/dtrace_user_take/
>
> It's about a utility called DTrace, providing realtime kernel stats/monitoring for Sun Solaris.
>
> The interesting part of the article is that the author states that the DTrace utility was written in D - and then also claims that the authors of DTrace take credit for inventing it!
>
> I wonder which is the fallacy: is this really written in D?  Did the creators invent a different language and call it D?  Did the creators arrogantly and falsely claim they created D?  Or did the author of the article make a mistake?
>
> Interesting...
>
>

FYI:
==========================================================

From: Bryan Cantrill <bmc@zion.eng.sun.com>
To: steve@waits.net
Cc: bmc@eng.sun.com, mws@eng.sun.com, ahl@eng.sun.com
Date: Thu, 8 Jul 2004 16:26:08 -0700 (PDT)
Subject: Re: "D" language..

Hey Steve,

> It's been around for awhile: http://www.digitalmars.com/d/ In
> fact it's the first link returned when googling "d language".

When we started work on DTrace (in 2001), the D language wasn't around
sufficiently to show up on a Google search; it seems that the D's were
developed roughly in parallel.  We don't intend to change the name of our D, and we think that our D and Digital Mars's D are sufficiently different that no one will be confused.  And neither "D" can be trademarked, so there's no legal issue to speak of.

Hopefully this didn't cause too much confusion...

> Were you all aware of this, and, are you considering renaming your
> language so as not to be so confusing?
>
> Nice work on the tool BTW.

Thanks!  Have you used DTrace, or did you see us at USENIX last week?

       - Bryan

--------------------------------------------------------------------------
Bryan Cantrill, Solaris Kernel Development.       http://blogs.sun.com/bmc

D first appeared on slashdot in August, 2001, but had been started in December 1999.

"Stephen Waits" <steve@waits.net> wrote in message news:cckm40$88v$1@digitaldaemon.com...
>
> FYI: ==========================================================
>
> From: Bryan Cantrill <bmc@zion.eng.sun.com>
> To: steve@waits.net
> Cc: bmc@eng.sun.com, mws@eng.sun.com, ahl@eng.sun.com
> Date: Thu, 8 Jul 2004 16:26:08 -0700 (PDT)
> Subject: Re: "D" language..
>
> Hey Steve,
>
>  > It's been around for awhile: http://www.digitalmars.com/d/ In
>  > fact it's the first link returned when googling "d language".
>
> When we started work on DTrace (in 2001), the D language wasn't around sufficiently to show up on a Google search; it seems that the D's were developed roughly in parallel.  We don't intend to change the name of our D, and we think that our D and Digital Mars's D are sufficiently different that no one will be confused.  And neither "D" can be trademarked, so there's no legal issue to speak of.
>
> Hopefully this didn't cause too much confusion...
>
>  > Were you all aware of this, and, are you considering renaming your
>  > language so as not to be so confusing?
>  >
>  > Nice work on the tool BTW.
>
> Thanks!  Have you used DTrace, or did you see us at USENIX last week?
>
>         - Bryan
>
> --------------------------------------------------------------------------
> Bryan Cantrill, Solaris Kernel Development.       http://blogs.sun.com/bmc

Walter wrote:
> D first appeared on slashdot in August, 2001, but had been started in
> December 1999.

I figured as much; however, it's between you and them.  It probably doesn't bother you..  probably shouldn't anyway as your D has already become "D".

--Steve

Stephen Waits wrote:

> Stephen Waits wrote:
> 
>> DTrace author's blog: http://blogs.sun.com/roller/page/bmc (includes a link to his Usenix paper on DTrace)
> 
> 
> Further examination of the DTrace paper (http://www.sun.com/bigadmin/content/dtrace/dtrace_usenix.pdf) reveals that it is indeed another language named "D".
> 
> Somebody want to mail them and let them know that the name "D" is taken?  :)  Along those same lines, has Walter done anything to legally reserve rights on that name???

In the circles I more usually frequent, D is the language of True Relational Database Management Systems (TRDBMS), as espoused by C J Date and H Darwen since the mid-90's.  So, that makes three languages known as 'D' -- I would not be surprised to find there are others too.

-- 
Jonathan Leffler                   #include <disclaimer.h>
Email: jleffler@earthlink.net, jleffler@us.ibm.com
Guardian of DBD::Informix v2003.04 -- http://dbi.perl.org/

In article <cckg4v$5h$1@digitaldaemon.com>, Stephen Waits says...

>  :)  Along those same lines, has Walter done anything to legally
>reserve rights on that name???

You can't trademark a letter of the alphabet! If you could, Microsoft would have trademarked all twenty six letters decades ago, and would by now be charging everyone royalties for everything ever written in Latin script.

A couple of years back, ISO tried to charge royalites on ISO language and country codes. Since these are part of the XML and HTML spec, this would have affected pretty much the entire internet. They backed down due worldwide outrage.

Now, I have a patent somewhere for the wheel... <g>
Jill