IPA (Information-Technology Promotion Agency, Japan) had invited
public participation in drafting of Ruby International Standard, and
my company, NaCl (Network Applied Communication Laboratory Ltd.), was
adopted.
IPA is an affiliated organization of the Ministry of Economy, Trade
and Industry of Japan. They supports development of open source
softwares, and have an interest in Ruby. Ruby is spreading quickly in
the world. However, there could be problems in some domains from lack
of officially documented language specification of Ruby, especially
among the public sectors. That's the reason why IPA launched this
drafting project.
We, NaCl have been sponsoring development of Ruby since 1997, but Ruby
has been, and will remain Matz's personal project. NaCl will not
claim any ownership nor control over the language. We believe that
community driven development is the best way to develop Ruby. So
we'll extract the language specification from implementations, and
just write it down. We don't want Ruby to be a straight-laced
language, so take care our draft (which should be the international
standard) not to prevent free development of Ruby.
The project is just started, and we're preparing a lot of works.
We'll inform you further information in the near future.
Is your extraction of the language specification from the various
implementations going to be separate from the RubySpec [ http://rubyspec.org/ ] effort? If so, why?
IPA (Information-Technology Promotion Agency, Japan) had invited
public participation in drafting of Ruby International Standard, and
my company, NaCl (Network Applied Communication Laboratory Ltd.), was
adopted.
The project is just started, and we're preparing a lot of works.
We'll inform you further information in the near future.
Is there anything we (the community) can contribute?
Is your extraction of the language specification from the various
implementations going to be separate from the RubySpec [ http://rubyspec.org/ ] effort? If so, why?
Same questions came to mind. IMHO, an executable spec is the
only way to go for a programming language.
Is your extraction of the language specification from the various
implementations going to be separate from the RubySpec [ http://rubyspec.org/ ] effort?
It's a separate work.
If so, why?
ISO or any standard organization would no accept RubySpec like
specification for various (and some good) reasons.
matz.
···
In message "Re: Drafting of Ruby International Standard" on Sat, 25 Oct 2008 23:27:52 +0900, "Craig Demyanovich" <cdemyanovich@gmail.com> writes:
>IPA(Information-Technology Promotion Agency, Japan) had invited
> public participation in drafting of Ruby International Standard, and
> my company, NaCl (Network Applied Communication Laboratory Ltd.), was
> adopted.
> The project is just started, and we're preparing a lot of works.
> We'll inform you further information in the near future.
Is there anything we (the community) can contribute?
Thank you.
We will prepare a public mailing list for this project, and notify
drafting
policy, progress, and drafts in process of creation. Any comments
will be
welcome.
>Is your extraction of the language specification from the various
>implementations going to be separate from the RubySpec [
>http://rubyspec.org/ ] effort?
It's a separate work.
>If so, why?
ISO or any standard organization would no accept RubySpec like
specification for various (and some good) reasons.
matz.
So, can we take this to mean that the intention is to eventually have an ISO Ruby standard (as opposed to say, Ecma)? Also, while I realize the logic behind not using a test-suite as a spec, couldn't the ruby-spec work be included as an official test-suite similar to how POSIX has the PCTS? Anyway, I'm glad to see work on this front!
-Josh
P.S. Should the ruby-design wiki be updated, or will this work be happening independent of the ruby-design work?
···
On Oct 25, 2008, at 7:17 PM, Yukihiro Matsumoto wrote:
In message "Re: Drafting of Ruby International Standard" > on Sat, 25 Oct 2008 23:27:52 +0900, "Craig Demyanovich" <cdemyanovich@gmail.com > > writes:
So, can we take this to mean that the intention is to eventually have
an ISO Ruby standard (as opposed to say, Ecma)?
Our goal is establish open standard for Ruby language. ISO Ruby
standard will satisfy our goal, but we don't deny other possibility
yet. But we don't think for Ecma, that tends to require more money
than we can afford.
Also, while I realize
the logic behind not using a test-suite as a spec, couldn't the ruby-
spec work be included as an official test-suite similar to how POSIX
has the PCTS? Anyway, I'm glad to see work on this front!
I am not sure yet. If RubySpec can be a part of the official spec,
that would be great.
P.S. Should the ruby-design wiki be updated, or will this work be
happening independent of the ruby-design work?
It's independent, although we've already discussed with other
developers. ruby-design wiki and developer meeting have not seen
progress for months due to our unfortunate discommunication.
matz.
···
In message "Re: Drafting of Ruby International Standard" on Mon, 27 Oct 2008 02:37:49 +0900, Joshua Ballanco <jballanc@gmail.com> writes:
>So, can we take this to mean that the intention is to eventually have
>an ISO Ruby standard (as opposed to say, Ecma)?
Our goal is establish open standard for Ruby language. ISO Ruby
standard will satisfy our goal, but we don't deny other possibility
yet. But we don't think for Ecma, that tends to require more money
than we can afford.
In addition, one of reasons why we need open standard for Ruby is
that the basic guideline for the government procurement in Japan
(http://www.soumu.go.jp/gyoukan/kanri/pdf/070301_1.pdf\) require it.
In that policy, we should refer to open standards instead of specific
products for fair competition.
>Also, while I realize
>the logic behind not using a test-suite as a spec, couldn't the ruby-
>spec work be included as an official test-suite similar to how POSIX
>has the PCTS? Anyway, I'm glad to see work on this front!
I am not sure yet. If RubySpec can be a part of the official spec,
that would be great.
Indeed, it would be great. I think RubySpec and our draft can be
complementary.