Kuro5hin.org: technology and culture, from the trenches
create account | help/FAQ | contact | links | search | IRC | site news
[ Everything | Diaries | Technology | Science | Culture | Politics | Media | News | Internet | Op-Ed | Fiction | Meta | MLP ]
We need your support: buy an ad | premium membership

[P]
Open Standards Update

By codemonkey_uk in News
Thu Apr 11, 2002 at 02:28:34 PM EST
Tags: Technology (all tags)
Technology

Programming language standards act as contract between the programmer and the compiler & library vendor. If you write code, either professionally, or as a hobby, you probably want it to conform to an open standard.

Read on for a selection of small (mostly good) news items concerning various open programming standards ...


  • C, C++ & C#

    A reasonably priced hardback edition of the BSI/ANSI C99 Standard is to be published by John Wiley & Sons within the next few months, thanks to a licence deal brokered by the ACCU. A companion C++ volume is also planned once the corrections for the 98 standard have been voted on.

    C++ guru Herb Sutter has accepted a job with Microsoft, and will be filling the role of "C++ Community Liaison". This along with Stan Lippman's recent move from the C# group to the C++ group seems like strong indication that Microsoft's attitude towards Standard C++ has shifted, and hope for a standard conforming VC++ is returning.

    Meanwhile the C# and CLI standards submitted to ECMA sponsored by Microsoft, along with Hewlett-Packard and Intel, have been ratified. Work continues on the standard, and it is now in the hands of ISO. If you want to contribute to the ISO standardisation process for C# and CLI, please email: standards@accu.org.

  • Posix / Unix

    ISO/IEC SC22 working group 3, IEEE, and the Open Group have announced Version 3 of the Posix system specifications, which and is available as HTML free online, or can be ordered on CD-ROM.

  • Java

    The Java Executable File Format (JEFF) specification has been accepted by ISO. This provides a ready to execute format for object orientated programs which drastically reduces runtime requirements, and halves the usual class file format without compression, making deployment on small devices such as cellphones more feasible. For more details visit the J Consortium website.

  • BSI & ISO

    Finally, BSI and ISO are both experimenting with selling copies of the standards under there control at "lower prices" in order to see if it generate more sales. While the sentiment is right, in the programmers opinion the prices are still way to high.

Sponsors

Voxel dot net
o Managed Hosting
o VoxCAST Content Delivery
o Raw Infrastructure

Login

Poll
Open Standards
o Important, worth paying for 21%
o Important, but should be free 76%
o Irrelevent 1%

Votes: 51
Results | Other Polls

Related Links
o hardback edition of the BSI/ANSI C99 Standard
o John Wiley & Sons
o ACCU
o Herb Sutter
o Stan Lippman
o C#
o CLI
o standards@ accu.org
o Version 3 of the Posix system specifications
o the J Consortium website
o BSI
o ISO
o Also by codemonkey_uk


Display: Sort:
Open Standards Update | 37 comments (15 topical, 22 editorial, 0 hidden)
Dammit (3.25 / 4) (#1)
by tombuck on Thu Apr 11, 2002 at 12:31:02 PM EST

One fine day I shall learn C++ properly. I have the book, but not the time. Damn this alcoholism.

On the other hand, it looks like I'm going to be put on C# at work which I'm really not sure if that's a good thing or not...

--
Give me yer cash!

Don't. (2.50 / 2) (#16)
by porkchop_d_clown on Thu Apr 11, 2002 at 01:47:14 PM EST

If you want to learn object oriented programming, pick up Java or Smalltalk or even Objective C first.

If you want to go back and learn a sort-of oo language later, go ahead and learn C++.


--
I'm not a sexist pig!
I'm a plain-old-everyday pig!


[ Parent ]
Did he mention OOP? (none / 0) (#37)
by codemonkey_uk on Mon Apr 15, 2002 at 07:10:01 AM EST

No, he did not, so why do you assume he did? Perhaps he already knows Java et al. Or perhaps he wants to do some template meta-programming.

C++ is a widely used, high performace multi paradigm language. The same cannot be said for any of the other languages you suggest. There are lots of reasons he might want to learn it.

Check your assumptions.
---
Thad
"The most savage controversies are those about matters as to which there is no good evidence either way." - Bertrand Russell
[ Parent ]

Contract? (3.40 / 5) (#3)
by prometheus on Thu Apr 11, 2002 at 12:34:39 PM EST

More like a verbal agreement which no one ever remembers about until they're taking the stand in a messy divorce trial involving incest, sex with farm animals, and the wife's father's brother's daughter's son's former male transvestite's uncle's pedophilia collection.
--
<omnifarad> We've got a guy killing people in DC without regard for his astro van's horrible fuel economy
Yay! (3.33 / 3) (#7)
by seebs on Thu Apr 11, 2002 at 12:40:16 PM EST

The best thing is, no reference to Herb Schildt's annotations for this one. Much hilarity has been had about the annotations he did to C89.

I might even buy a hardcover C99, although I've been well served by my PDF copy. (No, I didn't pay for it. No, it's not illegal. I paid my dues, quite literally, and I'm *allowed*.)

Java (3.50 / 2) (#18)
by bosk on Thu Apr 11, 2002 at 02:08:04 PM EST

So while the file format for java bytecode is being standardized the language still remains under the control of Sun. I did receive something in my email the other day though from Apache that Sun has recently allowed open source implementations of Java specifications. A step in the right direction but a fully standardized, open source version of Java the language and Java the VM is still a long way off.

Sun vs. MS (5.00 / 1) (#36)
by cpfeifer on Sat Apr 13, 2002 at 05:41:03 PM EST

A step in the right direction but a fully standardized, open source version of Java the language and Java the VM is still a long way off.

I agree. I was severly disappointed when I first heard the Sun had pulled the Java language from consideration of standardization.

From what I've seen of the various standards bodies(OMG, W3C...), I believe the optimal approach to standardization is:

  • standards bodies produce standards documents that describe the standard (the structure and format of documents (e.g. XML standards), the functional requirements & basic usage flows (e.g. CORBA standards)
  • open source projects produce the "reference implementation" of aech standard, as well as a test suite to judge standard compliance
  • Commercial ISVs (indenpendant software vendors)produce standard-compliant, value-added products. Commercial ISVs can differntiate their product through enhanced fault-tolerance, scalability, and additional platform support (for starters)
Through this process, everyone wins:
  • standards are not vendor controlled, they are controlled by standards corsotiums (which are made up of vendor representatives, but the hope is that there are enough voices that one doesn't dominate)
  • developers have free products (open source) to start developing solutions with. This accelerates the adoption of the standard and gets the technology in the hands of it's user base quickly.
  • once developers build their solutions, they can assess their non-functional requirements (the 'ilities': scalability, reliability...). If the open source products do not fulfill these requirements, they can look to the commercial marketplace for a vendor implementation (that is compliant with the standard) that does.
This pattern has emerged by watching Apache's Jakarta project. IMHO, the Tomcat servlet engine is a large contributing factor for the widespread adoption of Servlets in the development community, not to mention XML, JSP...


--- "What's the point of waking up in the morning if you don't try to match the enourmousness of the known forces in the world with something powerful in your own life?" - Don Delillo, "Underworld"
[ Parent ]

are people actually using/complying with C99? (3.50 / 2) (#23)
by aphrael on Thu Apr 11, 2002 at 02:53:44 PM EST

from what i've been able to tell from marketing surveys, etc, the new features in C99 (as opposed to C92) aren't being used, and none of the major commercial compilers have implemented them.

They can't use them (5.00 / 1) (#30)
by pfaffben on Fri Apr 12, 2002 at 02:23:37 AM EST

because the vendors haven't implemented them yet. On the other hand, the vendors haven't implemented them because there's not a great demand for them (yet).

I suspect that GCC's increasing C99 compliance will act as a push in the right direction.

[ Parent ]

Yeah. (3.00 / 1) (#35)
by aphrael on Fri Apr 12, 2002 at 02:45:06 PM EST

I work for a compiler vendor, so i'm familiar with this phenomenon. I was mainly enquiring to see if anyone had any strong evidence that people were demanding the c99 features, because then i could use that evidence to browbeat people. :)

[ Parent ]
Open Protocols (3.66 / 3) (#27)
by Adam Theo on Thu Apr 11, 2002 at 04:01:02 PM EST

Yes, open standards for programming are very good, and also (thankfully) very popular. I just wish everyone was so concerned about Open Protocols, especially Instant Messaging. It's an up-hill battle to convince people to build off of the completely open and free Jabber protocols instead of the popular yet proprietary .NET Messenger or AIM platforms. The reason so many people stick with the proprietary platforms, in my experience, is because they are popular and "everybody uses them". Why isn't this the case with programming languages so much? or is it just because most popular programming languages are already open that it's a non-issue?

-- "A computer geek free-market socialist patriotic American Buddhist."

simple difference (4.00 / 2) (#28)
by jbridge21 on Thu Apr 11, 2002 at 04:17:30 PM EST

A programming language only has to be the same within a project; once the program is programmed, it can be used by anyone (well, if it's made portable, or the userbase on the one platform is large enough). Furthermore, one person can use two different programs written in two different languages with no problems.

With the current methods of instant messenging, though, two users must use the same "standard", or else the interoperability is exactly zero. It is this fundamental difference between programming languages and instant messenging protocols that encourages grouping behavior in the one and not the other.

Of course, the way Jabber is built, once everyone is using it, you are free to use a different server that is perhaps written in a different programming language, and you still have access to talk to everyone else who uses Jabber.

[ Parent ]
K & R (3.00 / 1) (#31)
by S1ack3rThanThou on Fri Apr 12, 2002 at 03:51:22 AM EST

Still the bible, still all ya need... POSIX, now THERE's the crucial thing.

"Remember what the dormouse said, feed your head..."
Barrier to entry (5.00 / 1) (#33)
by Maniac on Fri Apr 12, 2002 at 10:06:02 AM EST

I recall a presentation several years ago that indicated that large companies like DEC, IBM, AT&T and GM would typically have 1000 to 2000 people working on standards at any given time and that "standards" was a 2 billion dollar a year industry. With that kind of background, the reluctance of standards organizations to make copies "free" or at low cost becomes obvious. I look at standards as another barrier to entry. Not a particularly high one - any real company would be paying the bills - I know mine does (it subscribes to several services and actively participates in several standards bodies).

Most companies consider the whole standards process another part of their competitive advantage. Get an industry group to adopt some method you have already implemented & get a head start on time to market.

There are exceptions as you mentioned, but here are a few more...

  • Military standards (though many US ones are no longer in use)
  • The whole suite of Internet Standards and the RFC process.
  • The X consortium has made all its standards available here.
  • The Ada language standard and rationale (and several other Ada related items).
As well as many more several of us could list. I don't expect this behavior to change much. The standards that are rapidly changing will cost plenty of money to acquire (and to influence). The ones that are stable & widely used will be free. Just another barrier to entry to be at the leading edge of development.

two things (3.00 / 1) (#34)
by mikpos on Fri Apr 12, 2002 at 10:12:06 AM EST

The C99 standard has, I think, widened the division between C and C++. Bjarne's elusive upcoming standard will probably make this even worse. The days of converting a C program to a C++ program with only a few touch-ups here and there are probably coming to an end.

And why so skimpy on the Unix standards? SingleUnix? Hello?

Open Standards Update | 37 comments (15 topical, 22 editorial, 0 hidden)
Display: Sort:

kuro5hin.org

[XML]
All trademarks and copyrights on this page are owned by their respective companies. The Rest 2000 - Present Kuro5hin.org Inc.
See our legalese page for copyright policies. Please also read our Privacy Policy.
Kuro5hin.org is powered by Free Software, including Apache, Perl, and Linux, The Scoop Engine that runs this site is freely available, under the terms of the GPL.
Need some help? Email help@kuro5hin.org.
My heart's the long stairs.

Powered by Scoop create account | help/FAQ | mission | links | search | IRC | YOU choose the stories!