Joe Gottman
--
[ See http://www.gotw.ca/resources/clcm.htm for info about ]
[ comp.lang.c++.moderated. First time posters: Do this! ]
> When I tried to download the most recent draft of C++11
> (http://www.open-std.org/jtc1/sc22/wg21/prot/14882fdis/n3291.pdf) it
> demanded I enter a user name and password. I was able to download
> previous versions without a problem. What's going on?
Now the FDIS ballot has started, ISO rules require that the document is
not publicly available. Seems daft to me, since people could already
have downloaded it, but they'll be charging for the final standard, and
this is only editorially different, so it sort of makes sense.
Anthony
--
Author of C++ Concurrency in Action http://www.stdthread.co.uk/book/
just::thread C++0x thread library http://www.stdthread.co.uk
Just Software Solutions Ltd http://www.justsoftwaresolutions.co.uk
15 Carrallack Mews, St Just, Cornwall, TR19 7UL, UK. Company No. 5478976
> When I tried to download the most recent draft of C++11
> (http://www.open-std.org/jtc1/sc22/wg21/prot/14882fdis/n3291.pdf) it
> demanded I enter a user name and password. I was able to download
> previous versions without a problem. What's going on?
>
The most recent draft is identical to the FDIS, except for the title
page and the running heads. Since the FDIS will, in all probability,
become the standard, ISO doesn't allow the FDIS or the most recent
draft to be distributed for free.
--
Pete
Roundhouse Consulting, Ltd. (www.versatilecoding.com) Author of "The
Standard C++ Library Extensions: a Tutorial and Reference
(www.petebecker.com/tr1book)
This is very close, if not identical, to the standard that ISO and
ANSI intend to sell later. If everyone could download it for free,
that would obviously hurt the future sales.
Bo Persson
> When I tried to download the most recent draft of C++11
> (http://www.open-std.org/jtc1/sc22/wg21/prot/14882fdis/n3291.pdf) it
> demanded I enter a user name and password. I was able to download
> previous versions without a problem. What's going on?
I guess they're preparing for the final version, which will be sold by
ISO.
That latest version is the final draft IS as such it will come under ISO
rules with restricted circulation to those that require access 'for the
purpose of standardisation'. In other words it is effectively what they
intend to sell in about 3 months time. Up to this stage the document is
a consultative one which is why WG21 has got away with allowing public
access.
It has been explained to me that this restriction is in agreement with
ISO JTC 1 rules, but I'm unaware of any further details or the specific
reasons for this recent change, sorry.
HTH & Greetings from Bremen,
- Daniel Krügler
> The most recent draft is identical to the FDIS, except for the title
> page and the running heads. Since the FDIS will, in all probability,
> become the standard, ISO doesn't allow the FDIS or the most recent
> draft to be distributed for free.
ISO does allow free distribution of standards. It's just that it has
to be negotiated upfront. Now it's too late, of course.
On the other hand, almost everybody in the C++ community hopes that
the FDIS continues to leak to the Internet and is consulted by
programmers because everything else would have a disastrous impact on
adoption.
--
Is it me or it just does not make sense?
There seems to be so many things wrong with such a process that I
can't even begin enumerating. The worst certainly is that charging for
something that should be on the public domain simply does not sound
right. Yes, there are costs involved in the process and I would be
glad to pay for them if it were worth the money. But we don't usually
want pay for highly suboptimal stuff, do we?
Come on, they couldn't even include a feature as simple as static_if!
While there are some great ideas in the new standard, they all look
like half baked, even though it took 8 years to get ready. For
example, why must I use the pack operator only for the last arguments
in a variadic template? Dude! Initializer lists are so wrong! Perfect
forwarding is not perfect! A good interface for valarrays?
C++0x isn't even ready and there is so much already to be done. The
worst thing that ever happened to C++ was Stroustroup leaving it for
ISO to handle.
Elias.
[..]
> Come on, they couldn't even include a feature as simple as static_if!
I'm unaware of any official proposal of static_if. Maybe this proposal
had a different name, but I couldn't find any, therefore I would say
that this lack is a failure of those who wanted it. Participate in the
process of standardization and you can improve the standard. It is much
easier to criticize a state that has been baked by others.
> While there are some great ideas in the new standard, they all look
> like half baked, even though it took 8 years to get ready. For
> example, why must I use the pack operator only for the last arguments
> in a variadic template? Dude! Initializer lists are so wrong! Perfect
> forwarding is not perfect! A good interface for valarrays?
This does not sound like any reasonable form of criticism to me.
"Perfect forwarding" is not defined in the standard, so I don't see how
you can argue against the usage of a non-official term. I miss to see
any specific arguments for "Initializer lists are so wrong!" and
"A good interface for valarrays?", therefore these are buzz words to me.
If there are problems that you are aware of (not handled yet by existing
issues), why don't you submit a corresponding issue?
> C++0x isn't even ready and there is so much already to be done. The
> worst thing that ever happened to C++ was Stroustroup leaving it for
> ISO to handle.
A standard is never perfect and you really *need* at some point in time
complete (or near to complete) implementations to find more subtle
defects. An ISO standard has a very specific mechanism to handle defects
and problems, these are issues and technical corrigenda, so I cannot
agree with your conclusion. Based on your suggestion we should better
never publish a standard, because we will have a hard time that this
standard is perfect.
Greetings from Bremen,
- Daniel Krügler
--
> A standard is never perfect and you really *need* at some point in time
> complete (or near to complete) implementations to find more subtle
> defects. An ISO standard has a very specific mechanism to handle defects
> and problems, these are issues and technical corrigenda, so I cannot
> agree with your conclusion. Based on your suggestion we should better
> never publish a standard, because we will have a hard time that this
> standard is perfect.
>
Thanks for your excellent rebuttal. As I read your final paragraph 'The Art of Computer Programming' sprang to mind. I have often wondered about submitting that work to The Guinness Book of Records. Has any other book ever reached a 3rd edition of its first volume whilst the last volume is yet to be published? Has any other book had a volume consisting of two chapters (planned to be) published in three parts (i.e. sub-volumes)? Has any other book ever been published in draft sections which themselves consist of part of more than one chapter (indeed IIRC parts of sections of chapters)
What is the relevance? Well this is what happens when you decide that a text must be definitive and almost perfect before publication. If we took the same attitude to C++ we would never have a standard. Some might think that would be a good thing (TM) but most understand that having agreed fixed points even if imperfect and necessarily incomplete benefit the majority. How I wish that DK would actually finish The Art of Computer Programming (indeed had finished the first edition at least a decade ago).
The simplest way to get an entirely legal free copy of the C++ Standard is to become actively involved in work on Standardising C++. That entitles you to a copy from your NB. However most people find a Standard a difficult document, ill-suited to the normal working programmer. Of course authors of explanatory text books tend to expect to eat and so usually charge for them :)
Yeah it's counter-productive (particularly in this modern age), and
arguably even unethical, but I suppose it more or less worked OK back
in the day when only companies and academic institutions really cared.
Of course standards organizations are ollllld and still in many cases
creaking along with methods and practices from many ages
ago... Institutional inertia is a powerful force... :(
-Miles
--
Justice, n. A commodity which in a more or less adulterated condition the
State sells to the citizen as a reward for his allegiance, taxes and personal
service.
I certainly hope that they make it available for a reasonable fee. Some ISO standards cost $30, and some cost $300. If it ends up in the $300 category... then very few programmers will buy it.
Right now, there isn't any good free online reference documentation for the C++ standard. The SGI docs are the best, but they only partially cover the standard library. There's no iostreams documentation for instance. They also don't cover any of the new additions in C++0x.
I think there's an expectation among the standards committee that programmers will just buy a bunch of C++ books. However, today that's unrealistic.
First of all, C++ is a huge language, and you need maybe 5 or 6 books to get what's in the standard. That's a pretty big investment for a new C++ programmer. The main reason people learn C instead of C++ is that learning it is less of an investment in time and money.
Secondly, programmers no longer expect to have to buy books. Personally, when I learn a new language now, I expect there to be free online documentation. Python, Ruby, Java, C#, Haskell, Clojure, etc, all have free authoritative online references for their standard library and language.
I think that the "programmers will buy a bunch of books to learn a language" expectation comes from an era before the internet.
In short, what is the most recent draft available for downloading? is
it the n3242.pdf or a more recent version is also available ?
Thanks,
Nicola Bonelli
Maybe I should have proposed it, you are right.
> This does not sound like any reasonable form of criticism to me.
> "Perfect forwarding" is not defined in the standard, so I don't see how
> you can argue against the usage of a non-official term.
I've read an article
http://www.artima.com/cppsource/rvalue.html
which states it as a key motivation for rvalue references, but Scott
Meyers has already collected a few examples were perfect forwarding
fails miserably:
http://tinyurl.com/3rzwsao
> I miss to see
> any specific arguments for "Initializer lists are so wrong!" and
I've tried to discuss something in this forum, but got no responses:
http://tinyurl.com/6b6c2av
> "A good interface for valarrays?", therefore these are buzz words to me.
Josuttis heavily criticises the valarray interface in his "The C++
Standard Library - A Tutorial and Reference". It seems that the issue
has not, so far, been addressed. You probably know that valarrays are
fundamental to turn some optimizations on that are not possible due to
the well known aliasing problem.
>
> If there are problems that you are aware of (not handled yet by existing
> issues), why don't you submit a corresponding issue?
>
I would if it were under the direction Stroustrup or some "benevolent
dictator", but the mere idea of going through a bureaucratic committee
makes me itch. Sorry, I know you won't agree with such a posture.
>
> A standard is never perfect and you really *need* at some point in time
> complete (or near to complete) implementations to find more subtle
> defects.
Agreed.
> An ISO standard has a very specific mechanism to handle defects
> and problems, these are issues and technical corrigenda, so I cannot
> agree with your conclusion. Based on your suggestion we should better
> never publish a standard, because we will have a hard time that this
> standard is perfect.
It is not quite the mechanism that bothers me. It is the fact that
decisions seem not to be taken based on the scientific merit of the
proposals, but rather on political influences and the such. Of course
this may be only an impression, but is it?
Forget all the technical issues, and try explaining me the fact that,
as Weimer pointed out:
> On the other hand, almost everybody in the C++ community hopes that
> the FDIS continues to leak to the Internet and is consulted by
> programmers because everything else would have a disastrous impact on
> adoption.
So, they want to sell something expecting it to be leaked, because if
not there will be no adoption. Does it make sense?
Thank you for your time and attention,
Elias.
> Joe Gottman wrote:
>> When I tried to download the most recent draft of C++11
>> (http://www.open-std.org/jtc1/sc22/wg21/prot/14882fdis/n3291.pdf) it
>> demanded I enter a user name and password. I was able to download
>> previous versions without a problem. What's going on?
>>
>>
> This is very close, if not identical, to the standard that ISO and ANSI
> intend to sell later. If everyone could download it for free, that would
> obviously hurt the future sales.
But that doesn't make sense! I do have a copy of n3291.pdf downloaded on
April, 18th.
See also
http://groups.google.com/g/71a7ff08/t/766b6c40206ac6a0/d/3394b17661b12ef2
Helmut.
> Secondly, programmers no longer expect to have to buy books. Personally,
> when I learn a new language now, I expect there to be free online
> documentation. Python, Ruby, Java, C#, Haskell, Clojure, etc, all have
> free authoritative online references for their standard library and
> language.
And which of those languages did you learn from its 'Standard'? It is the authoritative references and tutorials that are useful. And learning a langauge as large as C++ involves learning a useful subset (and progressively extending that). The Standard is not really accessible as a vehicle for learning a useful subset of C++.
People's energies would be much better directed to encouraging gifted teachers to share their insights. Note that very few books on programming generate royalties in excess of $10000, yet they take much longer to write and hone than it takes to earn $50000 for consultancy or course presentation. (Perhaps that is an excuse for the way certain commercial authors continually plagiarise themselves by copying vast chunks of earlier books - warts errors and all - into their latest offering)
>
> I think that the "programmers will buy a bunch of books to learn a
> language" expectation comes from an era before the internet.
Anyone who expects to learn the whole of C++ ever has an unreasonable expectation. The real power of C++ is not in the Standard but the insightful way that talented individuals have used it.
I find it hard to grasp an economic model where talented people work for free. What does happen is that some talented people place some of their work online. This leads to them being employed in various ways. Even those who write for 'profit' largely earn their living through highly paid tutorial and consultancy work.
It was obviously a mistake to publish them initially. The documents
have since been moved to the protected committee-only part of the
site.
Those of us fast enough to take advantage of the mistake can enjoy
reading the documents now. Others will have to wait until they become
official (and probably with a price tag).
Bo Persson
The money doesn't go to the committee but to the standards
organizations.
The people doing the work have to pay as well, for being voting
members and allowed to do the work. I'm sure they are happy if as many
as possible get the results of their work, but for it to be an
official international standard we need the international standards
organization as well. And that has a cost.
Bo Persson
ISO has salaries to pay, buildings to rent, web servers to hire, etc.
Would you volunteer to pay higher taxes to pay for all this? Or do you
think that ISO should make its money by selling something?
If you live in the USA you could try to get Congress to give more money
to ANSI so online standards can go back to being low cost again. (But I
don't rate your chances very highly).
Or you could put the language under the control of a large company
willing to pay the costs. But what influence do you have over the next
issue of Java and C#?
John
--
John Harris
[... TC5 -- ALGOL ... TC6 -- COBOL ... ECMA proposed standards will be
made available without restriction to all interested parties as soon as
they have been approved ... ]
regards,
alexander.
> I certainly hope that they make it available for a reasonable
> fee. Some ISO standards cost $30, and some cost $300. If it ends up
> in the $300 category... then very few programmers will buy it.
ISO standards are priced roughly based on page count. This means that
the new standard will probably cost more than 500 USD if you buy it
from ISO.
Some national standardization organizations publish less expensive
editions of the same text, but it's not the ISO standard. There is a
print version from Wiley of the British standard, and ANSI published
comparatively cheap PDFs of the American standard. Most national
standardization organizations charge fees in excess of the ISO fees,
though. There is also no guarantee that anyone will offer the
standard add reduced prices. (AFAIK, nothing comparable to the Wiley
book was published for the original C++ standard.)
> Right now, there isn't any good free online reference documentation
> for the C++ standard. The SGI docs are the best, but they only
> partially cover the standard library. There's no iostreams
> documentation for instance. They also don't cover any of the new
> additions in C++0x.
Microsoft publishes extensive reference documentation, but it only
reflects C++ as far as it is implemented in their toolchain.
> I think there's an expectation among the standards committee that
> programmers will just buy a bunch of C++ books. However, today
> that's unrealistic.
If you need the standard (because you want to settle a dispute between
you and an implementation, for instance), some random C++ book just
won't do.
I think the important development is not so much the medium, but
increased interest in truly cross-platform development (across
operating systems, compiler versions and compiler vendors), so
fiddling with the code until it's accepted by your local compiler
isn't feasilble anymore for many of us.
> Thanks for your excellent rebuttal. As I read your final paragraph 'The Art of Computer Programming' sprang to mind. I have often wondered about submitting that work to The Guinness Book of Records. Has any other book ever reached a 3rd edition of its first volume whilst the last volume is yet to be published? Has any other book had a volume consisting of two chapters (planned to be) published in three parts (i.e. sub-volumes)? Has any other book ever been published in draft sections which themselves consist of part of more than one chapter (indeed IIRC parts of sections of chapters)
I would not argue against DK's intents, he is a true genius. In fact
TeX, even 40 years after its inception, is still a better suited
language than any other for what is its intended use.
You've chosen an awful example since TACP is about theory, not a
practical document. The TeXBook, on the other hand, is ready, nobody
has actually found any serious need for major revisions and all of the
recent developments (e.g. LuaTeX) are not following the archaic
methodology that the ISO committee does.
> What is the relevance?
I believe none, since TACP is mostly about theory. It is not about
describing a programming language. When Knuth wanted to create a
language he did so. It was not perfect from the beginning (neither it
is now), but has always been very good. Yes, it took quite a while to
reach its final form, but decisions were made and implemented in a far
more timely manner than ISO has been doing with C++.
> The simplest way to get an entirely legal free copy of the C++ Standard is to become actively involved in work on Standardising C++.
How is that simple? Imagine all programmers involved in the process.
Would it be practical?
> That entitles you to a copy from your NB. However most people find a Standard a difficult document, ill-suited to the normal working programmer. Of course authors of explanatory text books tend to expect to eat and so usually charge for them :)
So is this the logic? Deny distribution of the standard so text book
writers can get what to eat? In fact, as you just said, standardese is
not for everyone, so text book writers have no reason whatsoever to
worry about the standard being freely distributable.
I would say there is no reason why to charge for the distribution of
the standard (aren't there sponsors to the standardisation process?).
There is little arguing against this point, I haven't actually seen
any.
Elias.
P.S.: Sorry, it was not my intent to start a flame war in this
respectable forum. In fact, it is the absence of such discussions what
makes me interested in reading the posts here. So, no more answers
from me to this topic.
What matters to the working programmer is what the implementation(s) they use actually do. The Standard is most important to those who provide those implementations. For them the cost of a copy is a minuscule part of the cost of development.
The time for a programmer to become concerned with the specific requirements of the Standard is when the code they write fails to behave in the way they expect. There are several excellent resouces to tackle such issues.
First you can test your code with Comeau C++ to see if it conforms with the Standard. Whilst that product is not definitive it is near as we can get to a reference implementation (and those concerned with conformance to the C++ Standard really should invest in buying a copy)
Next you can consult here or in comp.std.c++ and get a considerable amount of cost free help. Anyone who has ever struggled with the actual Standard will know just how difficult and time-consuming it can be. Even the authors sometimes get surprised by the way different parts can interact.
I can see little relationship between adoption of features in C++ and the words in the Standard. Take the struggle C99 has had. Even with relatively cheap access to that Standard it was many years before implementors had elected to support more than a few parts of it. And even today many C programmers stick with the earlier version.
We can only use what implementers provide. At the moment C++ implementers seem to be highly motivated to produce close to fully conforming implementations. They work from the Standard and they are the people for whom it is most relevant.
Efforts by several people and organisations have drastically reduced the cost of buying a copy of the C++ Standard (on the earlier -- early 1990s business model NBs would be charging you well over $1000 for a copy)
Francis
Have you any idea how much it costs to participate in developing an ECMA
Standard? At least ISO allowed me to attend meetings as a member of my
NB delegation without charging me for it. ECMA Standards are usually
developed by a very small number of people, usually representing large
companies with vested interests.
>> The simplest way to get an entirely legal free copy of the C++ Standard
is to become actively involved in work on Standardising C++.
>
> How is that simple? Imagine all programmers involved in the process.
> Would it be practical?
It is as simple as asking a contact person from your NB; while I'm not
active in the ISO C++ process, but rather in SC29 (JPEG), I can ensure
you that getting into a committee is very easy indeed. However, it
requires patience and effort to follow the process and contribute to it.
The reason why these "archaic methods" take so long is because lots of
people with differing opinions meet, and have to come to an agreement.
That is, the proposals have to be written, read, checked, tested and
then edited. It is quite a different process from writing up a book or a
programming language of your own.
>> That entitles you to a copy from your NB. However most people find a
Standard a difficult document, ill-suited to the normal working programmer.
Of course authors of explanatory text books tend to expect to eat and so
usually charge for them :)
>
> So is this the logic? Deny distribution of the standard so text book
> writers can get what to eat?
No, the purpose of the standard is quite different from the purpose of a
text book. A standard is like a law-book: It's intended audience is that
of the professional reader, everything is spelled out in very specific
details, such that every detail is technically correct, and a correct
implementation can be derived from following the text. This does not
mean that the text is very pedagogic - it doesn't have to. It has to be
correct, and has to spell out all details. Nor does the text have to
tell you how to program or use the language - that's the job of a
textbook. Nor does the text tell you how a specific implementation
should look like - only which requirements it has to satisfy. This is
something very different from what a textbook has to do.
> In fact, as you just said, standardese is
> not for everyone, so text book writers have no reason whatsoever to
> worry about the standard being freely distributable.
A standard is not for the average person to read, indeed. It is a
toolkit for those who want to implement a standard-conforming compiler
(for ISO C++) or for those who want to implement a standard-conforming
codec (for JPEG or MPEG), or... If you want to learn about the language
features, then leave it to the members of the committee to write text
books, or to text book authors to buy the standard and write this up as
a book.
> I would say there is no reason why to charge for the distribution of
> the standard (aren't there sponsors to the standardisation process?).
Because ISO or rather the corresponding national organizations also have
to cover the cost, namely to pay for the secretary and the
administration. In Germany, for example, the DIN is an association (a
"club" if you want to say so), and while the technical experts are not
paid by the DIN, the DIN hosts a whole infrastructure that eases the
standarization process.
> There is little arguing against this point, I haven't actually seen
> any.
I guess the whole process is a bit hard to follow from outside, and
while probably not perfect, it is still a democratic process that tries
to find compromises between all those that want to contribute. If you're
not willing to contribute, you can hardly claim that your opinion didn't
count.
So long,
Thomas
John G Harris wrote:
[...]
> Or you could put the language under the control of a large company
> willing to pay the costs. But what influence do you have over the next
> issue of Java and C#?
Java aside, C# is Standard ECMA-334:
http://www.ecma-international.org/publications/standards/Ecma-334.htm
and it is available free of charge.
It is also an ISO standard and freely available just like a whole bunch
of other ISO standards:
http://standards.iso.org/ittf/PubliclyAvailableStandards/index.html
"Freely Available Standards
In accordance with ISO/IEC JTC 1 and the ISO and IEC Councils these
International Standards are publicly available.
ISO Copyright for the freely available standards
The following standards are made freely available for standardization
purposes. They are rotected by copyright and therefore and unless
otherwise specified, no part of these publications may be reproduced or
utilized in any form or by any means, electronic or mechanical,
including photocopying, microfilm, scanning, reproduction in whole or in
part to another Internet site, without permission in writing from ISO.
Requests should be addressed to the ISO Central Secretariat.
The documents you are about to download are a single-user, non-revisable
Adobe Acrobat PDF file, to store on your personal computer. You may
print out and retain one printed copy of the PDF file. This printed copy
is fully protected by national and international copyright laws, and may
not be photocopied or reproduced in any form. Under no circumstances may
it be resold."
regards,
alexander.
--
Francis Glassborow wrote:
>
> On 09/06/2011 00:24, Alexander Terekhov wrote:
> >
> > C++ should get rid of ISO and go to ECMA:
> >
> >
>
http://www.ecma-international.org/publications/files/ECMA-MEMENTOS/Ecma%20me
> mento%201962%20public.pdf
> >
> > [... TC5 -- ALGOL ... TC6 -- COBOL ... ECMA proposed standards will be
> > made available without restriction to all interested parties as soon as
> > they have been approved ... ]
>
> Have you any idea how much it costs to participate in developing an ECMA
> Standard? At least ISO allowed me to attend meetings as a member of my
Well,
http://www.ecma-international.org/activities/General/presentingecma.pdf
"Five membership categories, to reflect the diversity of stakeholders in
the ICT industry:
- Paying: Ordinary (OM), Associate (AM), Small and Medium-sized
Enterprise (SME), Small Private Company (SPC)
- No charge: Not-for-Profit (NFP)
Note: Members are organizations, not individuals"
See also:
http://www.ecma-international.org/memento/join.htm
regards,
alexander.
> John G Harris wrote:
> [...]
>> Or you could put the language under the control of a large company
>> willing to pay the costs. But what influence do you have over the next
>> issue of Java and C#?
>
> Java aside, C# is Standard ECMA-334:
>
> http://www.ecma-international.org/publications/standards/Ecma-334.htm
>
> and it is available free of charge.
Last time I looked, it would have cost me $15,000 per year to
participate in ECMA. For ISO, through INCITS, the fee is $1200. It's
very generous of you to suggest that I pay over ten times as much as I
currently do so that you can get a free copy of the standard.
--
Pete
Roundhouse Consulting, Ltd. (www.versatilecoding.com) Author of "The
Standard C++ Library Extensions: a Tutorial and Reference
(www.petebecker.com/tr1book)
Pete Becker wrote:
>
> On 2011-06-09 05:49:35 -1000, Alexander Terekhov said:
>
> > John G Harris wrote:
> > [...]
> >> Or you could put the language under the control of a large company
> >> willing to pay the costs. But what influence do you have over the next
> >> issue of Java and C#?
> >
> > Java aside, C# is Standard ECMA-334:
> >
> > http://www.ecma-international.org/publications/standards/Ecma-334.htm
> >
> > and it is available free of charge.
>
> Last time I looked, it would have cost me $15,000 per year to
> participate in ECMA. For ISO, through INCITS, the fee is $1200. It's
> very generous of you to suggest that I pay over ten times as much as I
You do know that not for profit orgs such as open source orgs, research
institutes, and government orgs don't pay ECMA membership fees (and
still can vote on technical committee level, but not on general assembly
meetings), right?
"British Library ... Mozilla Foundation ... Stanford University ... The
Library of Congress ... US Department of Defense"
http://www.ecma-international.org/memento/NFP.htm
"... next category is the small private company. This has a maximum of 5
employees and annual revenues of $5 million per year. Usually startups
or very small private companies. The membership fee is 3,500 swiss
francs per year. The next one is a better known term, the SME (small
medium enterprise). The only qualification is that it must have a
turnover (annual revenues) of $75 million. SMEs pay 17,500 swiss francs
per year in dues. There's no initiation fee. If you apply in a certain
six month period, invoicing starts at the beginning of the next
"semester." Semesters begin on January 1 and July 1. But you get
voting rights on a technical level as soon as you join, even though you
haven't paid yet. Ordinary members, the highest level pays 70,000
francs. Associate Members pay 35,000 francs per year. If you want to
be in the management of the association — in other words be involved at
both a technical and the management level — getting the rights to decide
on budget, who is president, etc., then you must apply for ordinary
membership. Theoretically, everyone can apply. but if you were a small
private company, I guess you wouldn't do that because it is too
expensive. If you want to have the rights to chair technical committees
and do not qualify for one of the first three categories, then you can
apply for an associate membership. "
http://www.zdnet.com/blog/btl/page/interview-with-ecmas-secretarie-general/2259
> currently do so that you can get a free copy of the standard.
regards,
alexander.
--