carl@malamud.com (Carl Malamud) (01/23/91)
> From: Einar Stefferud <uunet!ics.uci.edu!Stef> > Date: Mon, 21 Jan 91 20:17:57 -0800 > Message-Id: <18135.664517877@nma> > > > I think it is entirely clear at this point how INTERNET folk feel aobut > the high cost of paper only ISO/CCITT/ETC standards, adn the desire to > see a change in the method of publication. > > I think that the real question now is more of "What is a way to work out > an alternative that meets what we clearly percieve as a critical need?" > > I don't think that continued simple lamenting will get us what we want. Stef - My original query to this list was not totally benign. There will be a "standards summit" in March between ISO and CCITT. One of the items on the agenda is this very topic. My purpose in asking if anybody else felt strongly about the issue was to send a signal to those folks that people really do care. That signal was received in no uncertain terms. There is another alternative to waiting for the standards bodies. As you may know, most standards appear in public domain form before they are "published" and copyright is asserted. Most legal scholars I've spoken to feel that an assertion of copyright is not justified and that even the FINAL documents are still in the public domain. A little OCR, a scanner, and somebody with a law degree are all that's needed to test this theory. Carl carl@malamud.com
Stef@ICS.UCI.EDU (Einar Stefferud) (01/24/91)
OK -- We agree that the time is here to come up with constructive proposals. I know that the ISO/CCITT community has been considering ways to improve the publication accessibility and provide electronic publication, but I am not aware of any of the issues hey are trying to resolve. Do you suppose we could get a factual report on what progress they have been achieving and what issues they are addressing? I don't see a lot of reasons to plow the ground that they have already plowed. I also must agree with whoever it was that pointed out the an extra $200 to $1000 for Standards Documents is not really significant compared to the cost of hiring two good programmers to implement without the standards in hand! Not smart economics there I fear. I recall the famous quote from Commodore Vanderbilt when asked by a plebian what his yacht cost. "If you have to ask, you can't afford it!" Best...\Stef
enag@ifi.uio.no (Erik Naggum) (01/25/91)
In article <21575.664711596@nma> Stef@ICS.UCI.EDU (Einar Stefferud) writes:
I also must agree with whoever it was that pointed out the an extra
$200 to $1000 for Standards Documents is not really significant
compared to the cost of hiring two good programmers to implement
without the standards in hand! Not smart economics there I fear.
I object to this arrogant stance. There are those among the
readership of this list and of the ISO standards who wish to be able
to follow the state of the art, to be able to know of what people
speak when in learned circles, and who would appreciate not being
treated as Palestinians because they don't intend to become the white-
robed Central Computer Attendants of times past.
I recall the famous quote from Commodore Vanderbilt when asked by a
plebian what his yacht cost. "If you have to ask, you can't afford
it!"
I'm sorry to see that OSI has become a luxury, unlike the prospect
that many had of it becoming about as ubiquitous as electricity and
telephones.
OK -- We agree that the time is here to come up with constructive
proposals. I know that the ISO/CCITT community has been considering
ways to improve the publication accessibility and provide electronic
publication, but I am not aware of any of the issues hey are trying to
resolve. Do you suppose we could get a factual report on what
progress they have been achieving and what issues they are addressing?
I don't see a lot of reasons to plow the ground that they have already
plowed.
I seem to remember having heard that ISO standards are stored and
coded using SGML. Contrary to popular belief, SGML documents should
normally be easy to read (with markup intact), and I would think that
SGML applications would soon emerge if these documents were released
in electronic form, thus achieving two goals with one move.
Feel free to think that this fails to meet your standard of "con-
structive proposal". I'm but a plebian in this yacht-owner's club.
--
[Erik Naggum] Snail: Naggum Software / BOX 1570 VIKA / 0118 OSLO / NORWAY
Mail: <erik@naggum.uu.no>, <enag@ifi.uio.no>
My opinions. Wail: +47-2-836-863 Another int'l standards dude.BILLW@MATHOM.CISCO.COM (William "Chops" Westfield) (01/25/91)
I made this point on tcp-ip, and I guess it needs to be made here too. A lot of work in the tcp world, both in terms of defining protocols, and in creating implementations, has been done by students, hackers, systems programmers, hobbiests, and other types of "unfunded researchers". While $1000 might not be significant to a company actually intending to eventually sell a product, it is certainly enough, together with the difficulty and time lag inherent in obtaining the documents, to stop much work before it has even had a chance to begin. Bill Westfield cisco Systems. -------
my@dtg.nsc.com (Michael Yip) (01/26/91)
In article (Erik Naggum) writes: >In article (Einar Stefferud) writes: > I also must agree with whoever it was that pointed out the an extra > $200 to $1000 for Standards Documents is not really significant > compared to the cost of hiring two good programmers to implement > without the standards in hand! Not smart economics there I fear. > >I object to this arrogant stance. There are those among the >readership of this list and of the ISO standards who wish to be able >to follow the state of the art, to be able to know of what people ... > I recall the famous quote from Commodore Vanderbilt when asked by a > plebian what his yacht cost. "If you have to ask, you can't afford > it!" > >I'm sorry to see that OSI has become a luxury, unlike the prospect ... I totally agree with Mr. Erik Naggum. I think that the Standards Documents should be available on line, may be for a small fee of maintance, just like the RFC documents for the Internet stuffs. $200 to $1000 is definitely not a significant cost for any company but there are many many people who works on their own and also many universities what are tight in budget. For example, the code like routed was first developed in UCB and being copied/referenced by many companies. That helps! But if the cost of development is so high that universities and individuals will not be able to develop public domain software, then I think that everyone and every companies will suffer. By the way, I would like to have ISO standards not just for some big and elite companies but for everyone. Every programmers, engineers and every users. I think that a very small transaction or maintanence fee is very reasonable. But with an electronic distribution, the cost will be a little bit lower. Just my 0.02 buck! -- Mike
Stef@ICS.UCI.EDU (Einar Stefferud) (01/26/91)
Hello Bill (and all) -- I have to agree that the cost and difficulty inherent with access to the ISO/CITT standards does not facilitate or foster grad student or academic (or itinerant hacker) interest in working with these standards. One proposal would be for the standards bodies to make a special case for academic use, and I hope that this proposal will be put before the standards bodies by whoever will capture the ideas from this list for contribution to the standards bodies. (Not me!) However, the observation remains valid that is is not good economics to hire good programmers at current programmer salaries (plus overhead) and then deprive them of the documents because they cost $200-$1000! Even for a two person company. [Lets see, where else can we save $200-$1000 by eliminating something?] I am sorry if some people are offended by my comments. I don't mean to be facetious here, but I do want to focus in on the central issues. In summary, in some cases the price is too high, while in others the price is not too high. So, we need to do something about the cases where the price is too high. The two classical things to do are to (1) arrange for special focused discounts or (2) subsidize in some way. I DO NOT MINIMIZE THE CONCLUSION THAT SOMETHING SHOULD BE DONE. Cheers...\Stef
OSmith@acorn.co.uk (Owen Smith) (01/28/91)
In article (Einar Stefferud) writes: > I also must agree with whoever it was that pointed out the an extra > $200 to $1000 for Standards Documents is not really significant > compared to the cost of hiring two good programmers to implement > without the standards in hand! Not smart economics there I fear. I used to work with the ISO standards. Having changed job I don't now. However, having become an expert in certain ISO areas I wish to stay an expert, and the only way I can do that is by reading the new specs as they come out. And unless they are free I am not going to be able to do that. It is in the interests of the promotion of ISO standards for people like me to be able to keep their knowledge up to date, so that the pool of ISO experts is larger. Owen (ASN.1, X.411 and X.420 1984 and 1988 expert). The views expressed are my own and are not necessarily those of Acorn.
klee@wsl.dec.com (Ken Lee) (01/29/91)
While I agree with many of the postings here that free access to ISO standards would be nice, you have to remember that developing and administering standards is a very expensive proposition. Most of the costs are born by the volunteer technical experts who write the standards, but there are many other expenses, including the costs associated with open public reviews, administrative overhead, etc. ISO, ANSI, and other standards groups generally receive no government funding, so must rely on some sort of fund raising. Note that standards bodies are not the only people who sell documents. Many technical societies raise a large portion of their budget by selling conference proceedings and journals that were written by volunteer technical experts. Many technical professionals will claim that this information is just as critical to their work as are standards documents. -- Ken Lee DEC Western Software Laboratory, Palo Alto, Calif. Internet: klee@wsl.dec.com uucp: uunet!decwrl!klee
rossj@cognos.UUCP (Ross Judson) (01/29/91)
It is interesting that you mention the difficulties that students have in finding these documents. I am currently doing a survey of compression techniques and would like to include the MNP 5 and V.32bis methods in my report. Does anyone have document names or pointers for me? There is a fairly large technical library (CISTI Canada) I have access to, but I'm not too sure of where to start. There are so many standards... -- Ross Judson ++ uunet!mitel!cunews!cognos!rossj + end of the road i'm travelling Cognos Inc. ++ rossj@cognos.uucp + i will see jordan beckoning...
PWW@BNR.CA (Peter Whittaker, P.W.) (01/29/91)
> Note that standards bodies are not the only people who sell documents. > Many technical societies raise a large portion of their budget by > selling conference proceedings and journals that were written by > volunteer technical experts. Many technical professionals will claim > that this information is just as critical to their work as are standards > documents. > > -- > Ken Lee The IEEE does this: conference proceedings are sold to raise funds, but members get a discount (usually 50%). Memebership fees vary, but there are several types, including student fees. Perhaps the ISO and CCITT should institute some similar scheme, where members receive discounts, and where stduent memberships are cheap enough to allow students to buy the standards. Other interested parties would foot the full bill. (If you were interested, why wouldn't you join?). Perhaps some organization like the IEEE should take it upon itself to make such a deal with ISO/CCITT and start offering discount blue books to its membership. Peter Whittaker [~~~~~~~~~~~~~~~~~~~~~~~~~~] Open Systems Integration pww@bnr.ca [ ] Bell Northern Research Ph: +1 613 765 2064 [ ] P.O. Box 3511, Station C FAX:+1 613 763 3283 [__________________________] Ottawa, Ontario, K1Y 4H7
melby@daffy.yk.Fujitsu.CO.JP (John B. Melby) (01/30/91)
As far as I am aware, all implementations of MNP 5 must be licensed from Microcom. Apparently, MNP 4 is not subject to these terms, and I have heard of public-domain implementations in the MS-DOS world. Disclaimer: I speak only for myself and cannot guarantee the correctness of the above info, etc.... ----- John B. Melby Fujitsu Limited, Machida, Japan melby%yk.fujitsu.co.jp@uunet
lstowell@pyrnova.pyramid.com (Lon Stowell) (01/30/91)
McGraw Hill has available a set of the '88 standards (most of them except Internet RFC's) for $795. This is CONSIDERABLY cheaper than the competition, and hopefully will cause Omnicom to rethink their pricing policies. Students and educational orgs should have a cheaper source if ISO is to ever become as flexible and ubiquitous as TCP/IP. More importantly, the smaller high-tech developers should be able to obtain these w/o risking bankruptcy...these small shops provide fast-footed products to enhance/compete with the larger organizations which fund ANSII. BTW, I believe ANSII, not IEEE, "owns" the copyrights to all o f the ITU standards in the U.S..... How is it in other countries? Are the standards reasonably available at prices suitable for small firms and students? Is the U.S. in the position of possibly losing it's competitive position in ISO, or are most other nations similarly handicapped by IMHO such short-sighted policies?
csg@pyramid.pyramid.com (Carl S. Gutekunst) (01/31/91)
>...the observation remains valid that is is not good economics to hire good >programmers at current programmer salaries (plus overhead) and then deprive >them of the documents because they cost $200-$1000! What everyone seems to be forgetting is that, with ISO, you can spend as much as $200 FOR ONE DOCUMENT. The complete X.25 set (ISO7776 and ISO8882, with addendums) is $180. The ISO8882 conformance test is $165. If you are trying to actually build a library, these add up *very* quickly. By comparison, CCITT Blue Book Fascicle VIII.2, which includes everything from X.1 to X.32, is $62 from the U.N. Bookshop. ISO7776 and ISO8882 are better documents than X.25, but not *that* much better. Considering how much Omnicom price gouges on CCITT Blue Books (they want $80 for that same Fascicle), where they have competition, I'm a wee suspicious of the prices for ISO documents, where they have none. <csg>
enag@ifi.uio.no (Erik Naggum) (01/31/91)
Lon,
I think you're a little confused.
When you speak of "the '88 standards", I believe you're referring to
CCITT documents, not ISO documents. CCITT docs are considerably
cheaper than their ISO counterparts from the outset. I've observed a
factor of three as the norm, with peaks as high as a factor of 20.
When you speak of "most of them except Internet RFC's", I'm not sure
what you mean. Internet RFC's are not "standards". They are not
published every four years, either, or yearly or any other fixed time
interval.
I'm quite certain it's "ANSI" with only one I, as opposed to one of
the more ubiquitous standards known as ASCII.
When you speak of ANSI "owning" the copyrights of all ITU standards in
the U.S., I believe you intended to indicate a licensing agreement.
Is this so? Last I heard, Global Engineering had a license agreement
with ITU, and ANSI is rumored to have given Global exclusive right to
sell or duplicate their standards.
You ask about other countries. I hope Norway qualifies. I've bought
about two shelf-feet of CCITT's Blue Book and it's quite inexpensive.
Only $500 a shelf-foot or something in the vicinity thereof. I've
also bought about three shelf-inches of ISO standards, and that
totalled $2,000, so we're talking $4,000 a shelf-foot. Some of the
more perversely priced documents would extrapolate to $10,000 a
shelf-foot, but that's unusual. Most notably, the ISO ODA specs cost
12 times more than the CCITT specs, and they are supposed to be
"technically aligned". There are things in the ISO ODA which are not
in the CCITT ODA, but there are also issues of decency in pricing,
here.
I've wanted to acquire the ISO documents for MOTIS, but CCITT X.400
was available at 1/8 of the cost, so I bought that instead. I also
wanted to acquire the ISO X.25 specifications because of the higher
quality of the description of several of the "sub-layers", but I've
had to shelve that. I bought SGML and its amendment and associated
SDIF and registration stuff, and the TR (9573), and then I recently
bought two copies of Goldfarb's {The SGML Handbook} shipped with DHL
from London or maybe it was New York, never figured out, and both of
these books cost much less than the above mentioned standards. Note
that this book _contains_ the ISO 8879 text, including amendment and
suggested improvements.
In my opinion, ISO's pricing policy is somewhere between insane and
mentally warped. Also, I hate Helvetica for regular text, but that's
another of their wrong decisions, along with their insistance on two-
column text. CCITT goes for what looks like Times Roman and has full-
page layout, and it's a lot easier to deal with. I have one gripe
with CCITT, though: They make a lot of typos and even though I get
"corrigenda" sheets in the mail every four weeks or so, it's annoying.
I have found ISO with their incessant Amendments are even worse to
keep track of. Finding out what section foo in the amended SGML said
was a real pain in the butt.
Here's one who would really appreciate electronic versions, and diffs
each time they made changes. Not only would it be possible to find
out what the current version actually says, one could make hypertext
links into this mass of mangled English prose. (Goldfarb's SGML
Handbook has made this possible, _on_paper_, and I'm really impressed.
I wish more publishers would get the idea.)
From what I've seen, there's no danger of the U.S. losing its
competitive position in ISO because of unfair pricing policies.
BTW, I'm not opposed to copyright on standards, restrictive dupli-
cation policies or covering costs, it's just that with a little more
sense of marketing these things, they could easily sell an order of
magnitude more of them and make a lot of money in the process. If ISO
8859 was available in the bookstore, I'm sure several students over
here would have had their own copy, for instance.
The times when only manufacturers were interested in ISO standards are
gone. ISO doesn't seem to have noticed its recent popularity.
--
[Erik Naggum] Snail: Naggum Software / BOX 1570 VIKA / 0118 OSLO / NORWAY
Mail: <erik@naggum.uu.no>, <enag@ifi.uio.no>
My opinions. Wail: +47-2-836-863 Another int'l standards dude.csg@pyramid.pyramid.com (Carl S. Gutekunst) (01/31/91)
>When you speak of "the '88 standards", I believe you're referring to >CCITT documents, not ISO documents. The McGraw-Hill Encyclopedia to which Lon was referring contains a good cross- section of ANSI, IEEE, ISO, CCITT Blue Book, and ECMA standards. The now-obso- lete 1984 edition sold for considerably less than the cost of the individual ISO docs it contained. I'd guess that the same is true for the 1988 edition, but haven't actually checked. >When you speak of "most of them except Internet RFC's", I'm not sure >what you mean. Internet RFC's are not "standards". Some of them certainly are. RFC's have a status associated with them, just as ISO documents do. The difference is that when the IETF issues an update, it issues a new RFC number. ISO and CCITT keep the same number regardless of the number of revisions through which a document has gone. <csg>
lstowell@pyrnova.pyramid.com (Lon Stowell) (01/31/91)
In article <ENAG.91Jan30204136@svarte.ifi.uio.no> enag@ifi.uio.no (Erik Naggum) writes: >Lon, > >I think you're a little confused. Likely, but I don't think you read very carefully. > >When you speak of "the '88 standards", I believe you're referring to >CCITT documents, not ISO documents. CCITT docs are considerably >cheaper than their ISO counterparts from the outset. I've observed a >factor of three as the norm, with peaks as high as a factor of 20. > To be equally picky, CCITT produces only "Recommendations". I believe you missed my point,,,,there are many ISO documents which are based on CCITT ones....rather than including these, a cross-reference to the CCITT number is used. There USED to be such a price differential here, and maybe that is why the mentioned books are cheaper than Omnicoms...perhaps they only publish the related CCITT and x-ref the ISO....I don't know why the price difference, just guessing. >When you speak of "most of them except Internet RFC's", I'm not sure >what you mean. Internet RFC's are not "standards". They are not >published every four years, either, or yearly or any other fixed time >interval. > I am aware what the initials stand for. Again you seem to have deliberately missed the point. The Internet doc's are not included in the bookset I mentioned. Semantics on whether they are "standards" or "request for comments" aside, these are not typically included in commercial publications here in the US, nor are they copyrighted by the same folks with delusions of pricing that control ISO. >I'm quite certain it's "ANSI" with only one I, as opposed to one of >the more ubiquitous standards known as ASCII. No useful information in this statement. If you insist on using a spell-checker, please pipe it to /dev/null, not the net--thanking you in advance, etc. > >When you speak of ANSI "owning" the copyrights of all ITU standards in >the U.S., I believe you intended to indicate a licensing agreement. >Is this so? Last I heard, Global Engineering had a license agreement >with ITU, and ANSI is rumored to have given Global exclusive right to >sell or duplicate their standards. > In the US, the Congress actually controls all treaty type info. My understanding is that the copyrights have been assigned to ANSI for the ISO docs...although I have recently seen mention of availability from the IEEE....if anyone has FACTS about this please contribute. >You ask about other countries. I hope Norway qualifies. I've bought >about two shelf-feet of CCITT's Blue Book and it's quite inexpensive. >Only $500 a shelf-foot or something in the vicinity thereof. I've >also bought about three shelf-inches of ISO standards, and that >totalled $2,000, so we're talking $4,000 a shelf-foot. Some of the >more perversely priced documents would extrapolate to $10,000 a >shelf-foot, but that's unusual. Most notably, the ISO ODA specs cost >12 times more than the CCITT specs, and they are supposed to be >"technically aligned". There are things in the ISO ODA which are not >in the CCITT ODA, but there are also issues of decency in pricing, >here. > Pricing in each country is set by the organization in that country which participates in the international body. I don't see '88 CCITT's over here much cheaper than ISO's...but it looks like there MAY be some emerging price competition and it is IMHO, about time. >I've wanted to acquire the ISO documents for MOTIS, but CCITT X.400 >was available at 1/8 of the cost, so I bought that instead. I also >wanted to acquire the ISO X.25 specifications because of the higher >quality of the description of several of the "sub-layers", but I've >had to shelve that. I bought SGML and its amendment and associated >SDIF and registration stuff, and the TR (9573), and then I recently >bought two copies of Goldfarb's {The SGML Handbook} shipped with DHL >from London or maybe it was New York, never figured out, and both of >these books cost much less than the above mentioned standards. Note >that this book _contains_ the ISO 8879 text, including amendment and >suggested improvements. > >In my opinion, ISO's pricing policy is somewhere between insane and >mentally warped. Also, I hate Helvetica for regular text, but that's >another of their wrong decisions, along with their insistance on two- >column text. CCITT goes for what looks like Times Roman and has full- >page layout, and it's a lot easier to deal with. I have one gripe >with CCITT, though: They make a lot of typos and even though I get >"corrigenda" sheets in the mail every four weeks or so, it's annoying. >I have found ISO with their incessant Amendments are even worse to >keep track of. Finding out what section foo in the amended SGML said >was a real pain in the butt. > Fonts, etc. are set by the publisher....unless one is looking at the raw document copies... >Here's one who would really appreciate electronic versions, and diffs >each time they made changes. Not only would it be possible to find >out what the current version actually says, one could make hypertext >links into this mass of mangled English prose. (Goldfarb's SGML >Handbook has made this possible, _on_paper_, and I'm really impressed. >I wish more publishers would get the idea.) > I would like electronic versions as well, but I bet that first there would be a CCITT Recommendation for the electronic format... >:-) I prefer MakerInterchangeFormat, but that's 'cause I hate the *roff formats. Likewise would not like to see IBM/GML formatting. Maybe that's why they are not available...other than the $$$ issues of reproduction, perhaps. >From what I've seen, there's no danger of the U.S. losing its >competitive position in ISO because of unfair pricing policies. > IMHO (and not an official statement by any means) the US is not overly competitive in ISO, but it is not, IMHO, due to pricing. Rather the widespread use of TCP/IP and the existing installed base is more likely the cause. Remember the old saying that Creation only took 6 days, 'cause there was no install base to worry about.. >BTW, I'm not opposed to copyright on standards, restrictive dupli- >cation policies or covering costs, it's just that with a little more >sense of marketing these things, they could easily sell an order of >magnitude more of them and make a lot of money in the process. If ISO >8859 was available in the bookstore, I'm sure several students over >here would have had their own copy, for instance. > >The times when only manufacturers were interested in ISO standards are >gone. ISO doesn't seem to have noticed its recent popularity. Even in larger companies, generally only one or two copies of ISO doc's are available. This keeps all but a few personnel from perusing them at leisure. Also keeps one from high-lighting, margin noting, adding comments/corrections, etc....all the little things one does to enhance one's recall and understanding of otherwise BORING technical information. I really feel that this lack of ready access to the commercial crowd and the educational crowd is actually hurting roll-out of OSI...
csg@pyramid.pyramid.com (Carl S. Gutekunst) (02/01/91)
In article <143009@pyramid.pyramid.com> I wrote: >The now-obsolete 1984 edition [of the McGraw-Hill Encyclopedia] sold for >considerably less than the cost of the individual ISO docs it contained. This is exagerated, for which I apoligize. There are only five documents in the '84 edition, 72 pages, whose list price total is about $100. <csg>