[comp.protocols.iso] OSI Documents

dave@TACKY.CS.OLEMISS.EDU (David E. Johnson) (11/13/89)

From the information recieved thus far, it seems that to obtain
the specifications for the OSI protocols requires a substantial
outlay of money. Other larger universities may be able to afford
to do such, but as I suspect, most do not have the resources to
apprehend them.

As Gunner Lindberg stated in a previous message, it seems that
large corporate entities are able to afford the set of protocols.
As we have seen with the DOD TCP/IP protocols, the driving force
behind their use and availability has been the presence of public
implementations and complete availability of the protocol specs.

I understand that the fee charged for the documents is to cover
printing and distribution costs, but it seems to me that an
electronic version supplied on disk or tape would cost almost
nothing and provide availability to those who cannot afford the
current selling price of the printed versions. Or as another
possible solution, the basic protocols for each layer could be
made available at a smaller fee than more in depth specifications
at least promoting common knowledge. It seems to me that with
the vast array of ISO documents available for diverse areas of
standards, a reduction in cost would actually increase profit
by increasing purchase volume.

 David E. Johnson               ** Title:        Systems Programmer
 Department of Computer Science ** Telephone:    (601) 232-7396
 The University of Mississippi  ** Internet:     dave@cs.olemiss.edu
 336 Weir Hall                  ** 
 University, MS  38677          ** 
 

mrose@CHEETAH.NYSER.NET (Marshall Rose) (11/13/89)

Long, but informative--

Well, I'm the last person to defend OSI (some say the first person), but
there are rational reasons as to why the ISO documents are distributed the way
they are.

It is important to appreciate that anyone can get a copy of any ISO
standard.  It's true that the documents are costly, but this is strictly
a cost recovery mechanism.  ISO breaks even on the deal.  In comparison
to the popular press or even university texts, there isn't really that
much demand for standards, so not many copies get printed and the cost
per unit is high.  Keep in mind these things are typeset by gnomes in
CH, they aren't ASCII-set by the actual authors.  Typesetting is
enormously personnel and time expensive.  This costs big bucks, or CH francs.

If you want to deal directly with the ISO you can, though typically it
is preferable to go to your national standards body (e.g., ANSI in the
US) for copies.  As you might expect, you will pay a little more because
there is personnel and handling involved.  There are even some companies
licensed to sell copies of standards.  The one I recommend is "The
Document Center" in Belmont, California at 415-591-7600.  They (and
companies like them) will do all the leg work to find copies of things
for you at whatever prices are there.

So why copyright the things and why not make them available on-line?
The copyright is there because it is legally necessary to make sure that
people don't change the documents.  They aren't available on-line,
primarily due to historical reasons.  There really isn't a uniform
format or anything else: it's all up to the editor of each individual
document, some still use typewriters, some use MACs, some use PCs, some
use LaTeX, etc.  The more forward-thinking editors usually make an on-line
copy of interim and final documents available to their buddies.
However, these aren't official in any sense.  Only things published by
the ISO are official.

The big-vendor conspiracy is a crock.  What you have with ISO standards
is the real world.  Although it might surprise some, freebies and
subsidies are the exception in the business world, not the rule.
It costs time, people, and money to attend standards meetings.  It costs
money to buy copies of standards.  And so on.  This is called "the cost
of doing business" or more colloquially, "the cost of admission".  At
previous companies I've been at, I've had the library order standards
for me.  They have a budget for things like this.  Unless you're a small
shop, it's almost certainly the case that someone has money for
something like this.  If not, then the organization in question should
re-evaluate priorities accordingly.  If you're a small shop, you can
sponge off someone else.  I know half-a-dozen of consultants who do this.

/mtr

ps: before someone goes calling me some kind of greedy, big-business
shill, it should be pointed out that for the last four years I have
orchestrated the largest collection of openly available OSI software
since time's dawn (now how's that for self-aggrandizement!).  I don't
have any sympathy for the ISO, the standarization process, standards
documents, or standards committees.  They are all flawed, but they are
reasonable processes for the imperfect world.  We don't live in a
perfect world, and my goal is simply to damage control the pain.

LYNCH@A.ISI.EDU (Dan Lynch) (11/13/89)

I believe that the previous messages about the high cost of ISO documents
are innacurate as to the reason for the high cost.  ISO is a large organization
that does the administrative work of organizing meetings and dealing with
appeals processes, etc.  It has no dues to support that activity.  It raises
a large part of its overhead expenses by charging more than printing and
distribution costs for its ONLY product, documents.  Wehn you look at it
in that light, things make "sense".  The RFC series has been mainly
paid for by the US Government in its many contracts to protocol developers
and to network information centers.  Thus it appears "free" to us.
Nothing is free, as we all know.  

I doo agree that it would be better for the spread of ISO's OSI protocols
if the documents were freely available.  

Dan
-------