Re: How to pay $47 for a copy of RFC 793

[John C Klensin <[e-mail]>]

--On Sunday, May 08, 2011 15:06 -0700 Bob Braden
<[e-mail]> wrote:

> I just discovered an astonishing example of misinformation,
> shall we say, in the IEEE electric power community. There is
> an IEEE standards document C37.118, entitled (you don't care)
> "IEEE Standard for Synchrophasors for Power Systems
> C37-118(TM)-2005", which is currently of great importance for
> the instrumentation of the national power grid. I just noticed
> that it references RFC 793, and for curiosity looked to see
> how it was referenced. I found:
> 
>     [B8] RFC 793-1981,Transmission Control Protocol DARPA
> Internet Program Protocol Specification.[12]
>...
> Now, it has always been IETF's (and even before there was an
> IETF, Jon Postel's) policy to allow people to sell RFCs. What
> astonishes me is that clever people in the IEEE don't know
> RFCs are available free online. I guess RFCs remain so
> counter-cultural that industrial types don't get it. I wonder
> how many other IEEE standards contain similar RFC-for-pay
> references..

Bob,

What you presumably remember, but others reading this may not,
was just how many comments Jon made about the impossibility of
preventing fools from throwing their money away.  And, of
course, it is in the interest of Global Engineering Documents
--which, in the era in which few folks had direct access to the
Internet was one of the better sources for miscellaneous
technical standards documents-- to let people continue to
believe that they are a convenient and standard (sic) source.



--On Sunday, May 08, 2011 21:26 -0400 "John R. Levine"
<[e-mail]> wrote:

>...
> This isn't an enormous project, but it requires figuring out
> which online libraries are worth getting into, making the
> necessary arrangements with them (which may or may not involve
> money), a batch process to load in all the existing RFCs, and
> arrange with the production house to ensure that each new RFC
> gets listed as it's published.  Most of these systems include
> abstracts and forward and backward references, which will
> doubtless require some debugging to make them work reliably.
> 
> Like I said, it's a good project for the new RFC series
> editor.  It should be a lot easier than deciding how to put
> Chinese names into RFCs.

+1

I do note, however, that RFCs appear to be listed in ACM's Guide
to Computing Literature (essentially part of the ACM Digital
Library at this stage).  Putting "Transmission Control Protocol"
into the search mechanism turns up RFC 793 in a hurry.  And,
behold, they have full text available and retrieving it works
without any charges other than the access fees for the Digital
Library itself.  "RFC Editor" is even on their list of
publishers for search purposes.  

The problem is that the titles they index do not contain the RFC
numbers, so looking up "RFC793" or "RFC 793".  That is not a
decision to avoid indexing the series (which would require the
process John outlines to reverse) but a bug.   I have filed a
bug report as Digital Library feedback.

    john