[rfc-i] Character sets, was Comments on draft-iab-rfcformat

Heather Flanagan (RFC Series Editor) rse at rfc-editor.org
Wed Dec 19 16:54:34 PST 2012

On 12/19/12 4:13 PM, Martin Rex wrote:
> Phillip Hallam-Baker wrote:
>> Why does normative text need to be displayable on every machine?
> Why NOT?

In my own opinion, which I am forming from a librarian background
rather than a computer science background, normative text does not
need to be displayable on *every* machine because there comes a point
when information such as links and line art make a document more
readable and therefore useable to a broader audience.

If one said "normative text needs to be displayed on most machines", I
think that gets more in to the spirit of things, but I have no way of
getting to a quantitative definition of "most".  This is where common
sense and advice comes in to play for me.

> Not being able to discuss normative text on an IETF mailing list 
> because it can not be represented in a plain-ASCII EMail would be 
> just as bad as not being able to display an RFC through a plain SSH
> tty-based dialin, or discussed in f2f meetings.
> Everything outside outside of US-ASCII is a barrier to
> comprehension for various communication scenarios.

I have heard this argument and documented it in the requirements
document.  I recognize the concern, but I do not agree that it is one
that should prevent the RFC Editor from allowing text that includes
more than US-ASCII characters in an RFC.  Exactly how and where still
needs clarification.

> And the same KISS principle applies in order to keep our documents 
> highly "accessible" to those, where the limitations are not caused 
> by the favorite physical computing environment, but by reduced 
> vision, hearing or both capabilities.

I agree that we need to keep these documents accessible, but I think
there are ways to do that that in turn support more than one type of
reading and printing mechanics.

> For many years there has been the possibility to publish an RFC in
> an additional format with colorful text, fancy type faces and
> sizes, graphics and all.  Considering how rarely it is used, there
> evidently is no signficant demand, and neither is there necessity.

I am also curious as to why more individuals have not chosen this
option, but I don't think there is enough information to assign a
conclusion.  My own perception is that it could be people just don't
realize that option, OR they feel it's more of a hoop than they want
to jump through given all the other things they've done to get the I-D
to an RFC publication, OR there is not a need.

In terms of additional options, I suspect people do want more than
just the very very plain text given how popular the datatracker html
version of the RFCs are, as judged by how often I see the datatracker
version referenced in various bibliographies rather than the
rfc-editor.org version.


> For those who do have an urge, the possibility to publish an
> alternative document *IS* already there.  Stop whining and *USE*
> it.
> -Martin _______________________________________________ 
> rfc-interest mailing list rfc-interest at rfc-editor.org 
> https://www.rfc-editor.org/mailman/listinfo/rfc-interest

More information about the rfc-interest mailing list