[rfc-i] [IAB Trac] #266: Requirement for "Clear Printing"

RJ Atkinson rja.lists at gmail.com
Fri Feb 15 07:00:15 PST 2013


  A key feature of the current RFC publication process is 
that a given RFC (or I-D) will have identical pagination 
-- regardless of whether it is printed on A4 paper, 
printed on US-Letter paper, or even viewed in a web browser 
(i.e., browser showing text/plain format, rather than 
a browser showing HTML format).  

  This means that when working with colleagues, one
can talk about some RFC (or I-D) and use the page 
number as a reference/anchor point in our discussions.
Today, even folks working from an electronic copy
(if using the text/plain ".txt" format, rather than the
HTML format) also have that pagination visible to them.

  It would be a nightmare to have a colleague refer
to (his/her) "page 4" and have that be "page 3" or
"page 5" on my own copy.  That would be a huge waste
of a lot of folks' time, and seems easily avoidable.

  While I'm sure that individual custom varies, nearly
all of the implementers that I know well normally work 
from a printed copy.  Similarly, many reviewers of I-Ds 
and readers of RFCs find it more effective to work from
a printed copy.  

  Curiously, in my own (non-scientific) sample of people, 
there is no "generational" aspect to this.  This practice 
of working from print when coding or designing an implementation
seems as common among folks who are 20-something as it is
with folks who are 40-something.

  While I have no objection to making RFCs available
in other formats (e.g., HTML without pagination), I would
be greatly obliged if a print-oriented format with existing
fixed/predictable/consistent pagination and fixed/predictable
line-length remained available for those who wish to continue 
to work from the decades-long text/plain (".txt") format.

  The 2 current text/plain format rules cause printing
to work equally well with A4 and US-Letter paper sizes.
So one can work trans-Atlantic or trans-Pacific and all
be on the same page:
	* 58 lines/page
	* 72 characters (plus CR and LF)/line

   So my requested action is that Section 3.3. be revised
to delete the "retirement" of the 58 lines/page and 
72 characters (plus CR and LF)/line rules.  

  I'd be quite happy if these 2 rules were scoped to apply only 
to the "text/plain" format of RFCs (and I-Ds), such that those 
2 rules did NOT apply if one were generating an HTML format 
or some other format.  Of course, this implies continuing the 
automatic generation of a print-oriented format (i.e., text/plain 
with ".txt" filename extension) that DOES comply with those 
2 rules.

  This change to the RFC Format Requirements draft would retain 
the critical uniformity of pagination in the existing widely used 
printable format, while allowing flexibility in line-length and 
pagination in other formats (e.g, HTML).


Ran Atkinson

More information about the rfc-interest mailing list