[rfc-i] RFC Format - final requirements and next steps

Ole Jacobsen ole at cisco.com
Tue May 15 15:44:37 PDT 2012

On Wed, 16 May 2012, Iljitsch van Beijnum wrote:

> On 15 May 2012, at 23:52 , Ole Jacobsen wrote:
> > The downside is that it MAY completely ruin carefully designed and 
> > "beautiful" pages that were /designed/.
> I don't see how that can possibly be a downside. Obviously the 
> person changing the parameters and thereby ruining the design thinks 
> the ruination is an improvement, because she or he wouldn't be 
> making the changes otherwise.

What makes you think "the person changing the parameters" has ANY idea
about the intent of the designer? And why do you assume that he or she
changed ANYTHING. Lot's of these issues are badly chosen defaults,
something somebody else picked.

> Do we do that today? Unless my memory is failing me I've never seen 
> ASCII art with the text continuing beside it in an RFC.

Yes, ironically we DON'T do that today, but it IS done in most modern
publishing systems, and has been done for hundreds of years.

> This is a false dichotomy. We want to move away from imposing the 
> formatting that works best on a given output device on the file 
> format. If the file format is flexible, it can adapt to a large 
> range of display options. If the file format is HTML, CSS can be 
> used to adapt it to different displays and different user 
> preferences without the need to perform conversions or modify the 
> original file.

Sure, but the trick is to find a source format that allows this.
The fact that even a simple ASCII diagram can get really messed up
if only ONE line wraps (or you chose a variable-width font) tells
us that this isn't a simple problem to solve.

> To make a parallel with the publishing world: authors don't concern 
> themselves with the layout of their books, they just tag headlines, 
> different kinds of text, table and image captions etc etc with the 
> right style so later the designer can apply the chosen design. The 
> difference here is that rather than having one design, or perhaps 
> one for print and one for the ebook, there can/will be many designs.

Whoa! Define "authors" here. I would argue that authors have exactly 
"concerned themselves with layout" since the days of Gutenberg (at 
least). The fact that we have tools that allows untrained 
non-designers to do this today is a different matter, but don't assume 
please that authors do not or should not concern themselves with 


More information about the rfc-interest mailing list