[rfc-i] Does the canonical RFC format need to be "readable" by developers and others?

Julian Reschke julian.reschke at gmx.de
Fri Jun 22 14:17:34 PDT 2012

On 2012-06-22 23:09, Iljitsch van Beijnum wrote:
> On 22 Jun 2012, at 22:45 , Julian Reschke wrote:
>>> As to the discussion about XML2RFC as the canonical new RFC format: I don't think we can assume the ability to work with this format meets our longevity goals.
>> Can you explain how you came to that conclusion?
> As far as I know, there are two ways to display XML2RFC documents:
> 1. Convert them to something displayable
> 2. Get a browser to display them

Both are the same; they just differ on when the conversion happens.

> As for 1, we now have RFCs that are about 40 years old. Do we really expect XML2RFC to be around and be compatible with how it operates today in 40 years?

I don't see why not, unless we break it.

> About 2: I don't really know how this works. But as far as I can tell, this is a pretty obscure browser feature. It works well with the last ftp64 draft that I wrote in Safari, but Firefox 3.5 (yes, I know it's old) doesn't like the references to RFCs that I used (should be fixable, but it's strange that Safari has no problem and Firefox does) and Chrome just shows an empty window. All of this fails to inspire the requisite confidence.

It's totally not obscure; it has a stable spec (XSLT 1.0) and several 
compatible implementations (IE, Webkit, Firefox, Opera).

Browsers *do* differ in how they allow including remote data; the advice 
is not to try if you want to run it in a browser.

>>> Adopting XML2RFC as the canonical format would also make changes to the format much harder, depriving XML2RFC users of innovation.
>>> ...
>> It makes *incompatible* changes harder. How would that be different for other formats, though?
> Not sure how something could be different but "compatible". Small things can have large consequences.
> It's not different from other formats, but the likelihood of the IETF changing other formats seems smaller.

How so?

Best regards, Julian

More information about the rfc-interest mailing list