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

Paul Hoffman paul.hoffman at vpnc.org
Thu Jun 21 12:47:11 PDT 2012


On Jun 21, 2012, at 12:36 PM, Joel M. Halpern wrote:

> I have thought of the term canonical as being akin to normative.

Now we are throwing around multiple terms that have different meanings to different people. What do you mean by "normative"?

> If we have transformations, periodically there will be inconsistencies.

Exactly right. For example, a transformation to "plain text" and a transformation to "HTML" will be inconsistent if there is external art.

> It seems to me helpful, and arguably necessary, to have a specific, singular, form which can be read in the case o finconsistency.

What do you mean by "read"? You can read XML and HTML: I believe I have seen you do both.

> thus, I am very uncomfortable with a canonical form that is the xml, etc.

By "etc" do you also mean "HTML"?

> Further, I note that in practice, checking for correctness will be done by the authors (e.g auth48)on the output form, not the input form. Hence, I just can't see how the input can be canonical.

That would argue that the RFC Editor could only publish one output format, the one that was reviewed by the authors. Is that what you mean?

--Paul Hoffman


More information about the rfc-interest mailing list