[rfc-i] v3imp #2 Control over paginated output

Paul Hoffman paul.hoffman at vpnc.org
Sat Jan 24 08:12:31 PST 2015

On Jan 24, 2015, at 1:05 AM, Julian Reschke <julian.reschke at gmx.de> wrote:
> On 2015-01-24 09:35, Sean Leonard wrote:
>> On 1/24/2015 12:14 AM, Julian Reschke wrote:
>>> Before we add even more than that, we need a *concrete* example of
>>> where that really is needed.
>> I feel that every time these issues are raised, you ask for a "concrete
>> example". When a concrete example is given, you respond that it is "not
>> a real example", "not a concrete example", "trivial", "not good enough",
>> "not really needed", or "solved some other way" (which does not actually
>> solve the problem, just pushes it out of xml2rfc to somewhere else,
>> where it remains unsolved).
>> I gave you a large number of concrete examples: they are littered
>> throughout draft-josefsson-pkix-textual.
>> Please provide concrete examples of what you consider to be acceptable
>> concrete examples.
> A concrete example would be one that shows the markup in the source file, and a page break that xml2rfv has generated when it shouldn't have.
> I'm not going to do that research; it's you who claims that what the spec has is insufficient.

Sean: I agree with Julian here. So far, your theoretical examples have seemed to me like "nice to have", not "need". That is, if such a break appeared in a real paginated document, a reader might have a momentary blank, but then would completely understand the actual contents.

The RFC series has had 40 years where probably one or two RFCs a year had bad page breaks, and the series is doing just fine. If eliminating bad page breaks were free, of course we would do it. Giving authors new elements that they feel that they have to enter in their documents, and even then might still have surprising results at the end, is not free. Thus, you need to show convincing examples of "need" in order to overcome the cost.

--Paul Hoffman

More information about the rfc-interest mailing list