Re: [Tools-discuss] .txt? [I-D Action: draft-xxx.txt]

[Julian Reschke <[e-mail]>]

Am 29.06.2021 um 23:41 schrieb Toerless Eckert:
> On Tue, Jun 29, 2021 at 08:08:48AM +0200, Julian Reschke wrote:
>> Am 29.06.2021 um 01:47 schrieb Toerless Eckert:
>>> ...
>>>> I've had no trouble at all editing XML in v3.
>>>
>>> Some tags got more convoluted. And figuring out how to put RFC references into v3 was also more convoluted han in v2.
>>
>> a) Which?
>
> I forgot the details and it would take me effort to put that together,
> and right now i do not have the trust that spending that effort would
> achieve anything in getting the tool improved.
>
> So, for now i have submitted a ticket for what i consider from my own experience to
> be the worst problem, and if that actually does get fixed, i will go back and look
> into reporting the other issues.
>
> https://trac.ietf.org/trac/xml2rfc/ticket/650

Please see the xml2rfc --help:

V2-V3 Converter Options:
   --add-xinclude        replace reference elements with RFC and
Internet-Draft
                         seriesInfo with the appropriate XInclude element
   --draft-revisions, --draft-revs
                         Reference explicit draft revisions when inserting
                         XIncludes for draft references
   --strict              be strict about stripping some deprecated
attributes



>> b) It's just a slightly different syntax (a standard one instead of a
>> what actually was a hack).
>
>
> v2 was already a fairly bad markup for human editing, even if many
> of us seem to suffer from Stockholm syndrom and claim to like it.
>
> When we make v3 even harder to edit by humans because it needs to
> be better for machine processing, then we should really talk about
> a standard tools-team supported text-edit format like kramdown.

"...because if needs to be better for machine processing..." - who said
that?

There are places where there are currently discussions to remove
authoring features in favor of easier processing, and it would be nice
if I wasn't the only one arguing for being author-friendly (in this
case: content model for lists).

> ...

Best regards, Julian