Re: [Gendispatch] Updating the IETF Discussion List Charter (was: Fwd: New Version Notification for draft-eggert-bcp45bis-02.txt)

[Keith Moore <[e-mail]>]

On 6/26/22 18:04, Brian E Carpenter wrote:

>>
>> Is there a way to skip xml2rfc entirely?  IMO that's the biggest part 
>> of the problem.   I can accept the utility of XML for an RFC that 
>> needs to be maintained for a long time, but I'm not at all sure it's 
>> worth the trouble for a -00 Internet-Draft. Maybe the right time to 
>> move to XML is when an author or WG is ready to submit to IESG.
>
> That's exactly where kramdown fits in. There will be an XML file, but 
> it is automatically generated and you never need to look inside it. 
> I'm no expert, I've only done a handful of drafts in kramdown, but 
> it's clearly more approachable than raw XML, or even nroff.
>
> (I have to say that there are corner cases where you *do* need to look 
> at the XML, but I think those will get less as time goes on.)

It's generally been my experience that there's a "leaning tower" effect 
resulting from layering one hack on top of another, resulting in a 
system that is overall both more fragile, and harder to 
understand/debug, than a single well-designed layer. And I've never 
really thought that either XML or SGML were fit for ordinary authors to use.

But I'm happy to try kramdown the next time I write an I-D.  It's got to 
be worth a try.   If it only reduces the effort of writing 50% of 
drafts, that's still a big improvement.

>
> Have you tried authoring in an ACM or IEEE format recently? They are 
> *seriously* harder to satisfy than the IETF, even if you are 
> experienced in using LaTeX. 

Not recently.  I haven't worked in academia for about 15 years, and the 
occasional document that I wrote that was published by either 
organization was formatted for me by someone unknown to me.

I just think that IETF needs to impose as few barriers as possible to 
submit new ideas and proposals.   We shouldn't require the XML stuff 
until there's clearly some support for a document.

Keith