Re: [Rfc-markdown] [xml2rfc-dev] <br> is back, was: New xml2rfc release: v2.32.0

[Henrik Levkowetz <[e-mail]>]

On 2019-10-04 17:40, Carsten Bormann wrote:
> Something is not right with the process if we have a clear
> requirement and an obvious solution, but instead have to resort to
> using an obscure, widely shunned Unicode feature so we appear not to
> violate the specification.

I couldn't agree more.

My recollection is that the pushback from design team members against
permitting <br> generally when this was discussed in October 2018 was so
strong that it was not accepted.  Few other people participated in the
discussion:

https://mailarchive.ietf.org/arch/browse/xml2rfc-dev/?q=subject%3A%2337

	Henrik

> 
> Grüße, Carsten
> 
> 
>> On Oct 4, 2019, at 17:34, Henrik Levkowetz <[e-mail]> wrote:
>> 
>> Signed PGP part
>> 
>> On 2019-10-04 17:25, HANSEN, TONY L wrote:
>>> On 10/4/19, 10:49 AM, "xml2rfc-dev on behalf of Julian Reschke" <[e-mail] on behalf of [e-mail]> wrote:
>>> 
>>>    On 04.10.2019 14:41, Henrik Levkowetz wrote:
>>>> ...
>>>>   * Improved the handling of U+2028 in text output, and fixed a bug in the
>>>>   * handling of U+2028 in the HTML output.
>>>> ...
>>> 
>>>    So U+2028 is Unicode "LINE SEPARATOR". What this means is that xml2rfc
>>>    now supports forced line breaks, just a few weeks (months?) after there
>>>    was a decision not to include the <br> element.
>>> 
>>>    I think this is a really bad idea, as opposed to having an explicit <br>
>>>    element, because:
>>> 
>>>    1. It's kind of obscure (hint: browsers do not process it as line break).
>>> 
>>>    2. The grammar doesn't help people to understand where it is allowed.
>>>    Actually, where *is* it allowed? Anywhere?
>>> 
>>>    So, AFAIC, if we identify cases where we want to allow forced line
>>>    breaks, we should allow them explicitly (and in the same way HTML does).
>>> 
>>> I have always preferred having an explicit <br/> element (however it
>>> gets spelled). I fought in the design team to have support for it in
>>> at least some limited cases, and think removing it entirely was
>>> completely the wrong decision. I'm too much of a pragmatic engineer.
>> 
>> I'm strongly for having an explicit <br/> element.
>> 
>> Given a clearly expressed need from the RPC, I will always try to provide
>> tools to make it possible for them to do their work.  Without having <br/>
>> available, this was a fallback solution.  Continuing to ignore clearly
>> expressed needs of the RPC seems unproductive.
>> 
>> 
>> 	Henrik
>> 
>> 
>> 
> 
>