[rfc-i] Following up from Atlanta

Joe Hildebrand (jhildebr) jhildebr at cisco.com
Wed Dec 5 22:18:49 PST 2012


On 12/5/12 1:02 PM, "Nico Williams" <nico at cryptonector.com> wrote:

>On Wed, Dec 5, 2012 at 2:51 PM, Joe Hildebrand (jhildebr)
><jhildebr at cisco.com> wrote:
>> On 12/5/12 11:51 AM, "Nico Williams" <nico at cryptonector.com> wrote:
>>>Agreed.  If browsers had what I need then I might drop my request for
>>>text.  Eventually the browsers will all get there (and maybe Opera has
>>>all that I need); maybe I should preemptively stop asking for text.
>>>Something for me to think about.
>>
>> The prototype I did allowed you to overwrite the RFC editor's default
>>CSS
>> with your own fonts, colors, etc.  The idea was that if you're using the
>> default look and feel, you could treat the document as canonical.
>
>I'll have to look at it.  Link, again?

Tooling:
https://github.com/IETF-Formatters/html-rfc


Example:
http://cursive.net/draft-hildebrand-html-rfc-2012-07-29.html


> How do I get regexp support?

elinks -dump draft-hildebrand-html-rfc-2012-07-29.html | grep -i unicode

Or just search the .html, if you prefer.


>(I also would love it if xml2rfc were to set id attributes that are
>useful for navigating the DOM, and for browsers to have nice,
>keyboard- and mouse-based DOM navigation.  These missing features are
>why I like a regular text format that I can trivially search for the
>things I am looking for.  Yes, I'm very lazy, which is why I wrote
>lyx2rfc...)

The tooling above generates id's for most everything.

>Yeah, but I don't buy it.  Sorry, if you're reading an RFC you should
>expect UTF-8.  

Maybe.  Some of the experiments we did a couple of months ago (including
putting the first non-ASCII codepoint several pages down into the text
below where a heuristic search would guess UTF-8) had several folks
thinking it wasn't good enough.

>Also, garbage on a tty happens -plenty- and it's not
>the end of the world.  If someone is still running a non-UTF-8 locale,
>well, they can still find ways to view UTF-8 content; also: they
>should switch as soon as practicable (I know, legacy, legacy, legacy).

I personally am fine with that; was just repeating where we had ended up
last time.

-- 
Joe Hildebrand





More information about the rfc-interest mailing list