On Sun, Nov 4, 2018, at 13:59, Jiankang Yao wrote:
> Your suggestion is very good. RFC7997 is an informational RFC, and
> shows a direction.
Not just a direction but possible things right now.
Because other formats than text are possible now, like PDF.
> Because the popular form of RFC is TXT, the TXT can not display non
> ASCII very well.
No, see RFC 7990 and 7995.
> I can use the natvie unicode characters in draft, but many readers
> will not display it correctly.
The RFC cited says how things can be dealt with and the difference between
the txt and PDF format for example.
If someone has to deal with internationalization and hence is interested
by your document, I guess he knows how to deal with that and properly display
characters.
> According to my current understanding, U+XXXX is still the proper
> way to be easily understood by most readers.
I still feel your document to be difficult to read, just because of that.
Things like:
<b-dn:rdn uLabel="U+5B9E""U+4F8B".example>xn--fsq270a.example</b-dn:rdn>
are not easy, besides being invalid XML.
(I pity the document shepherd that will need to validate all XML
snippets, this can NOT be just copied and pasted in some validtor)
Which is exactly why the RFC7997 says that directly using the unicode characters
as is is permissible in examples.
> Many years ago, IETF EAI WG also discussed this related issue.
> I still do not see which rfc uses the natvie unicode characters
> directly.
RFC8187 for one.
> If possilbe, I may suggest to add some texts in the draft, which says
> "in future, the natvie unicode characters instead of U+XXXX notation are
> suggested to use in the document"
As outlined, it is not in the future (hence this sentence is not necessary),
you can do it today already with RFC7997 guidance.
If you do not want to do it, then I do not recommend adding such a sentence.
And if you want to do it (apply RFC 7997 guidance) then you do not need
this sentence either, but you can put one explaining that examples contains
unicode characters, as permissible by section such and such of RFC7997.
If your draft goes further in the process, I guess this issue will come again
at various last calls and reviews.
--
Patrick Mevzek
p...@dotandco.com
_______________________________________________
regext mailing list
regext@ietf.org
https://www.ietf.org/mailman/listinfo/regext
