Re: [precis] I-D Action: draft-ietf-precis-framework-10.txt

[Peter Saint-Andre <[e-mail]>]

On 10/16/13 5:32 AM, Florian Zeitz wrote:
> Am 16.10.2013 00:52, schrieb Peter Saint-Andre:
>> On 10/15/13 4:26 PM, [e-mail] wrote:
>>
>>> The IETF datatracker status page for this draft is:
>>> https://datatracker.ietf.org/doc/draft-ietf-precis-framework
>>>
>>> There's also a htmlized version available at:
>>> http://tools.ietf.org/html/draft-ietf-precis-framework-10
>>>
>>> A diff from the previous version is available at:
>>> http://www.ietf.org/rfcdiff?url2=draft-ietf-precis-framework-10
>>
>> Reviews of -10 or the diff would be helpful. I plan to check them both
>> to make sure that I've accurately addressed all of the last call feedback.
>>
>> Thanks!
>>
>> Peter
>>
> I've had a quick look at the diff, and some areas I felt I should look
> at again. I'm generally very happy about the outcome. 

Excellent, thanks for reviewing the changes.

> Thank you for your
> hard work. Some minor comments:
> 
> Sections 3.2.2 and 3.3.2 say «Certain characters from the Exceptions
> ("F")». That's a bit awkward as a reader. A forward reference to section
> 7.6, noting they are explicitly listed there would help.

I agree about the forward reference. Will add. (Also the xref in that
text was pointing to Section 7.5, not Section 7.6.)

> In Section 4.1.3 about case mapping we currently have:
> «In general, the combination of case preservation and case-insensitive
> comparison of internationalized strings is NOT RECOMMENDED; instead,
> application protocols SHOULD either (a) not preserve case but perform
> case-insensitive comparison or (b) preserve case but perform
> case-sensitive comparison.»
> I wonder what the rational for this is? Both SASLprep-bis and Nicknames
> seem to want case-preservation, but case-insensitive comparison. I may
> be mistaken though.

You make a good point. I can't recall the rationale for that text, but
it does seem to be violated by several of our profiles, so I'm inclined
to remove it.

> In Section 10.5 "Visually Similar Characters" I'd suggest a reference to
> UTS #39, describing confusable detection and mixed-script detection.

That document is referenced in Section 10.1, but mentioning it again in
10.5 would be good. (Also the reference was incorrect since it said
UTR39, not UTS39.)

Peter

-- 
Peter Saint-Andre
https://stpeter.im/