--On Monday, November 3, 2025 11:36 +1300 Brian E Carpenter
<[email protected]> wrote:

> Wondering where to jump in, and towards the end of this message
> seems best, but I have read the whole thread up to date:
> 
> On 03-Nov-25 05:12, John C Klensin wrote:
>> 
>> 
>> --On Sunday, November 2, 2025 15:29 +0000 Paul Hoffman
>> <[email protected]> wrote:
>> 
>>> On Nov 2, 2025, at 10:17, Pete Resnick
>>> <[email protected]> wrote:
>>>> 
>>>> [Hatless]
>>>> 
>>>> I'll give an even more likely example than John's: If some kind
>>>> person at Huawei Technologies Co., Ltd. gave their company name
>>>> in Han characters as "华为技术有限公司", though I would
>>>> find a Latin script equivalent of the Pinyin version including
>>>> tone marks, "Huáwéi Jìshù Yǒuxiàn Gōngsī" helpful,
>>>> because I like getting the pronunciation right, I imagine most
>>>> people would not. I think ASCII-only (and perhaps English
>>>> translation) is really what we're looking for.
>>> 
>>> Do you want a must-be-ASCII policy just for the company name, or
>>> also for author names?
>>> 
>>> [author hat]
>>> 
>>> Specific wording for the draft would be preferred over examples.
>> 
>> I don't know if we are back to "ASCII only" in  the "interpretation
>> string, and can't tell Pete's preference from his example.  If we
>> are going that way, I'll leave to to others to propose text.   If,
>> instead, we don't believe an ASCII restriction there  is
>> appropriate (with Pete's example serving as an illustration of why
>> some non-ASCII characters may be needed if a phonetic
>> interpretation is intended, I believe fairly specific wording on
>> the two separate issues here, with two variations on one of them,
>> has already been proposed:
>> 
>> (1) Martin's suggestion to eliminate the distinction between
>> personal names and company/geographic names.  The differences are
>> not as great as the document implies and Pete's example shows.  If
>> his comments were not specific enough, that could be done by
>> dropping the second paragraph of 3.1 and inserting ", or
>> associated company or geographic names," after the first
>> occurrence of "names" in the first sentence of 3.1, although I
>> think you can come up with better ways to accomplish that.
>> 
>> (2)  Martin's suggestion was to replace "ASCII' with "Latin
>> script". Pete's example illustrates, IMO, the importance of that.
>> In the part of my note Pete did not quote, II expressed some
>> concern about that because "Latin script" may be over-broad and
>> allow reasonable people, acting in good faith to create far more
>> problematic strings than the use of diacritical markings as tone
>> marks that Pete suggested. 
> 
> Indeed. And in my opinion, no amount of wordsmithing in a policy
> document can specify what the RPC should do in enough detail, so we
> have to stay at a general level and be explicit that the RPC has the
> final say (in the case of failed negotiation with the authors).
> 
> And at the general level, I think what we can say is
> 
> "Proper names in a non-Latin script, or in a Latin script including
> unusual characters, should be accompanied by an equivalent in
> normal Latin script. The RPC makes the final decision about the
> Latin script version to be used."
> 
> [So we don't have to discuss here what's 'normal' or 'unusual'.]
> 
> Note that "proper names" covers both people and companies.

Brian,

FWIW, the above works for me although I hope we can find a better
word than "version".  I'm not sure what "version" means in the case
of Latin script and it might cause even more confusion and
bikeshedding.  Perhaps "subset" or a rewrite to "... about the
collection of Latin script characters to be used".

I read this to also leave the choice of what constitutes an "unusual
character" to the RPC, even, if necessary in their judgment, on a
per-document basis.  If others don't have that reading, some added
tweaking might be needed.

   john


   john



-- 
rswg mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to