-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 3/19/13 7:56 AM, Marc Blanchet wrote:
> 
> Le 2013-03-19 à 10:31, Peter Saint-Andre <[email protected]> a
> écrit :
> 
> On 3/19/13 7:22 AM, Matt Miller (mamille2) wrote:
>>>> 
>>>> On Mar 19, 2013, at 6:33 AM, Matthew Wild <[email protected]>
>>>>  wrote:
>>>> 
>>>>> On 19 March 2013 12:23, Joseph Yee <[email protected]> 
>>>>> wrote:
>>>>>> On Tue, Mar 19, 2013 at 3:52 AM, Peter Saint-Andre 
>>>>>> <[email protected]> wrote:
>>>>>>> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
>>>>>>> 
>>>>>>> On 3/18/13 7:03 PM, Alexey Melnikov wrote:
>>>>>>>> Hi Peter,
>>>>>>>> 
>>>>>>>> On 18/03/2013 17:18, Peter Saint-Andre wrote:
>>>>>>>>> During the WG session last week, I got the sense 
>>>>>>>>> that people find the term "NameClass" misleading 
>>>>>>>>> ("why can't I represent my name using the 
>>>>>>>>> NameClass?!"). Thus I propose changing it to 
>>>>>>>>> "SafeClass". Any objections?
>>>>>>>> 
>>>>>>>> I like SafeClass much better.
>>>>>>> 
>>>>>>> It's the best I could come up with on the plane 
>>>>>>> yesterday. :-)
>>>>>>> 
>>>>>>> My only concern is that people might think "safe" is 
>>>>>>> some kind of guarantee that nothing bad could ever
>>>>>>> happen if they use the SafeClass...
>>>>>>> 
>>>>>> 
>>>>>> Hi Peter,
>>>>>> 
>>>>>> I have the same concern about SafeClass too.  Can't
>>>>>> think of a better name, but don't see it as big concern. 
>>>>>> Probably just more text needed to explain 'safe' under
>>>>>> what condition.
>>>>>> 
>>>>>> Just throwing names out from my brain...
>>>>>> 
>>>>>> RestrictedClass? IdentifierClass? ResourceNamingClass?
>>>>> 
>>>>> I was also going to suggest IdentifierClass just as your 
>>>>> message came through. Many people already understand an 
>>>>> 'identifier' to consist of characters from a restricted
>>>>> set (as in 99% of programming languages, and various
>>>>> protocols). Usernames usually draw from a similar set of
>>>>> characters in most systems.
>>>>> 
>>>> 
>>>> +1 to IdentifierClass.
>>>> 
>>>> The suggestion that first came to my mind was probably worse:
>>>>  BasicClass
> 
> Yeah, I thought of that one too. The idea behind SafeClass is that
>  it's intended to be safe, just as the FreeClass is intended to be
>  free. But IdentifierClass is OK with me.
> 
>> - Safe is an overloaded word. How can we claim to have something 
>> safe with the zillions unicode codepoints. - I prefer 
>> IdentifierClass, but just too long. - suggesting: IdClass.

The length of the term 'IdentifierClass' doesn't bother me, and it's
very clear. Plus, "id class" feels a bit Freudian to me. ;-)

Peter

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


-----BEGIN PGP SIGNATURE-----
Version: GnuPG/MacGPG2 v2.0.18 (Darwin)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
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=26DQ
-----END PGP SIGNATURE-----
_______________________________________________
precis mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/precis

Reply via email to