Re: [KEYPROV] SASL stringprep Was: FW: New Version Notification - draft-ietf-keyprov-dskpp-13.txt

It was not my request.

It was a request from the IESG that is changed with worrying about
consistency across the specs.

It does seem better to me to have one mechanism in use than two
different ones for different specs. But it is a material change so I
thought I should notify the group to see if anyone had a reason to do
it differently.

On Tue, Sep 7, 2010 at 2:33 PM, Anders Rundgren
<anders.rundgren@telia.com> wrote:
> Phil,
>
> It seems that your request for SASL stringrep was honored in DSKPP
> so this is not a topic anymore.
>
>> You know of a token where you can input unicode sequences direct?
>
> In SKS/KeyGen2 all string data is assumed to be in UTF-8 which can be
> inputted from any computer and which has a defined byte-representation.
>
> So yes, SKS tokens *do* Unicode although they do not really know
> what Unicode is.
>
> If UTF-8 doesn't work with Chinese, Arabic, or Japanese without an
> extra layer of interpretation it IMO means tha  "XML failed" since it
> by default is supposed to be coded in UTF-8.
>
> Anders
>
>
>>
>>
>>
>> On Mon, Sep 6, 2010 at 12:11 AM, Anders Rundgren
>> <anders.rundgren@telia.com> wrote:
>>>
>>>  Thanx Philip!
>>>
>>> I had never heard about this before.  There is always something new to
>>> learn :-)
>>>
>>> http://www.ietf.org/rfc/rfc4013.txt
>>>
>>> Personally I would reject a scheme where a password in UTF-8 like
>>>
>>>   I<U+00AD>X
>>>
>>> is supposed to be translated to
>>>
>>>   IX
>>>
>>> The motives may be good but I think it creates more problems
>>> than it solves.  For KeyGen2/SKS I will absolutely not adopt this,
>>> that's for sure.  For encrypted passwords it would mean that the
>>> token would do this translation (?) and that's a really horrible idea.
>>>
>>> Anders
>>>
>>>
>>>
>>>
>>> On 2010-09-05 23:40, Phillip Hallam-Baker wrote:
>>>>
>>>> Thanks for doing this Andrea.
>>>>
>>>> One substantive change to the protocol that should be noted is that
>>>> SASL stringprep is now used for UNICODE passwords. I know we discussed
>>>> this but I cannot remember if anyone cared.
>>>>
>>>> Some members of the IESG did care, they thought that consistency was
>>>> important and in the absence of an actual reason not to use SASL
>>>> stringprep, I agree.
>>>>
>>>> Is there anyone who thinks this is an issue?
>>>>
>>>>
>>>> On Fri, Sep 3, 2010 at 1:56 PM,  <andrea.doherty@rsa.com> wrote:
>>>>>
>>>>> -----Original Message-----
>>>>> From: Internet-Draft@ietf.org [mailto:Internet-Draft@ietf.org]
>>>>> Sent: Friday, September 03, 2010 1:45 PM
>>>>> To: keyprov-chairs@tools.ietf.org;
>>>>> draft-ietf-keyprov-dskpp@tools.ietf.org; tim.polk@nist.gov;
>>>>> alexey.melnikov@isode.com
>>>>> Subject: New Version Notification - draft-ietf-keyprov-dskpp-13.txt
>>>>>
>>>>> New version (-13) has been submitted for
>>>>> draft-ietf-keyprov-dskpp-13.txt.
>>>>> http://www.ietf.org/internet-drafts/draft-ietf-keyprov-dskpp-13.txt
>>>>>
>>>>>
>>>>> Diff from previous version:
>>>>> http://tools.ietf.org/rfcdiff?url2=draft-ietf-keyprov-dskpp-13
>>>>>
>>>>> IETF Secretariat.
>>>>>
>>>>> _______________________________________________
>>>>> KEYPROV mailing list
>>>>> KEYPROV@ietf.org
>>>>> https://www.ietf.org/mailman/listinfo/keyprov
>>>>>
>>>>
>>> _______________________________________________
>>> KEYPROV mailing list
>>> KEYPROV@ietf.org
>>> https://www.ietf.org/mailman/listinfo/keyprov
>>>
>>
>>
>>
>
>

-- 
Website: http://hallambaker.com/