Re: [ipwave] 118

["Joel M. Halpern" <[e-mail]>]

Your earlier text included the line:

"A MAC address SHOULD be of length 48 decimal."

Even if the SHOULD were not capitalized, that would look like an effort 
in this document to specify the length of a MAC address.  With the 
capitalization, it appears even more strongly to be such an effort.

One of your suggestions removed that text, which would be a good way to 
resolve this problem.  Eric's question is based on the text you sent, 
which is taken directly from the document.

Yours,
Joel

On 4/17/19 10:29 AM, Alexandre Petrescu wrote:
> 
> 
> Le 17/04/2019 à 16:24, Eric Gray a écrit :
>> Adding to what Joel said, why are we attempting to define how long a 
>> MAC address SHOULD be?
> 
> We are attempting to define how long a better privacy Interface ID (not 
> MAC) should be.
> 
> There is a difference between an Interface ID and a MAC address.
> 
> An Interface ID is what to put near a prefix to make an address.  For 
> example, 3 in 192.168.1.3 is an Interface ID.
> 
> The longer IID the better privacy, intuitively speaking.
> 
> Why do you think we try to define how long a MAC address should be?
> 
> Alex
> 
>>
>> -----Original Message-----
>> From: ietf <[e-mail]> On Behalf Of Joel M. Halpern
>> Sent: Wednesday, April 17, 2019 9:39 AM
>> To: Alexandre Petrescu <[e-mail]>
>> Cc: <[e-mail]> <[e-mail]>; IETF Discussion 
>> <[e-mail]>; [e-mail]; 
>> [e-mail]
>> Subject: Re: 118
>>
>> Trimmed.
>>
>> On 4/17/19 3:46 AM, Alexandre Petrescu wrote:
>> ....
>>>
>>> I propose the following:
>>> OLD:
>>>>        A randomized Interface ID has the same characteristics of a
>>>>        randomized MAC address, except the length in bits.  A MAC
>>>>        address SHOULD be of length 48 decimal.  An Interface ID
>>>>        SHOULD be of length 64 decimal for all types of IPv6
>>>>        addresses.  In the particular case of IPv6 link-local
>>>>        addresses, the length of the Interface ID MAY be 118
>>>>        decimal.
>>>
>>> NEW:
>>>>        A randomized Interface ID has the same characteristics of a
>>>>        randomized MAC address, except the length in bits.  A MAC
>>>>        address SHOULD be of length 48 decimal.  An Interface ID
>>>>        SHOULD be of length specified in other documents.
>>>
>>> Do you disagree?
>>
>> "SHOULD be of a length specified in other documents" wihtout any 
>> reference to what the other documents are or how to find them seems 
>> like a recipe for implementor error.  Can we be somewhat more specific?
>>
>> Thank you,
>> Joel
>>
>