Re: [port-srv-reg] Four final points for discussion

[Joe Touch <[e-mail]>]

On 9/16/2010 2:11 AM, Magnus Westerlund wrote:
> Joe Touch skrev 2010-09-08 18:01:
>>
>>
>> On 9/7/2010 4:14 PM, Stuart Cheshire wrote:
>>> On 3 Sep, 2010, at 12:01, Michelle Cotton wrote:
>>>
>>>> I agree with all the changes below.
>>>> Regarding the last point, can the service name aliases for future
>>>> registrations go in the notes column?
>>>>
>>>> Michelle
>>>
>>>
>>> I don't think there will be any future aliases.
>>>
>>> I don't see any reason to be allowing further creation of aliases that
>>> add nothing except being a new name for something else that already exists.
>>
>> The burden falls on the owner of the port. If they want to ask for
>> aliases, e.g., to shift from an old product name to a new one, I can't
>> see why we would care. There is impact, but only on that port anyway.
>> However, I'd restrict it to the owner of the port only.
>>
>> A good example of this would be the STUN/TURN stuff we discussed recently.
>>
>
> I think we should strongly recommend against alias for the first case
> like http and www. However STUN and TURN are not aliases, they are
> compatible services that can co-exist on the same port. Thus I wouldn't
> call them aliases at all.

Any time more than one string maps to the same port number it's a kind 
of an 'alias'. The only way for the client to know the difference is via 
in-band information.

I.e., we would allow aliases for:
	- different services
	- that can be resolved in-band

Unless BOTH of those apply, we would not allow aliases except:

	a) legacy (www, http)
	b) to support changeover to the new namespace

Joe