What about using the term 'link' when referring to traditional virtual/physical 
interfaces. And leave the term 'interface' to indicate a routing system 
interface?

Marcelo


The interface term left alone will likely always be confused with an API 
interface. 
On 2012-11-28, at 7:22 AM, Russ White <[email protected]> wrote:

> 
> How about we push "interface" into an acronym in one case, so we don't
> end up repeating the word itself constantly? Maybe:
> 
> - "interface," used in the traditional sense, as a physical or virtual
> interface that connects a link to a packet processing engine of some type.
> 
> - "RSI," the routing system interface, defined as the bidirectional
> interface into the set of software and hardware elements that control
> the forwarding of packets through a routed (IP) network.
> 
> If we all used "RSI" consistently, then we could just leave "interface"
> where it is, minimizing confusion for readers in the future.
> 
> Thoughts?
> 
> :-)
> 
> Russ
> 
> 
> 
> On 11/27/2012 8:24 PM, Zach Seils (seils) wrote:
>> What about simple "control interface" and "data interface" designations?
>> 
>> Regards,
>> Zach
>> 
>> 
>>> -----Original Message-----
>>> From: [email protected] [mailto:[email protected]] On Behalf Of Noel
>>> Chiappa
>>> Sent: Tuesday, November 27, 2012 5:36 PM
>>> To: [email protected]
>>> Cc: [email protected]
>>> Subject: Re: [i2rs] Progressing the chartering effort
>>> 
>>>> From: "George, Wes" <[email protected]>
>>> 
>>>> We have an unfortunate overloading of the term "interface" here
>>>> ...
>>>> These are not the same sort of interfaces, and their proximity can lead
>>>> to confusion.
>>> 
>>> Good catch. I can't offhand think of a good synonym for either one (which we
>>> could/would always use to refer to that flavour of interface) - can anyone
>>> else?
>>> 
>>> Failing that, we'll have to use an attached modifier (which I suggest be
>>> included in all uses of the word 'interface') - 'physical interface' or
>>> 'traffic-carrying interface' for one, and something like 'control plane
>>> interface' for the other.
>>> 
>>>> Or am I inferring a distinction where none is desired?
>>> 
>>> No, there's definitely a real distinction between i) an interface over which
>>> user data packets are sent, to the next-hop router (or the final
>>> destination), and ii) the interface between two control plane subsystems,
>>> which the two subsystems use to interact with each other.
>>> 
>>>     Noel
>>> _______________________________________________
>>> i2rs mailing list
>>> [email protected]
>>> https://www.ietf.org/mailman/listinfo/i2rs
>> _______________________________________________
>> i2rs mailing list
>> [email protected]
>> https://www.ietf.org/mailman/listinfo/i2rs
>> 
> 
> -- 
> <><
> [email protected]
> [email protected]
> _______________________________________________
> i2rs mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/i2rs

_______________________________________________
i2rs mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/i2rs

Reply via email to