[Nml-wg] NML terminology: Port vs Interface

[Martin Swany]

Hi all,

There is a little more background here that I don't think that it's 
been touched on.  I apologize if so.

The NM-WG initially used the term interface in the topology schema.
When the OSCARS/perfSONAR group was trying to harmonize with 
the GN2-JRA3 effort, for what would be the Interdomain Controller
Protocol, this was one of the points discussed.  The JRA3 group
had used port in its design documents.  I remember Mauro Campanella
defending the choice and making many of the same points that Jason
did.  So, as is part of consensus building, we adapted and took this 
back to the NMWG.  

I personally preferred interface when starting out, but was won
over by port.  First, we "lost" the debate.  Next, it's shorter to type
and there is a time-honored tradition of lazy typists. :-) Lastly, I
agree with the arguments for using port and I find it slightly more
accurate when I compare the relevant dictionary entries.

That said, it's not a big enough deal for me to post the dictionary
entries and try to argue through them.  What I will say is that if we 
don't agree and then stick to the agreement, there will always be a 
group who will prefer a different term for the same concept.  As far as 
GENI goes, I don't think that the resource specification discussion is 
far enough along to say that the are settled firmly on interface.  

best,
martin

On Sep 27, 2011, at 9:32 AM, Freek Dijkstra wrote:

> All,
> 
> A few years ago the NML group agreed to describe an object equal to the
> G.800 "Forwarding Point", but name it "Port", after prior usage in the
> NM-WG.
> 
> Recently, it was proposed to rename this object to "Interface", after
> some confusion over the term "Port" in two projects.
> 
> To end this discussion, I ask everyone reading this mail to vote what
> they think is the best term. The options are:
> 
> * Port
> * Interface
> * ForwardingPoint
> * LogicalPort
> * LogicalInterface
> * No discussion, but further discussion on this term
> 
> This is a ranked ballot, so if you don't care about the name, but like
> this discussion to end, you can rank the first five options with high
> priority (rank 1), and the last option (further discussion) as rank 6.
> 
> Please vote at:
> http://www.cs.cornell.edu/w8/~andru/cgi-perl/civs/vote.pl?id=E_74473334af268499&akey=8c6d8b4da822da62
> 
> The voting system has some easy-to-circumvent methods in place to
> prevent people voting multiple times. If you have problems voting, let
> me know.
> 
> Thanks,
> Freek
> _______________________________________________
> nml-wg mailing list
> nml-wg at ogf.org
> http://www.ogf.org/mailman/listinfo/nml-wg