Purpose of terminology I-D

[Scott Michel <scottm@rush.aero.org>]

I may have been a little more caustic than usual when I replied to
IZUMIYAMA Hidetaka's reposonses. However, his reply was pretty
opaque wrt what it was he wanted, other than a redefinition of the
terminology.

We need to have a common language, so that understanding someone else's
I-D is possible. Thus, new terminology should never be necessary, unless
a concept or term is clearly missing.

What we don't want is a huge list of terms with similar meanings. Along
with the terms, I'm willing to concede that we need synonyms, or something
that says "Also known as ____, ____, ...".

I'm attempting to apply Occam's Razor. Keep it simple and flexible. If you
need synonyms, propose them. If there really is something missing, then
explain why it is missing.

Examples:

>  UDL -
>    Uni-directional link. Some kinds of cable net service and a satellite
>    channel are the sample.

We have UDD. I think UDL is a synonym. Or is UDL an uplink? There has to
be a clear seperation between the physical layer (links) and host/routers
which utilize them. (*)

>  Feed -
>    A host or a router which sends packets to UDL.

"Feed" is a sequence of packets which transit through an uplink. This is
the layer seperation issue again.
 
> Receiver -
>   A host or a router which receives packets from the UDL. It can not send
>   packets to the UDL.

AKA downlink. Right?
 
> VIF (Virtual Interface) -
>   An interface which emulate a UDL as a BDL.

Do we really need to talk about how bidirectional traffic is emulated by
UDLNs? When posing the problem, yes, we do need to talk about specific
solutions (encapsulation, etc.) But does it merely suffice to say that
there must be two seperate network paths? (No, there isn't a clear answer
to this -- but at least tell the mailing list WHY we need to add this
terminology.)

WRT to VIPRE: VIPRE is an ATM solution. Should we add ATM terminology to
this I-D?


-scooter


(*) Henning Schulzrinne's proposed penalty for violating layering is
    reading a thick ITU standards manual from cover to cover at one
    sitting, preferrably related ATM signalling. :-)