Dow Street allegedly wrote on 03 29 2009 4:38 PM:
>> locator       A locator is a name that has topological sensitivity and
>> must
>>        change if the point of attachment changes.  By convention,
>>        a locator refers to layer 3 by default.  It is also
>>        possible to have locators at other layers.  Locators may
>>        have other properties, such as their scope (local or global
>>        (default)) and their lifetime (ephemeral or permanent
>>        (default)).
>>
>> identifier An identifier is the name of an object; identifiers have no
>>        topological sensitivity, and do not change, even if the
>>        object changes its point(s) of attachment within the
>>        network topology.  Identifiers may have other properties,
>>        such as the scope of their uniqueness (local or global
>>        (default)), the probability of their uniqueness
>>        (statistical or absolute (default)), and their lifetime
>>        (ephemeral or permanent (default)).
>>
>> address    An address is a name that is both an interface locator and an
>>        endpoint identifier.
> 
> As a starting point, I think we can say that a system has objects and
> names for those objects.  

and functions that act on those objects and names for the functions.

> A given name could be a locator, an identifier, or both.  Whether it is
> considered to be a locator or identifier depends on *how the name is
> used*.  

I'm so pleased to meet a fellow believer in the wilderness.  Usage is
the primary issue.  However, _intended_ usage can also apply (for
example, if I pass you a message which has a field in it that contains
something I want you to use as a locator, it's a locator).  And some
names can only be usefully used a particular way due to the way they are
constructed or administered.

> A name alone is just a name - it is the binding of name to use
> (i.e. role) that makes it a locator or identifier.  As such, it is quite
> possible that a name can serve various roles, or be used in different
> ways, by different components of the system at the same time. 
> Furthermore, the scope and/or nature of the named object can differ
> between different users of the name.  (Is this the name of an IP
> interface, a stack, a host, etc?)
> 
> So, what makes a name into a locator?
> 
> "A name is defined to be a "locator" when it is used in the context of,
> or perhaps by, a system that has a notion of geography, topology, or
> distance, and the value of the name, combined with a location in the
> system, produces a vector of relevance to that system's geography or
> topology."
> 
> Looking at Tony's definitions above, there is something about the phrase
> "must change if the point of attachment changes" that does not seem
> quite right.  (Still trying to figure out why, though...)

Noel Chiappa allegedly wrote on 03 29 2009 5:13 PM:
>     > From: Dow Street <[email protected]>
> 
>     > So, what makes a name into a locator?
> 
> When it has _structure_ which relates to _where_ the named thing is.
> 
> I.e. MAC is not a locator, even when used in a bridging system, because
> there's no structure in it - MACs are just identifiers for the interface.

Right, it is administered in such a way that it cannot be used
effectively as a locator.  That limits its use ... but if tomorrow I
enforced assignment of my MACs hierarchically and dynamically, suddenly
<Ta Da> they could be locators.  It's not a property of the thing
itself, just how it is used.

>     > "A name is defined to be a "locator" when it is used in the context of,
>     > nor perhaps by, a system that has a notion of geography, topology, or
>     > distance, and the value of the name, combined with a location in the
>     > system, produces a vector of relevance to that system's geography or
>     > topology."
> 
> I _think_ this says the same thing as the above , but in a mathematical way,
> but I'm not sure.
> 
>     > there is something about the phrase "must change if the point of
>     > attachment changes" that does not seem quite right. 
> 
> Well, one thing is that a locator may change, even if the physical location
> the entity is plugged in hasn't changed, if the structure of the network has
> been rearranged (e.g. a chunk of topology has been re-homed, e.g. to a
> different ISP).

I think that falls under "change in topological location".  The topology
changes so the topological location changes, even if the attachment
point hasn't.


_______________________________________________
rrg mailing list
[email protected]
http://www.irtf.org/mailman/listinfo/rrg

Reply via email to