[standards-jig] jep-0080 (geoloc)

Matt Mankins mankins at media.mit.edu
Wed Sep 24 16:59:42 UTC 2003


> On Wed, Sep 24, 2003 at 12:25:36AM +0100, milk wrote:

> > what about having an element that describes if the resource's location
> > is static or dynamic? i.e., if the client was wireless/linked to
>
> RalphM said:
> What is the advantage of that? If your location doesn't change, well then
> it doesn't change. Do you have a specific application in mind?


I've seen this differentiation before--static objects are used as points
of reference for dynamic objects, and make the foreground/background 
division easier for applications.

I've heard a scenario (not a particularly good one) go something like: I'm
a traveling salesman and have negotiated a deal with a client and want to
print out the changed contract.  I look at all of the static objects in my
vicinity to find something I recognize as a printer, or failing that, ask
another static object who must know where the printer is.

In this example, the static attribute allows a foreground/background 
differentiation to take place, underscoring the architectural "lived" from 
the fluttering "living," which supposedly reduces "location discovery" 
time.

I'm not convinced that this dramatically reduces location discovery, for 
there are probably a zillion other protocols available for finding the 
nearest printer.

Admittedly, there are various other ways of accomplishing the same thing.  
An "expires" tag in some ways does what a static/dynamic attribute might
strive to do.

Bill Mitchell recently noted that "everything is dynamic, if you have the
patience."

Mankins





More information about the Standards mailing list