Re: [Geopriv] Long review of draft-ietf-geopriv-loc-filters-07

["Hannes Tschofenig" <Hannes.Tschofenig at gmx.net>]

Exactly. 

The current specification settles for the 'by' only and does not require you to implement anything else because that was the requirement in the previous spec. 

I tend to agree that defining a new purpose-built filter would probably more powerful. 

-------- Original-Nachricht --------
> Datum: Wed, 4 Nov 2009 12:52:54 +0800
> Von: "Thomson, Martin" <Martin.Thomson at andrew.com>
> An: Hannes Tschofenig <Hannes.Tschofenig at gmx.net>, \'Brian Rosen\' <br at brianrosen.net>, \'James Polk\' <jmpolk at cisco.com>, "geopriv at ietf.org" <geopriv at ietf.org>
> Betreff: RE: [Geopriv] Long review of draft-ietf-geopriv-loc-filters-07

> RFC 4661 allows you to request a notification if a value changes _to_ a
> value, _from_ a value, or _by_ a certain amount.  For speed, you could
> specify <changed to="360">  to request notification if the speed changes to 100
> km/h, however there's a good chance that it goes straight past the value you
> are interested in.
> 
> It's impossible to specify every possible target value for something
> continuous like speed or position.  You could cover 341, 342, 343, ..., 368,
> 369, 370, but when the value comes up at 364.3, you still miss the
> notification.
> 
> The <changed by="x"> element can be used for most use cases, although some
> complications are inevitable.  The advantage is that this is pretty simple
> and, for those with a 4661 implementation, zero cost.  Think of it as an
> opportunity for other specifications to define better filters.
> 
> --Martin
> 
> > -----Original Message-----
> > From: geopriv-bounces at ietf.org [mailto:geopriv-bounces at ietf.org] On
> > Behalf Of Hannes Tschofenig
> > Sent: Wednesday, 4 November 2009 11:30 AM
> > To: 'Brian Rosen'; 'James Polk'; geopriv at ietf.org
> > Subject: Re: [Geopriv] Long review of draft-ietf-geopriv-loc-filters-07
> > 
> > Hi James,
> > 
> > Thanks for your detailed review. I am in the process of updating it and
> > have
> > comments here and there.
> > 
> > Here is one:
> > 
> > Draft Text:
> > 
> > --------
> > 
> > <?xml version="1.0" encoding="UTF-8"?>
> > <filter-set xmlns="urn:ietf:params:xml:ns:simple-filter">
> >     <ns-bindings>
> >         <ns-binding prefix="dyn"
> >             urn="urn:ietf:params:xml:schema:pidf:dynamic"/>
> >     </ns-bindings>
> >     <filter id="123" uri="sip:presentity at example.com">
> >         <trigger>
> >             <changed by="3"> [<---James]
> >               //dyn:speed
> >             </changed>
> >         </trigger>
> >     </filter>
> > </filter-set>
> > 
> > ------
> > 
> > [James]
> > 
> > I get the XML,
> > but I don't see what the speed now is of
> > the device/target?
> > What if this device is moving at 200kmph,
> > where would that be placed in the XML?
> > Of, is this left for the application at the
> > subscriber to calculate?
> > If the latter is true, something needs to be
> > said about this expectation.
> > 
> > ------
> > 
> > The filter would not tell you what the current speed is.
> > 
> > The current document only supports 'by' and hence you cannot say
> > something
> > like 'tell me when he goes faster than 100kmph'.
> > 
> > Ciao
> > Hannes
> > 
> > _______________________________________________
> > Geopriv mailing list
> > Geopriv at ietf.org
> > https://www.ietf.org/mailman/listinfo/geopriv
>