[aprssig] Local Repeater Freq Object Ambiguity.

[Robert Bruninga]

> > Only a few APRS clients do position ambiguity properly.
> 
> "Properly" (mathematically) would be as a rectangle of 0.11,
1.15,
> 11.51, or 69.09 miles.

That is precisely what bothers me about some displays.  To say
that AMBIGUITY must be displayed precisely!
In my opinion it shouldn't.  I prefer a consistent "circle" (or
elipse) to imply ambiguity.

> If the clients plot them as circles or ellipses they're doing
it
> incorrectly, but we've had this discussion before.  Xastir
plots
> such things as rectangles, with the symbol at the center and
lines
> going from the symbol to each corner.  They are very
unmistakable
> from anything else on the map, and we know the symbol is
_somewhere_
> inside that rectangle or along an edge.

That unmistakeable display is good.  That is the most important
aspect, since it unmistakeably displays the symbol like no other
and so there is no opportunity for missinterpretation by someone
viewing the symbol and BOX.  But what about the display when the
person zooms in to a map scale smaller than the size of the box?
I hope then that the symbol disappears since otherwise it is
precisely plotting an ambiguous position and that violates the
integrity of APRS.

The reason we have this disagreement is that by defining
ambiguity as a precise rectangle that exactly matches the
lat/long box of ambiguity, the inference here is that you are
defining a PRECISE ambiguity.  (is the guy 60 feet over the
line?)  That is what I don't like about a rectangle.  Position
ambiguity has nothing to do with precise boundaries of
inprecision.  It is only a method of conveying in any position
packet, the degree of precision (or lack thereof) of the
position and therefore making sure that the receipent does not
attempt to view the position with any greater precision than
that which was sent.

The original APRS defined APRS position ambiguity to be:
1) At large view scales where the SYMBOL is larger than the
ambiguity, then the SYMBOL is displayed normally.
2) At zooms where the SYMBOL becomes smaller than the ambiguity,
then the symbol is no longer displayed (because it implies
PRECISION that does not exist), and only a circle approximating
the ambiguity is displayed and the callsign is displayed within
it.
3) At zooms below the range scale where even the ambiguity
circle is off the screen, then the callsign does not even
appear, nothing remains to imply a precision at that map scale
that does not exist.

I was hoping all software would display the APRS position
ambiguiy the same way so that users would learn a consistent
display of ambiguity and would become familiar with how it
looked across platforms.  Since the XASTIR box is clearly
unimistakeable to see, I cannot object too strongly, but the
precise edges of a precise lat/long box, seem to me to try to
define ambiguity too precisely for what it's inteneded purpose
is.  And it is different from other displays of ambiguity.

Bob, WB4APR