Some words of caution. As others have pointed out, the (clusters of)
satellites tend to be spaced at intervals of about 3 degrees around the
Clarke Belt (the usual term for the geo-stationary equatorial orbit; BTW
*not* the Ecliptic, that's something else entirely). This doesn't leave
much room for error, and potentially there are a number of sources of
error in aligning a dish. While any one on its own may not be too much
of a problem, combined with others, it could be, so it pays to be aware
of such errors, if only to be able to know in any one particular case
either that they're not a problem, or how to eliminate them if they are.
It is a given that there will be some error in aligning a dish, because
no such physical process is ever done perfectly. The pros here will
probably have their own opinions based on their greater experience, but
based on mine I suspect that unless the dish adjustment mechanism is
particularly good or particular naff, about one degree may result from
that alone.
It's a long time since I used Dishpointer, but it works essentially to
the way mine did when it was working ...
There is often an offset between the overhead imagery overlay and the
underlying map, see the example page on my site linked previously, and
this could lead to quite significant error. If when you examine the map
you find that the map of the streets doesn't exactly coincide with their
position in the imagery, then you have two problems:
1) Which is more likely correct? I suspect most probably the map.
2) How much is it out, and how to compensate for it?
Further, the maths is done assuming the earth is a sphere, but it's not,
and, from memory, this introduces an error of up to half a degree at
mid-latitudes for satellites that are at mid-range azimuths east or west
of the location's meridian, reducing to zero error for satellites that
are over the meridian.
With Dishpointer there is potentially a third source of error, from the
use of photographs. Right from the first, I suspected that this
functionality was a gimmick of very limited real life usefulness, so
never tried to copy it. My reasoning is simply this: a photo's scale
depends on the focal length of the lens, focussing, etc, which vary from
camera to camera, and further the photo may not have been taken quite in
the right direction or pointing truly horizontally. If Dishpointer
don't know accurately the photo's scale, horizontality, and direction,
how do they know where to annotate markings on it?