Position Ambiguity [was] Re: [aprssig] findu.com....

KC2MMI (Jared) kc2mmi at verizon.net
Sat Jun 10 13:56:12 EDT 2006


James-

<<The issue that I have with KC2MMI's complaints on position is that he wants me
to fix the data APRS World maps are drawn from. >>
  Not so. I don't know how you draw that conclusion, but I do not want or expect
you to fix their data. All that I would ask, in your case, is for you to alter
your presentation of that data, or, in the least, just stick a marginal note on
*your* screen that pointed out the error.
 Now, I don't know how you are drawing and presenting the data. If you are
calling up their map, and overlaying it with one glyph (sprite, icon, whatever
you know it as) for the data point, you can simply use a larger glyph. Instead
of a start that is 10(?) pixels across, use one that is 40 across. Same code for
everything except the glyph you use.
 If you are calling up the complete image as one image from the other source,
then perhaps you can't control it. OK. How about a little note someplace next to
the map that says "The position shown may be off by 1/10th of a mile." (I've
seen that much error, obviously the extent should be investigated.)
 You *can* easily do that, can't you?

<<That data is from the Census
Bureau. If they have money and a staff to put TIGER together, how can I (with
no money and no staff) possible be expected to "fix" their data. I don't even
have the GIS tools to easily modify the TIGER data.
... I'm not a cartographer. The US Census Bureau apparently is.
Submit your changes to them and when they come out with new TIGER data, >>

See, that's another set of misperceptions. I've spoken directly to the TIGER
folks at the Census Bureau. They will be the first ones to tell you that TIGER
is intended for the display of population demographics data, NOT NAVIGATION, and
it should not be used for any navigational purposes. What this means is that
literally, some of their road upates are made by sending out someone in a car
who said "I drove eight tenths of a mile approximately south-southwest on this
new street called....Please enter it that way on the map." Their purpose is to
show you where "123 Main Street" and "256 Broadway" are located, in relation to
the street and block entities, so you can see which block is in which voting
district, and so forth.

Please note that is all data from and for STREET ADDRESSES and not GEOGRAPHIC
data. While the USCB has correlated geographic positioning to their street
information ("geocoding" the street addresses, a word which can have several
slightly different meanings) they only intend the geographic information to be
incidental, so you can tell that "Smallville" is about eight miles away from
"Anytown", and nothing much more.

The USCB are *not* cartographers, they'll tell you that. They limit themselves
to demographic displays, and the rest is done as best they can with the limited
resources they can. They asked me to contact my CongressCritters (my word, not
theirs) to complain about funding, because they are very much aware of these
problems--but they have no budget to correct them.

Every year the USCB does what they can, typically correcting/updating 100
counties in the US. That's all they can fund. It is hard to blame the USCB for
bad data, when the problem is that all of us computer users have misappropriated
their data, and it is as much our error as theirs, for not finding out the
limits beforehand. (You've really really got to dig at the USCB to find out
their "maps" are not intended for traditional use, but for demographic
generalities only.)

<<But tell ya what: I'll happily give you a free week of use of the APRS World
map server.>> As I've said in the past, thank you. Seriously.

<<I can't fix the data. My choices are to present the data as given to me or to
not provide the service at all. I would be happy to add a notice to APRS
World that says it is not to be used for life support applications. Should I
do that?>>
  I have suggested something similiar many times. Not "life support" but
something that simply says the data source is NOT as precise as it appears,
there can and will be some error. I would expect most APRS users, like most
hams, and in turn most of the general population, knows nearly nothing about
mapping datums and error sources, especially about obscure ones like how the
TIGER maps happen. So instead of ignoring the error and allowing it to mislead
people, why not give them the heads-up, if that's all you can do? If you mop a
floor, you put a "wet floor, caution!" sign out. Or, you pay the jury award when
someone slips and falls on it and breaks a hip. (I know some businesses who have
paid those award, and still refuse to put out a sign because it is too much
effort.)


<<APRS World's source code is open source. ....
My asking you to contribute code is a perfectly reasonable thing to
do.>>

I wasn't aware of that. Asking me to contribute code would be reasonable--if I
were a programmer.<G> Being a "programmer once removed", if I may coin a phrase,
can I ask you again whether APRSWorld grabs one screen shot of the map plus the
overlaid position mark (star, glyph, icon, etc.) all from the same source? Or do
you have that option to simply change the size of the star, by using a larger
one?







More information about the aprssig mailing list