Re: [gpsd-users] Position accuracy of AIS data in gpsd json?

[Sven Geggus]

Gary E. Miller schrieb am Friday, den 10. August um 00:37 Uhr:

> Lost me.  By citation I mean an external specification we can comply
> to.  And I have no idea which feilds you are concerned about.

OK, I will try to explain this from scratch.

On a gpsd client I can either request the data in scaled or unscaled mode.

When using the latter I get longitude and latitude in minutes/10000
regarding the decription at:

http://catb.org/gpsd/AIVDM.html#_types_1_2_and_3_position_report_class_a

Regardless of any real world GPS accuracy minutes/10000 corresponds to a
resolution of 1/600000.0 degrees thus 6 decimal places are needed to express
this resolution.

Now when choosing scaled mode I will get 4 decimal places (%.4f in
gpsd_json.c) thus this resolution is truncated.

Unfortunately this is even below the resolution of a GPS device which would
need %.5f IMO.  However I still think that it should be up to the client to
interpret the accuracy of the result and would expect the digits to
represent the full NMEA resolution rather than a random decision made in the
server code.

This said, it will not matter that much to me as I can change my client to
use unscaled mode and scale the data myself.

Regards

Sven

-- 
"Der wichtigste Aspekt, den Sie vor der Entscheidung für ein Open
Source-Betriebssystem bedenken sollten, ist, dass Sie kein
Windows-Betriebssystem erhalten." (von http://www.dell.de/ubuntu)
/me is address@hidden, http://sven.gegg.us/ on the Web

signature.asc
Description: Digital signature