Author: steve[at]fractalus.com
[Added to the original trac issue at 10.42am, Tuesday, 25th April 2006]

could it be similar to http://www.openstreetmap.org/traces/user/SteveC/2628 in that its a long trace almost entirely in one direction? Is there a massive jump to a spurious point miles away? what's it look like in a viewer like JOSM?

Looks like it's in the DB fine.

Author: miki
[Added to the original trac issue at 12.27pm, Tuesday, 25th April 2006]

Actually it's a somewhat curly combined drive and walk from my home to my town center and back again.

I'm quite new at this (both GPS tracing and OSM) so I haven't yet been tinkering with editor applications but I'll probably do that to verify if there's a point which has "fallen off". Does this really happen?
I'm using an iPaq and bluetooth GPS (BT-48 from Sanav). Using VisualGPSce (http://www.visualgps.net/VisualGPSce/) to record log files and using http://gpsvisualizer.com/ for converting the NMEA log files to GPX.

Actually I just tried to draw the gpx at gpsvisualizer and it doesn't show any points at all when beeing feed the gpx file so its probably a data problem. I'll look into it later and post my findings.

Mikkel

Author: miki
[Added to the original trac issue at 12.55pm, Tuesday, 25th April 2006]

Okay, just grepped a little in the gpx file and found the points below (jump from 8 to 43 long). That'll do it I think ;)

But the timestamp is weird in the gpx too, it was done on the 23. but the dates jump around covering most months of 2006...

Better check the original NMEA file when I get back home...

<trkseg>
  <trkpt lat="55.4645133" lon="43.1500000">
    <course>099.0</course>
    <ele>22.1</ele>
    <hdop>03.70</hdop>
    <sat>5</sat>
    <speed>1.54</speed>
    <time>2006-06-03T08:21:00Z</time>
  </trkpt>
</trkseg>
<trkseg>
  <trkpt lat="55.4645067" lon="8.4518483">
    <course>096.0</course>
    <ele>21.7</ele>
    <hdop>04.40</hdop>
    <sat>5</sat>
    <speed>1.59</speed>
    <time>2006-06-04T00:01:13Z</time>
  </trkpt>
</trkseg>

Just uploaded an edition without this single point so lets see how it looks.

Sorry for the needless fuzz.

The dates puzzle me, though.

Mikkel

Author: erik[at]tiq.com
[Added to the original trac issue at 1.49pm, Sunday, 21st May 2006]

Mikkel did you understand why that point ended up there?