Kosovo LogisticsPlus Tracing
The Company Logistics Plus in Kosovo as donated millions of gps points from their fleet tracking software.
UPDATE
We have a better splitting of the data from Michael Schulze : please try these : Kosovo_LogisticsPlus_Tracing2
Whatever you do,
DO NOT UPLOAD THE data POINTS to osm!
Only upload street data that you have traced
How to process
NOTE:
When you download these files, save them to your computer, open them with josm. They are strips of data running from left to right across the country.
for example LogisticsPlusPoints_uw.osm.gz
You may need to Work on a few of them at a time to get a street that is running north/south.
Watch this video on how to process the data here : [1]
Watch my original video on processing the large point files here: [2]
The work is :
- get a wiki account [3]
- Pick a file to work on, edit the Person column, put your name in there. For example for me: [[User:H4ck3rmik3|James Michael DuPont]]
- You can leave the Finished column to NO until you have finished it. Then put in :yes
- DO NOT UPLOAD THE POINTS, trace them.
- Load the point osm layer into JOSM
- Create a new layer and download the OSM data for that area into it.
- Trace over the major streets in the point data, for trails that have only one tiny amount of points, look if you can see some streets in the landsat and yahoo wms layers.
- Upload the new streets layer into OSM. Delete the point layer.
- Write the changeset ID into the Table like this: | [http://www.openstreetmap.org/browse/changeset/3751578 changeset]
- Mark the finished column as yes, or put notes about problem in there.
- Please run the validator plugin from josm. I have found lots of problems.
- Have fun!
DO NOT UPLOAD THESE POINTS! THEY ARE FOR TRACING NOT UPLOADING!
WARNING!
DO NOT UPLOAD THESE POINTS! THEY ARE FOR TRACING NOT UPLOADING!
Data
Data B
Data C
Data D
Data E
Data F
Data G
Data H
Data I
Data J
Data K
Data L
Data M
Data N
Data O
Data P
Data Q
Data R
Data S
Data T
Data U
Data V
| File name | File size | Person | Complete | Comment
|
|---|---|---|---|---|
| LogisticsPlusPoints_va.osm.gz | 78309 | User:Andrewpmk | yes | changeset |
| LogisticsPlusPoints_vb.osm.gz | 76496 | User:Andrewpmk | yes | changeset |
| LogisticsPlusPoints_vc.osm.gz | 75580 | User:Andrewpmk | yes | changeset |
| LogisticsPlusPoints_vd.osm.gz | 82683 | User:Andrewpmk | yes | changeset |
| LogisticsPlusPoints_ve.osm.gz | 80437 | User:Andrewpmk | yes | changeset |
| LogisticsPlusPoints_vf.osm.gz | 83363 | User:Andrewpmk | yes | changeset |
| LogisticsPlusPoints_vg.osm.gz | 328 | No useful data in this segment - Adam Geitgey | Done | changeset |
Preprocessing
We have three deliveries of these points : | Files from delivery1 | Files from delivery2 | Files from delivery3
The current task is to split these into work packages.
I am attempting the following :
1. Merge all the files into one big one. cat *.osm > total.txt
2. Sorting them, removing the nodeid for sorting.
grep node total.txt | perl -p -n -e"s;id=\"\-[0-9]+\";;" | sort -u > total.txt.srt
3. Splitting them into work packages.
split -l 10000 total.txt.srt
where N is the line count
4. adding the osm headers... print "<osm version='0.6' generator='h4ck3rm1k3.pl'>\n"; my $id=0; while (<>) { $id--; if (s/<node/<node id=\"$id\" /) { print $_; } else { # warn "$_"; } } print "</osm>\n";
process the files and zip them for x in LogisticsPlusPoints_??; do echo $x; perl finish.pl $x > $x.osm ;done
for x in *.osm; do gzip $x; done
5. uploading to archive.org
Upload is finished : GpsDataFromKosovoDatapointsWorkPackages