Philippines/Data import Arivac Bohol

Arivac donated Bohol data

The research consortium Arivac has been gathering GIS data in Bohol during their medical research. In their effort to help the local community, they decided to release it for public use. The data was collected by driving with GPS units, and resulted in an extensive coverage and several thousands of POI's in more than 60 categories.

Current Map (Zoom here) Evolution (Zoom here)

Preview of the donated data:

Pois	Highways	Comparison of Arivac roads (red) and OSM (blue)

UPDATE

• 20100329 - received basic road data, provided an OSM test file for talk-ph review.
• 20100329 - Created osm user: http://www.openstreetmap.org/user/arivac_bohol
• 20100330 - initial road import (total of 420 km) using option 3.
• 20100331 - initial POI import (1979 nodes).
• 20100331 - Uploaded consolidated GPS tracks
• 20101126 - sent the letter for approval of the ODBL1.0
• 20110117 - data donors OK with the license change.
• 20110118 - Accepted ODBL and CT.

Proposed process

1. Create a separate wikipage and an osm username (arivac_bohol).
2. Matrix of data features and the corresponding osm k=v.
3. Convert a subset to osm format for review purposes.
4. Contact all bohol mappers and the osm-ph list for comments and review. Invite them to join the talk-ph list to continue discussions and options for importing the data.
5. Import the data when there is group consensus.

Proposed tags

Attribution

Tag for all ways and POI Nodes

source:arivac
website:http://hisdu2.sph.uq.edu.au/arivac/

POI tags

Proposed tagging for POIs

code	name	OSM key	OSM value	2nd key	2nd value	other key	other value
1	Barangay health station	amenity	clinic	clinic	public
2	Hospital	amenity	hospital
3	Sari-sari store	shop	convenience
4	School	amenity	school
5	Daycare center	amenity	kindergarten
6	Barangay hall	amenity	public_building
7	Church/Chapel/Worship	amenity	place_of_worship
8	Market	amenity	public_market
9	Shopping mall	shop	mall
10	Plaza	leisure	park
11	Sports field	leisure	sport_centre
12	Doctor's house	amenity	doctors
13	Private clinic	amenity	doctors
14	Pharmacy	amenity	pharmacy
15	Government/Private office	amenity	public_building
16	Hardware/Paints/Construction supply	shop	hardware
17	Videoke/Bar	amenity	bar	videoke	yes
18	Eatery/Restaurant/Bakery/Coffee shop	amenity	fast_food
19	Internet cafe	amenity	cafe	internet_access	yes
20	Pawnshop	shop	pawnshop
21	Beauty parlor/spa	amenity	hair_dresser
22	Car/motorcycle/bicycle shop/metal/car distributors	shop	car_parts (temporary tag)
23	Water refilling station	shop	water_refilling
24	Appliances/Electronics/Mobile phones shop	shop	electronics
25	Hotel/resort/Lodge/Inn	tourism	hotel
26	Purok	place	hamlet
27	Waiting shed	highway?	bus_stop?	shelter	yes	building	shed
28	Municipal office	amenity	townhall
29	Convent/Seminary/Monastery	amenity	place_of_worship	place_of_worship	convent
30	Cemetery	landuse	cemetery
31	Water well	man_made	water_well
32	Bridge	bridge	yes
33	Funeral home	amenity	mortuary
34	Corn/rice mill	shop	corn_rice_mill
35	Reservoir	landuse	reservoir
36	Gas station	amenity	fuel
37	Chicken farm	landuse	farm
38	Prison	amenity	prison
39	Orphanage	amenity	orphanage
40	Fire station	amenity	fire_station
41	Lotto outlet	shop	betting
42	Antique shop	shop	antique
43	Gym / Fitness Center	amenity	fitness_center
44	Bottler's corp	landuse	industrial
45	Tourist spot	tourism	attraction
46	Furniture shop	shop	furniture
47	Laundry shop	shop	laundry
48	NAPOCOR (National Power Corporation)	building	yes
49	Terminal	amenity	bus / jeepney / tricycle / ferry_station
50	Native products shop	shop	native_products
51	Bank	amenity	bank
52	Souvenir shop	shop	souvenir
53	Commercial building	building	commercial
54	Harbor	man_made	pier
55	Photocopier center	shop	copyshop
56	Photo center	shop	photo_studio
57	Tailoring	shop	tailor
58	Ticketing outlet	amenity	vending_machine	vending	public_transport_tickets
59	Lending institution	shop	pawnshop	-> I think this is similar to a pawn shop, not a money-change
60	Bridal gown shop	shop	apparel
61	Funeral home	amenity	mortuary
62	other
63	other
64	other

Sample tagging OSM Tags

amenity:clinic
clinic:public
source:arivac
source:url:http://hisdu2.sph.uq.edu.au/arivac/
FIXME:add names and proper tags

Arivac reference

arivac:name:L-01-606-010-06
arivac:type_id:1.000000
arivac:typetxt:Barangay health station

Roads

The received data don't have any useful attributes that matches the OSM tagging (no names and highway types).

highway:road
source:arivac
source:url:http://hisdu2.sph.uq.edu.au/arivac/
FIXME:verify road type
arivac_ID_ROA_BAS_: some number

Options for import

Option 1 : Create a several osm tiles. Announce the public download link to users. Allow mappers to merge and edit existing data.
Pro: Local users can merge data.
Con: Local users need to know how to import, merge...
Option 2 : One time import of all the data then allow mappers to fix.
Pro: All data is imported directly, arivac user keeps credit.
Con: Duplicated roads are sometimes difficult to remove.
Option 3 : Pre-process data to remove any duplicates from existing OSM data before upload.
Pro: Data is usable directly after import, arivac user keeps credit.
Con: Higher work load before import.

Tools

• Ogr2osm - to convert shapefile to OSM format
• JOSM - to edit OSM tags and conflation with existing OSM data
• GPSBabel - convert tracks to GPX

Roll Call

The collection and transfer of ARIVAC location data to OSM was performed by:

Bohol

• Diozele Sanvictores - Supervision
  • Michelle Andamon
  • Corazon Ardon
  • Sochette Bautista
  • Emilissa Fuerte
  • Marichu Guillena
  • Angelie Karaan
  • Bianca Camela Olalo
  • Jean Rosette Ramao
  • Mae Suaybaguio

Cebu

• Totor (OSM)

Manila

• Leilani Nillos (DOH - Research Institute for Tropical Medicine)
• Emmanuel Sambale (OSM)

Helsinki

• Antti Lehtinen (Aalto University, School of Economics)
• Antti Tanskanen (THL)

Post import tasks

• Some intersections look a bit weird (4 roads join 2 by 2, then a link joins the 2 nodes instead of 1 common node for the 4 segments). Remove the small way (with 2 nodes) and merge the 4 road ways into one node.
• Add road names and proper highway category. Remove the FIXME tag aftewards.
• Add POI names and correct POI tag.
• Remove duplicate POIs.
• Remove non-existing POIs.