Hi Steeve,
On Fri, Feb 8, 2013 at 8:44 PM, Steeve Ebener < steeve.ebe...@gaia-geosystems.org> wrote: > Dear Eugene,**** > > ** ** > > Thank you very much for your very good feedback.**** > > ** ** > > First of all, I just realized that attachments were not kept in the > discussion list. I have therefore posted here the proposed form for data > collection: > http://www.gaia-geosystems.org/PROJECTS/PHL/HF_fields_for_OSM_040213.doc.* > *** > > ** ** > > Now coming back to your feedback:**** > > ** ** > > **1. **Device used for location health facilities: Any device would > be fine as long as they allow to capture all the associated information > reported in the form (nbr of sat signal, horizontal accuracy, 5 decimals). > Hmmm. 5 decimal places leads to a precision of about 1 meter. Most consumer-grade GPS devices don't have that precision. > **** > > **2. **Use of Bing Map: Did you ever do an assessment of the > horizontal accuracy of the images you can access through this platform? If > yes, what is the maximum horizontal error you have encountered over the > Philippines? If not, I would be interested in discussing a way to estimate > this accuracy before these images are being used for collecting points. > What about the availability of high resolution images in Bing map? Would > you consider it has being equal or better than in Google Earth? > Actually, the aerial imagery found in Bing Maps has the same quality characteristics as that of Google Maps. As you may know, satellite imagery coverage, whether it comes from Google or Bing, is a patchwork of individual satellite images obtained from various satellite companies like DigitalGlobe and GeoEye. Sometimes the position is spot on, sometimes not and it may be off by around 5 meters or more. In some places, Google has higher resolution than Bing. In other places, it's the other way around. Sometimes Google has newer imagery, and in some places, Bing has newer imagery. In some places, Google has imagery where Bing has none and in other places, it's the other way around. Hmmm... now that I think about it, why do you even need an accuracy of 1 meter? Even 10 meters (4 decimal places) is good enough especially for hospitals and big clinics that are bigger than 10x10 meters. I know that it would be good to be as accurate as possible, but if you end up rejecting perfectly valid data especially in remote areas of the country just because a clinic is not located down to the meter, then the effort might be a waste. We don't need to have almost-perfect data on the first try. Even just "good enough" would be OK for most applications I can think of. **** > > **3. **Copyright: thanks for pointing us to the Intellectual > property code of the Philippines. I I downloaded it from here > http://www.aijc.com.ph/PCCF/observatory/pfd%20files/policies/IPR/ipr.pdf but > could not find the sentence you are referring to. Could you please > tell me in which section of the document it is mentioned? Maybe that the > above link does not point to the complete document. > The PDF you downloaded is just Part I of the law. You can find a copy of the complete law here: http://www.lawphil.net/statutes/repacts/ra1997/ra_8293_1997.html Please refer to Section 176. Eugene > **** > > > Thanks once more for your inputs and in advance for your answer on the > above.**** > > ** ** > > Best regards, Steeve**** > > ** ** > > *Steeve Ebener, Ph.D.* > CEO Gaia GeoSystems**** > > P.O. Box 795 – P.C. 114, Muscat - Oman > cell: +968 952 57 526 > email: steeve.ebe...@gaia-geosystems.org**** > > web: www.gaia-geosystems.org **** > > Twitter: @GaiaGeosystems**** > > LinkedIn: http://www.linkedin.com/in/steeveebener**** > > ** ** > > ** > > *From:* Eugene Alvin Villar [mailto:sea...@gmail.com <sea...@gmail.com>] > *Sent:* lundi 4 février 2013 20:04 > *To:* Steeve Ebener > *Cc:* osm-ph; Mark Anthonie Bello; gerald aguinaldo > *Subject:* Re: [talk-ph] Possible collaboration with OSM and DOH to > locate health facilities**** > > ** ** > > Hi Steeve,**** > > We have some interesting issues here...**** > > ** ** > > On Mon, Feb 4, 2013 at 1:27 PM, Steeve Ebener < > steeve.ebe...@gaia-geosystems.org> wrote:**** > > Dear Maning, > > Thank you very much for your email to the group and all the great work you > guys are doing. > > Few complement of information from our side: > - we need to standardize the way location are collected either using GPS > devices or Google Earth/map. The attached form is therefore proposed for > your review/feedback/comments. > - In order to avoid duplication of work, the DOH would provide OSM with > the list of facilities (including DOH codes) for which a coordinate is > missing**** > > ** ** > > If OSM contributors will help locating health facilities by adding data to > the OSM database, then our source will either be any of the following:**** > > 1. GPS device (standalone like Garmin or TomTom, or a mobile app like on > Android or iOS)**** > > 2. By locating on Bing Maps. We cannot use Google Maps or Google Earth to > add data to the OSM database, but we have permission from Microsoft to use > Bing Maps aerial imagery to add data ONLY to the OSM database. Other people > are free to obtain the same data but via OSM's database license.**** > > 3. By extrapolating from the existing OSM data. For example, given an > address of a facility, it may be possible to locate the facility just by > interpreting the address.**** > > ** ** > > - Regarding the data sharing issue. It is now clear that we will have to > keep OSM and DOH geo-location DB separated which means that: > - it will unfortunately not be possible to integrate DOH points in > OSM data**** > > ** ** > > This I find interesting. According to the Intellectual property code of > the Philippines (R.A. 8293), "No copyright shall subsist in any work of the > Government of the Philippines." If the DOH data has been collected by DOH > employees in the course of their official duties, then there is no > copyright. The non-profit clause has been argued to be a separate right > from copyright.**** > > If that is the case, then it *may be* should be possible to include the > pure DOH data into OSM.**** > > **** > > - OSM collected points will be integrated into DOH DB with a clear > mention of the source. I of course can't promise this will happen for sure, > but I will personally work on getting an official acknowledgement from the > DOH regarding OSM contribution to this work. In addition to that, the first > discussion I had seems to indicate that it will be possible for OSM to > mention their contribution to filling the gaps in DOH's DB. Please bare > with me on this as there is still some way to go but I will keep pushing.* > *** > > ** ** > > Have you read the ODbL, which is the open license under which OSM data is > licensed? Certainly, attribution to the OSM contributors is required, but > the share-alike property also needs to be followed.**** > > ** ** > > Regards, > Eugene > > **** > > **** > > Looking forward hearing from the group on your email and the above (I have > registered to the mailing list) and the possibility to work together on > this. > > Best regards, Steeve > > Steeve Ebener, Ph.D. > CEO Gaia GeoSystems > P.O. Box 795 – P.C. 114, Muscat - Oman > cell: +968 952 57 526 > email: steeve.ebe...@gaia-geosystems.org > web: www.gaia-geosystems.org > Twitter: @GaiaGeosystems > LinkedIn: http://www.linkedin.com/in/steeveebener > > One Planet, One System > > P Please don't print this e-mail unless you really need to. Thank you**** > > > > -----Original Message----- > From: maning sambale [mailto:emmanuel.samb...@gmail.com] > Sent: vendredi 1 février 2013 15:03 > To: osm-ph; Steeve Ebener; Mark Anthonie Bello > Subject: Possible collaboration with OSM and DOH to locate health > facilities > > Dear everyone, > > I had a meeting with Department of Health (DOH) IMS division on the > possibility of collaborating with the OSM-PH community. No clear plans yet > on the collaboration but I would like to put forward the discussions in > this list. > > = Overview = > Basically, DOH intends to improve their database for the location of > health facilities in the country. They will use the data on several > information management of the Department. At the moment, they have several > database developed in various projects. > Each db uses a different system but one thing they lack is a uniform > geographic identifier. An example of the db is available in the Unified > Health Management Information System portal [0]. This online portal lists > ~21,000 health facilities all over the country. > > At the moment they were able to geocode ~4,000 locations [2] (~20%). > > = What they need from OSM = > > A couple of ideas that were discussed are: > - Verify the location of the 4,000 geocoded facilities in the existing > OSM data. Right now, there isn't any clear evaluation on the accuracy of > the locations (a process of Q/A is now being developed). > Some facilities are actually in Sulu Sea! Check out the southern part of > Negros island [2]. > - Fill in the gaps. the IMS division have limited funds to individually > verify and take GPS readings of each facility. OSM data can possibly fill > in the gaps. Using the Overpass turbo [3], I was able to check that we > have ~1,100 amenity=hospital ways and nodes. > Note that some maybe duplicates and I have not included other health > facilities like barangay health centers (tagged as amenity=clinic > clinic=public [4]). > - Ask the community to map more health facilities. The only additional > tag they want included is the Health Facility Code (an alphanumeric code > similar to NSCB's PSGC [5]). This code will allow them to link the OSM > geometry into their health facility db. > - Train DOH staff to edit OSM. > - Build some easy to use tools to collect data in the field. > > = Pending issues = > As with most government data, they are available only on a non-commercial > use. Specifically, the DOH generated locations. For the other information > in the online db, they are OK with public access without restrictions. This > will be in conflict with OSM's ODbL. A workaround discussed was to > separate the DOH db with OSM's data points. No public db will be > distributed that integrates both data sources. Is this in compliance with > the ODbL? If not, what are other ways in which we can use OSM data in > relation to DOH's plans. > > Everything I've listed are open to discussions. I included here two > people from DOH to keep them aware of the discussions. (@ Steeve and Mark, > please subscribe to our mailinglist [6] to keep track of the > discussions) > > Regardless of what DOH intends to do, I encourage everyone to include > health facilities in your own list of things to map. > > Thanks! > > --- > [0] http://uhmis1.doh.gov.ph/UnifiedHMIS/ > [1] http://uhmis1.doh.gov.ph/doh_facility/rfacilitieslist.php?cmd=resetall > [2] http://maps.doh.gov.ph/facilities.html > [3] http://overpass-turbo.eu/ > [4] > http://lists.openstreetmap.org/pipermail/talk-ph/2010-March/001872.html > [5] http://www.nscb.gov.ph/activestats/psgc/default.asp > [6] http://lists.openstreetmap.org/listinfo/talk-ph > > > -- > cheers, > maning > ------------------------------------------------------ > "Freedom is still the most radical idea of all" -N.Branden > wiki: http://esambale.wikispaces.com/ > blog: http://epsg4253.wordpress.com/ > ------------------------------------------------------**** > > > _______________________________________________ > talk-ph mailing list > talk-ph@openstreetmap.org > http://lists.openstreetmap.org/listinfo/talk-ph**** > > ** ** >
_______________________________________________ talk-ph mailing list talk-ph@openstreetmap.org http://lists.openstreetmap.org/listinfo/talk-ph
