[Osmf-talk] Alternative Strategic Plan

Thu May 18 11:48:39 UTC 2023

We could start with a commitment to build a worldwide community as an integral component of our worldwide map and not to let anything, no matter how well intentioned it looks on the surface, detract from that priority.

--
Andrew
________________________________
From: Christopher Beddow <christopher.beddow at gmail.com>
Sent: 18 May 2023 11:40
To: Japheth Osumo <japhethosumo528 at gmail.com>
Cc: OSMF Talk <osmf-talk at openstreetmap.org>
Subject: Re: [Osmf-talk] Alternative Strategic Plan

I often see people writing that there are fewer contributors in countries outside of the EU/US/AU/UK/NZ/Japan etc because they lack the free time.

Certainly there are economic differences, and it limits time to spend on community, but I don't think this explains lack of contributions (as different from community activity).

Google Maps crowdsources a large amount of it's information in the form of reviews, contributed updates to business hours, and so on. But it does not require the large number of users to submit these edits via a compartmentalized editing tool. The users all have a Google account, and readily use Google Maps for navigating, local search, etc.

Restating what I've written before, OSM, especially in these areas where OSM can tend to be a better map, should rely on people who use the map to help submit edits (even just as notes or pending changes that require an experienced validator), instead of hobbyists or NGO related volunteers alone. Organic Maps and other apps: their users have incentive to participate in a user feedback loop.

While there are many hurdles and questions around how to approach these things, my proposition in addition to Steve's approach of laser focus on what to map, is that OSMF, or the community at large (it mostly operates very well without top down decrees), should actively promote use of and support development of projects like Organic Maps that give OSM data an applied use in everyday life.

Further, while Mikel stated that a corporate governance model is not good for a community--this is stated without any evidence and taken as granted. It's certainly work considering people who contribute to the map as shareholders and the board as having a responsibility to being value to shareholders, which could be defined by making the map itself valuable (useful to end users, and known, well marketed) hence increasing the value of the work that volunteer contributors provide. If a corporate governance model can get things done and provide value, and have accountability to shareholders, then what is the downside? Plenty of startups and companies have a corporate governance model and also a bustling community built around the product, even open source ones.

On Thu, May 18, 2023, 10:12 Japheth Osumo <japhethosumo528 at gmail.com<mailto:japhethosumo528 at gmail.com>> wrote:
Hi,

Dave has basically captured most of my thoughts applicable to Kenya. In addition, there are proportionately fairly few active local mappers

On Thu, May 18, 2023 at 9:23 AM Dave <dfjkman at gmail.com<mailto:dfjkman at gmail.com>> wrote:

Hi,

Sorry I am late to this conversation but would like to make a few observations sitting here in Zambia.

  1.  Making the map go blank or roads translucent where address information is missing will mean large parts of the map will go blank in Africa. Even in major cities like Lusaka street level addresses are not used although they do exist in some form, even mappers using smart phone apps would not be able to add any information as there would be no way of knowing what the actual address was. Out of the city centre in the outer suburbs some of the street addresses are based on the subdivisions of older plots and farms so you will get an address like 59 H/ 3777 (sub division H of plot 59 a subdivision of farm 3777), the next house on the street may be 59 E/ 3777. There have been efforts to rationalize this by using simple numbers so in the above example the above addresses may also be addressed 17 and 19 but you will find both forms being used. Even local mappers may be seriously confused. Many businesses will provide location information referencing a well known local landmark and sometimes this may be another business that was well known in the past but no longer exists or may have even changed its name. I am sure the government and local city administrations will have the information and the preferred method of addressing but it is not readily available. It is also being rationalised but this will take some years. The informal residential areas around the centre of Lusaka may have house numbers where there has been an attempt to formalise them, these are known as site and service areas, but the majority of them are pretty haphazard.

This is the situation in the larger cities. In the smaller towns there will be no address information at all.

  2.  Much of the imagery available in Zambia is many years out of date so people are mapping the past situation not the current. In larger cities the imagery is more up to date but as there is a great deal of development, particularly in Lusaka, even more up to date imagery is soon no longer valid. In the more remote areas whole clusters of buildings, usually houses, will disappear between imagery updates, especially as slash and burn farming is practiced and farmers move on to a new area. I have seen mention of using building information to extrapolate population numbers, this will definitely not be anywhere near accurate in Zambia, particularly as many anthills and trees have been mapped as buildings. Also Microsoft building footprint data has been used to map buildings and in Lusaka this has resulted in large advertising billboards being mapped as buildings.

  3.  Inaccurate data in Africa is not confined to OSM, Google Maps also suffers from this and I think is worse than OSM. Many small towns and settlements have the wrong name, the road reference numbers in some places  are plain wrong and Google Maps has somehow removed a major National Park, South Luangwa, which used to appear on their maps. It is not easy to correct information on Google Maps, I have tried and many of my corrections have been rejected, as a result I have given up informing them. Much of this incorrect information gets carried across to OSM as mappers copy the information from Google even though they shouldn’t, I have come across data in OSM giving the source as Google Maps.

Using Google Maps for navigation in Zambia can result in some interesting routes. A recent example was some tourists using Google Maps to navigate to the Lower Zambezi. There are two options for this route, the most straight forward is via the main T2 to Chirundu and then via a fairly rough but passable dirt road through Chiawa, The second is for the more adventurous down the Leopards Hill Road (D152) and then a very rough dirt road (RD481 Google has it as RD491 which is incorrect) requiring a 4x4 vehicle, Google routed the tourists down this route in a Mercedes car not 4x4. When I heard this story, I tried for myself and sure enough that is the route recommended by Google, obviously the state of the road and the type of vehicle required is not taken into account by Google, this is possible with OSM as the road state can be tagged. In fact, the RD481 is easier going from Chiawa to Lusaka than from Lusaka to Chiawa owing to the Mufundeshi river crossing that has a steep bank on one side.

  4.  Both OSM and Google suffer from POIs being inaccurately positioned. I have only been able to correct Google a few times.

These are just a few of my observations.

Cheers,

Dave

From: Graeme Fitzpatrick <graemefitz1 at gmail.com<mailto:graemefitz1 at gmail.com>>
Sent: Thursday, 18 May 2023 7:19 am
To: Steve Coast <steve at stevecoast.com<mailto:steve at stevecoast.com>>
Cc: osmf-talk at openstreetmap.org<mailto:osmf-talk at openstreetmap.org>
Subject: Re: [Osmf-talk] Alternative Strategic Plan

Not the only time something similar has been suggested!

https://github.com/openstreetmap/iD/issues/8919

Thanks

Graeme

On Thu, 18 May 2023 at 14:09, Steve Coast <steve at stevecoast.com<mailto:steve at stevecoast.com>> wrote:

Thanks for this, I think it shows very clearly how features are actually decided.

Best

Steve

________________________________

From: Cj Malone <CjMalone at mail.com<mailto:CjMalone at mail.com>>
Sent: Tuesday, May 16, 2023 3:48 PM
To: osmf-talk at openstreetmap.org<mailto:osmf-talk at openstreetmap.org> <osmf-talk at openstreetmap.org<mailto:osmf-talk at openstreetmap.org>>
Subject: Re: [Osmf-talk] Alternative Strategic Plan

On Sat, 2023-05-13 at 15:38 +0000, Steve Coast wrote:
> c.     Map notes will be turned on by default.

https://github.com/openstreetmap/openstreetmap-website/issues/3971

_______________________________________________
osmf-talk mailing list
osmf-talk at openstreetmap.org<mailto:osmf-talk at openstreetmap.org>
https://lists.openstreetmap.org/listinfo/osmf-talk

_______________________________________________
osmf-talk mailing list
osmf-talk at openstreetmap.org<mailto:osmf-talk at openstreetmap.org>
https://lists.openstreetmap.org/listinfo/osmf-talk

_______________________________________________
osmf-talk mailing list
osmf-talk at openstreetmap.org<mailto:osmf-talk at openstreetmap.org>
https://lists.openstreetmap.org/listinfo/osmf-talk
_______________________________________________
osmf-talk mailing list
osmf-talk at openstreetmap.org<mailto:osmf-talk at openstreetmap.org>
https://lists.openstreetmap.org/listinfo/osmf-talk
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/osmf-talk/attachments/20230518/64809e2e/attachment.htm>