Author states:<p>>And that’s the point, rules and complexity have completely unknowable downsides. Downsides like the destruction of the whole project.
With each rule and added complexity you make the system less human and less fun. You make it a Computer Scientists rube goldberg machine while sterilizing it of all the joy of life.<p>While too much rules and complexity can certainly be bad, some basic amount of standardization can actually reduce complexity and really doesn't cause a "destruction of the whole project".<p>As a counterpoint, too much flexibility can also increase complexity. For example, without defined rules, 5.6.2022 can mean 5. June 2022 or 6. May 2022. Nor user, nor parser can know for sure what it means, if standard isn't defined. This kind of flexibility certainly isn't fun.<p>Example from OSM wiki for "Key:source:date":<p>> There is no standing recommendation as to the date format to be used. However, the international standard ISO 8601 appears to be followed by 9 of the top 10 values for this tag. The ISO 8601 basic date format is YYYY-MM-DD.
<a href="https://wiki.openstreetmap.org/wiki/Key:source:date" rel="nofollow">https://wiki.openstreetmap.org/wiki/Key:source:date</a><p>Just define some essential standards. It won't lead to destruction of the project!<p>And while you are making breaking changes, please fix the 'way' element. Maps are big. Storing points in ways as 64bit node-ids, while coordinates in nodes are also 64-bit (32bit lon and 32bit lat), just leads to wasted space and wasted processing time. There are billions of these nodes and nearly all of these nodes don't have tags, just coordinates. There is no upside for this level of indirection. And in case tags are needed for a point, this can already be solved with a separate node and a 'relation' element.<p>OSM data format could certainly be improved and it would benefit end users, as better tools/apps could be made more quickly and easily.
The proposed improvements would obsolete a bunch of problems such as broken polygons [1] which happen regularly. They would also make processing OSM more accessible without needing to randomly seek over GBs of node locations just to assemble geometries which takes a significant runtime percentage of osm2pgsql.<p>For me Steve Coast lost his credibility when he joined the closed and proprietary what3words.<p>[1] <a href="https://wiki.openstreetmap.org/wiki/OSM_Inspector/Views/Multipolygons" rel="nofollow">https://wiki.openstreetmap.org/wiki/OSM_Inspector/Views/Mult...</a>
In the first part of the article I was thinking, oh, maybe Steve Coast isn't such a jerk after all.<p>Then I got to the meat of it. Oh dear.<p>As one of the many many people who has had to deal with OSM data, I curse people with this attitude that the mess is somehow desirable or necessary. It's not. There is a long spectrum between totally free form and completely constrained, and OSM's data model is painfully down the wrong end, and causes enormous harm to all kinds of potential reuses of the data.<p>It also causes harm to the people creating data. Try adding bike paths and figuring out what tags are appropriate in your area. Try working out how to tag different kinds of parks, or which sorts of administrative boundaries should be added or how they should be maintained. It puts many people off, me included.<p>Bah.
I think those are the important bits:<p>> The Engineering Working Group (EWG) of the OSMF has “commissioned” (I think that’s OSMF language for paid) a longstanding proponent of rules and complexity to, uh, investigate how to add rules and complexity to OSM.<p>> [...]<p>> Let us pray that the EWG is just throwing Jochen a bone to go play in the corner and stop annoying the grownups.<p>It's a "response" to <a href="https://blog.openstreetmap.org/2022/06/02/announcement-data-model-study/" rel="nofollow">https://blog.openstreetmap.org/2022/06/02/announcement-data-...</a>
> The harder you make it for them to edit, the less volunteers you’ll get.<p>And that is why dedicated area type (rather than representing areas with lines or special relations[0]) could help new mappers and new users of data.<p>There would be very significant transition costs, but maybe it would be overall beneficial.<p>It is possible to have objects that are both area and line at once. Or area according to one tool/map/edtor and line according to another.<p>And many multipolygon relations are in inconsistent state and require manual fixup.<p>Also, complexity of entire area baggage makes explaining things to newbies more complex. You can either try to hide complexity (used by iD in-browser-editor) leaving people hopelessly confused when things are getting complex or present full complexity (JOSM) causing people to be overwhelmed.<p>See <a href="https://wiki.openstreetmap.org/wiki/Area#Tags_implying_area_status" rel="nofollow">https://wiki.openstreetmap.org/wiki/Area#Tags_implying_area_...</a> for a start of a complexity fractal.<p>[0] <a href="https://wiki.openstreetmap.org/wiki/Area" rel="nofollow">https://wiki.openstreetmap.org/wiki/Area</a>
I know it’s nothing to do with the main thrust of the article, but the author fundamentally misrepresents KYC. Know-your-customer is a facet of anti-money laundering and anti-corruption regulation. It has nothing to do with talking to users.
> Let us pray that the EWG is just throwing Jochen a bone to go play in the corner and stop annoying the grownups.<p>That is neither helpful, not useful, nor making me more likely to treat this diatribe more seriously.
I’ve started playing with data from OpenStreetMap. It started with me trying to fetch all the places where I could get water when moving around Copenhagen, which turned out not to be as easy as first envisioned, because OSM seems to have a lot of different ways to categorise available water, which makes sense, OSM and the tagging system isn't there to support only my usecase, and describing my idea doesn't fit 1:1 with the model.<p>I identified the following tags to look out for:<p>amenity=drinking_water, <a href="https://wiki.openstreetmap.org/wiki/Tag:amenity%3Ddrinking_water" rel="nofollow">https://wiki.openstreetmap.org/wiki/Tag:amenity%3Ddrinking_w...</a><p>man_made=water_tap, <a href="https://wiki.openstreetmap.org/wiki/Tag:man_made%3Dwater_tap" rel="nofollow">https://wiki.openstreetmap.org/wiki/Tag:man_made%3Dwater_tap</a><p>amenity=water_point, <a href="https://wiki.openstreetmap.org/wiki/Tag:amenity%3Dwater_point" rel="nofollow">https://wiki.openstreetmap.org/wiki/Tag:amenity%3Dwater_poin...</a><p>drinking_water=*, <a href="https://wiki.openstreetmap.org/wiki/Key:drinking_water" rel="nofollow">https://wiki.openstreetmap.org/wiki/Key:drinking_water</a><p>It's a tough problem to map out the world and describe it, especially when everyone can add or modify the data, but anything that could improve the experience of importing like osm2pgsql would be welcome.
The exact same argument that praises OSM's super flexible tagged node data model should also praise MS Excel for the number of things that can be achieved in the world of business with just a grid of boxes.<p>Both have been hugely successful, and both have the same pile of downsides.
The claim that a dataset with billions of users has only <i>dozens</i> of people able/interested in doing data processing on it is a damning admission that the format is too hard to deal with
The author is ranting a lot about OSMF's recent decisions, gives all kinds of reasons why they will undoubtedly lead to horrible consequences and grants sage advice what should have been done instead.<p>The only thing I'm missing is any indication that OSMF's course actually did cause any problems in reality.
What happened to the title? It used to be “In Defense of OpenStreetMap's Data Model”, which is the literal blog post title. Someone has now <i>changed it</i> to the boring-sounding “OpenStreetMap's Data Model”, probably resulting in fewer clicks.
Just for context, here is the current OSM data model in a nutshell:<p>There are three data types - nodes, ways and relations. All of the three can have any number of "tags" (i.e. a map<string, string>), which define their semantic meaning. For example, a way with `barrier=fence` is a fence. This stuff is documented in the openstreetmap wiki.<p>A node is a point with a longitude and a latitude.<p>A way is a sequence of nodes.<p>A relation is a collections of any number of nodes, ways or other relations. Each member of this collection can be assigned a "role" (string). Again, the semantics of what each role means is documented in the openstreetmap wiki.<p>To modify data, simply new versions of the edited data are uploaded via the API.<p>---<p>The most prominent point that stands out here is that only nodes have actual geometry.<p>This means that...<p>1. to get the geometry of a way (e.g a building, a road, a landuse, ...), data users first need to get the locations of all the nodes the way references. For relations, it is even one more step.<p>2. in order to edit the course of a way, editors actually edit the location of the nodes of which the way consists of, not the way itself. This means (amongst other things) that the VCS history of that way does not contain such changes
I have been looking into improving kerb and traffic_signals data for some bboxen. It is daunting, and I figure I need to work backwards - try to find out what the accessibility map apps look for and use those pairs. If I know what target I am shooting for I guess it will be alright. This is like my first week looking into this and I hope to find these targets soon.
> The harder you make it for them to edit, the less volunteers you’ll get.<p>I'm not sure I like OSMs obsession with the data <i>model</i>. I guess this has to do with its business model, but then let's not pretend the product manager is tasked to optimize for end-users.<p>I'd like it to focus on the UI. The easier it is to input a geographic thingie and the easier it is to visualize the geographic thingie, the better OSM both for users and volunteers.<p>Two issues to strengthen my point:<p>1. The osm-tag mailing list regularly discusses how tags are visualized in various renderers when recommending which to choose.<p>2. Quick mobile-based correction are nearly impossible with OSMAnd. I'd love to take a picture and write a quick note like "speed limit changed", so that someone (perhaps a bot) can pick this up and update the data model. Same with restaurant opening hours. Or various POIs.
So, come up with an improved format that has/enforces various 'rules' and also provide a conversion program for moving between the new/old format, as desired.<p>Slowly deprecate the nastiest parts of the old, as people get used to the new format.
There's an amusing paradox I've seen many times amongst GIS managers - their complaints about how dreadful OSM data is are pretty much the direct opposite of their enthusiasm to use it!
Heh, their data model is 99% of the reason why I don't use OSM. It's scattered all over the place with <i>so many tables</i>! It's such a nice project, but <i>damn</i> is it impossible to work with programmatically, let alone poke around it to discover what's all in there.