imagico has commented on the following diary entries

Post When Comment
OpenStreetMap active users about 1 month ago

That is in line with the Active contributors per month on:

and No. of active members last 30 days on:

In principle although the choice of a 30 days window is perfectly reasonable as a you have to pick one solution it would be nice to have a more detailed spectrum of the contributor activity w.r.t. frequency - like with one week, 30 days, 3 month, six month, year. Data we have here right now is:

  • one week: ~10k
  • 30 days: ~25k
  • year: ~150k with ~50k recurring

which indicates a significant number of users who contribute regularly but less often than monthly.

Ideally such stats should also include users who contributed only in changeset discussions and with notes - it would be especially interesting to know if these are primarily a domain of otherwise active mappers or if there is a distinct group of users who primarily engage in discussion only and do not perform edits themselves.

A new version of the OSM Edit Report is here! about 2 months ago

Thanks for the replies.

I am aware that the work of the Mapbox data team is quite diverse and also consists of lower volume activities like fixing errors etc. The tool you introduced here however does not reflect this and showing the edit volume without differentiation could give the impression this is what counts most.

The only specific suggestion i would have regarding your data team work is to put weight on local knowledge, that means preferably let mappers work on areas they are familiar with and have them familiarize themselves sufficiently with areas that are less known to them.

One very basic thing Mapbox could do to better support community mapping is to provide capture date metadata with your satellite imagery. This has already been requested several times i think and would enormously improve ability of mappers to properly assess their image sources.

A new version of the OSM Edit Report is here! about 2 months ago

I wonder if this is your primary method to evaluate the work of your mappers and if not what your primary method is and what role this tool plays in your data team controlling.

The shown numbers in the order of 10k-20k changes per day are only achievable with fairly mechanical tracing work - in this case mostly buildings. If you do the math 15k changes in an 8 hour day gives an average of 2 seconds per change or 10 seconds for a simple rectangular building. This might not seem too bad but it essentially contains very little room for mapping any more complex semantic information or doing higher level verification work (like cross checking with different image sources).

In principle i think this kind of mapping work is somewhat questionable in its overall value. The primary data that would be required to automatically acquire this kind of information (high resolution orthoimagery or LIDAR data) might not be readily available right now and existing building outlines for the area might not be available as open data but having this building data in the OSM database - even if valuable for practical applications right now - has relatively little lasting value in the long term compared to classical on-the-ground mapping.

So to get back to the starting point - from the perspective of the OSM community it would be important to evaluate your data team mapping work w.r.t. the long term sustained value of the data for the project. It is perfectly understandable and legitimate if Mapbox also has short term data needs and tasks its mappers to fulfill these but you should always keep in mind that 10k features mapped in that context are of a different inherent value than 10k features mapped by a local mapper walking the streets and mapping addresses, trash bins, fire hydrants, addresses and all kinds of other stuff in addition to the plain building outlines.

Georeferenzierte Bilder aus Videos extrahieren – Low-Cost VIS about 2 months ago

Bei Messzügen gibt es übrigens wohl auch 'ne hochwertigere Variante:


Für die Aufnahme in seitlicher Richtung ließe sich die Bildqualität durch eine lichtstärkere Optik und entsprechend kürzere Belichtungszeiten vermutlich steigern, die Frage ist dabei natürlich in wie fern das für die manuelle Auswertung überhaupt einen Gewinn bringen würde, d.h. ob sich das mehr an Informationen mit vertretbarem Aufwand auswerten ließe.

The history and completeness of OSM 2 months ago

I can't say much about the visual inspection without knowing the exact procedure but there are a lot of things you can do wrong here and introduce bias. Since many countries do not have full high resolution satellite image coverage available i don't even think random sampling is possible.

And you always have other systematic errors when doing assessment based on satellite images. For example in heavily forested areas (like Brazil, Canada, Russia) you underestimate the actual number of smaller roads in rural areas so you overestimate completeness.

Cases of probably quite significant overestimates are for example Libya and Chile.

The history and completeness of OSM 2 months ago

Without knowing the details of the methods used not much can be said here but it seems unlikely you can make such estimates without pretty hairy assumptions regarding the distribution of roads in countries (both spatial and in terms of road types) or the pattern how roads are mapped in OSM.

That being said the results seem to overestimate completeness in many cases, especially for larger countries with limited and localized mapping, probably because you interpret saturation effects in urban road mapping as a sign for overall completeness.

The good thing about this is that overestimation of completeness is going to be much easier to falsify. So if you stay with the 90% completeness estimate you are likely to be proven wrong relatively soon...

The most inefficient way in North America 2 months ago

The Import Guidelines already contain a remark concerning this:

Of course CanVec - the scourge of OSM...

Cleaning up NHD in North Carolina 2 months ago

Yes, JOSM does it correctly, if you do it like JOSM you should be fine.

OpenStreetMap and Humanitarian OpenStreetMap Team Together 2 months ago

That is a very nice video, especially in the beginning where it explains the core of what OpenStreetMap is about, i.e. people mapping their environment and sharing their knowledge of the world with others.

I want to point out however that the narrative that both OSM and HOT are about creating "a free map of the world", which also was mentioned several times in recent discussions in some form, is not really correct. For me as someone who has mapped primarily in those areas in OSM that are probably most distant from this potential goal this seems pretty clear. For OSM the map of the whole world would be the ultimate conclusion when everyone on earth has become a mapper and is contributing his/her knowledge to the OSM database. But this does not make it a goal of the project, OSM is about the process of collecting data and sharing knowledge.

And for HOT the map of the whole world seems to be no goal either, HOT is about generating free map data where map data is needed, either actutely in case of a desaster/crisis or prophylactically where it might be needed in the future. In any case HOT mapping is generally based on data needs by specific interested parties (that is mostly the aid organizations HOT cooperates with). And these parties generally do not have a need for a "map of the world", there are huge parts of the world, both geographically and thematically, they have no interest in at all.

Cleaning up NHD in North Carolina 2 months ago

One problem of NHD as well as other waterbody imports is that river/stream classification is often either off or completely missing and this is often very difficult to assess properly from imagery alone.

Your observations re. accuracy comply with what i experienced with NHD data. This is very variable, both in terms of positional accuracy and age of the data. In some areas NHD data is clearly very old (probably 1950s-1960s).

With your 'data efficiency analysis' - make sure when you apply this worldwide you take into account projection distortion, otherwise you probably end up with very wrong results at high latitudes.

JOSM could by default offer a scale independent simplification (using the node density along the line to set the simplification threshold).

How large are our national contributor communities and how are they developing? 3 months ago

I suspect one major problem of your methodology is that quite a few mappers start off with either a remote mapping changeset (which does not necessarily have to be part of a HOT project etc.) or mapping in a location away from home (during vacation for example). This will probably not significantly affect numbers for countries like Germany or the USA but it will likely overestimate the number of mappers in countries with a low number of mappers.

OpenStreetMap Carto Complexity 3 months ago

There are two important things you analysis misses i think:

  • In addition to Mapbox Streets there are also other styles that use preprocessing. Like OpenTopoMap In fact you could say through the coastlines all styles make some use of external data preprocessing that is based on additional code. This is of course the same for all styles.
  • In addition code complexity is largely influenced by the feature set offered by the underlying software. Styles vary in what versions of the various tools they require for example and if they use custom extensions (like for PostGIS). Quite a lot of the code complexity in osm-carto is there to work around limitations of the capabilities of the software used.
  • many styles use external non-osm data which often essentially means externalizing processing complexity.
#Spotted - 1 3 months ago

Small recommendation: if you show images of non-urban areas for educative purposes it usually helps to include a scale bar.

Average highway node distance between 2 points in OpenStreetMap - September 2015 5 months ago

In general node distance only tells one side of the story since there are fairly straight roads that require few nodes for accurate representation and curved roads that would be extremely inaccurate with the same node density. So it is usually best to also look at the average derivation angle at the nodes, see here for an example for that.

By the way - is that spherical/ellipsoidal distance or in mercator meters?

New road style for the Default map style, the full version - high zoom 6 months ago

General roads look quite fine now. Would be even better with the brighter farmland of course. ;-)

The new pedestrian color you tried looks very close to landuse=residential.

New road style for the Default map style - the full version 6 months ago

I am worried about collision with water.

I don't think that's a problem as long as you keep it bright enough and on the reddish side of blue - water color is quite greenish.

New road style for the Default map style - the full version 6 months ago

For pedestrian areas - you could try something slightly blueish placing it somewhere between residential and retail, like:

pedestrian in blue

New road style for the Default map style - the full version 6 months ago

The weaker shields are better but you should probably make sure the visibility is strictly decreasing with decreasing highway size - currently secondary appears slightly stronger than primary.

And the secondary shield color is fairly close to heath color due to the secondary color already leaning towards green.

New road style for the Default map style - the full version 6 months ago

Looks fairly good to me - two remarks:

  • At z=7/8 where you don't have minor roads in gray you could sightly lighten the railways to avoid them appearing too strong.
  • The secondary road yellow fill looks somewhat extreme at the higher zooms - i hope there is room for making it somewhat more like the old teritary with the lighter farmland color.
New road style for the Default map style - the second version 7 months ago

Ok - will try to explain briefly on the problem of too strong colors.

Your original choice of colors was a selection of moderately bright, moderately strong red-orange-yellow tones. These form a distinct unit within the color palette of the style that is not much used for other elements. This makes them work quite well. Problems arise (as you noticed) primarily where this set of colors is farily close to area colors used elsewhere, most notably things like beach and farmland.

Now when you make the colors stronger, i.e. move each of the colors closer to the color space edge you increase the distance between the different road colors. This means the color palette for the roads no more forms a clear unit within the overall color palette of the style. The perceptual distance between your new motorway red and the new secondary road yellow is so large for example that each of the various road colors is probably closer to a whole bunch of other colors of the style than to the other road colors. All the advantages of moving from the full color red-green-blue scheme to a more compact set of colors are gone.

A good collection of various sources with background information on color design and some specific discussions of rainbow palettes can be found on:

My specific suggestions here would be:

  • stay with your previous choice of colors as a starting point
  • give up the idea of using the same color for high zoom casing and for low zoom fill - this gives you more options to adjust things, in particular for the yellows
  • try tweaking the farmland color. As previously discussed this is tricky since there are several other colors closeby that constrain you.
  • don't test for 'bad eyesight' - sounds unfair but this is about optimizing understandability vs. optimizing readability. The primary reason for people using rainbow palettes is because they think by using all available colors they can transport more information than otherwise. But this is of no use if can't undestand the information. And basic readability is always an artificially constrained problem with interactive maps - if you can't read it you can always zoom in to get a better view.