Could we somehow focus a little bit more on the end user?

I’m a OSM contributor since nearly two years now, I’ve found it rather easy to get some mapping Software onto my Java-Phone, doing some mapping by bike, uploading the GPX, edit the map via Potlach and JOSM, and done some other stuff. The wiki is a great help on all these topics.

Now that the maps here in Germany are pretty much complete and accurate, definitely more so than Google Maps (I know there are other map services, too, but to make it short I will always refer to google in the rest of the post), I’m actually starting to use OSM as an end user - and try to spread the word so that others use it, too. But I often find myself overstrained, because there is very few documentation for the end user - or I just don’t know where to find it. So I’ve collected some issues that IMHO need fixing before I can truely recomend OSM to everyone.

First, something about what is already there:

  1. The map at has four base maps: Mapnik, Osmarender, Cyclemap and No Name, only one of which (do I need to say that I meand the Cycle Map?) has a self-descriptive name to anyone new to OSM.

  2. There are two overlays available, Maplint and data. Both names are not very descriptive, and Maplint is not very useful for the end user.

  3. The same goes for the 6 tabs at the Top: Map and Export may be interesting to everyone, the rest is just for geeks and nerds like me and you.

  4. The search on the left side is slow and rather picky about which kind of input it accepts. I know, this is a whole new issue in itself and porbably isn’t easy to improve, but for most users the search will be the main entry point to navigate somewhere in the map

And now about those things missing on

  1. There are many other renderings, like openpistemap, openseamap, openarealmap (well, not actually a rendering, i guess) and de-public-transport. (The last one I once saw in AndNav2, a routing software for Android Phones, and it is really cool, showing all the Trains, busses, Trams, Metros, etc. but I was not able to find it anywhere on the net, so i need to use my phone to look at that map) And u guess there are many more renderings that even I am not aware if. I think many of those layers should be included in the layer dropdown on, along with some desription (like: “Mapnik - the deault map view on OSM, suited for almost any use” and “Cycle Map” - Highlights cycle ways and other bike-relevant amenities")

  2. The wiki is not very descriptive about viewing the map. There’s a big “show me the map” that links to, and plenty of stuff for mappers, developers and press. I know there once was a page listing the possibilities to just look at the maps, but i can’t find it any more.

  3. does not work very well on mobile phone browsers. For all those phones, installable applications exist to show those maps. When a phone browser loads the homepage, they should be given the choice: “Click here for a list of mobile phone software that displays OSM maps, or click here to continue to the desktop version of OSM.” Even if the page is loaded by a desktop browser, there might be a hint for the user like “OSM on your mobile phone - click here to learn more”. In these days, this is a major “selling” point!

  4. There are cool services like It works very well for me, so I think this should be linked from the homepage or even directly embedded. While I have only minor problems with its usability, other users told me that they did not understnad the interface. People LOVE routing, so this is a must-have feature for OSM and proves that it is possible - so why do not offer it on

  5. Web developers tend to embed google maps into their pages. Openstreetmap already makes it really easy via the export tab, and I hope this will be used a lot. I also noticed the nixe opportunity to position a marker by mouseclick, this one is really cool. But those developers want a JS Api to embed other, site-specific stuff into their maps. I’m retty sure those APIs exist, but couldn’t find those. There should be a link right where outputs the html-igrame-code.

  6. If you really want end users to become contributors in the end, then give them a smooth way to climb up the ladder. Yeah, there is, and it is easy to use, but why isn’t it integrated with If a user sees bug, how does he know that he should navigate to openstreetbugs? (Btw., if he is not into software, does he really know that “bug” is not an animal in this context?) And yes, there is an edit tab that the user can click, and Potlatch is easy to use compared to some off-the-shelf geo software on the market, but clicking on the edit tab still will scare my grandmother to death. It’s not THAT easy.

I think the usual developement of from a google user to an osm contributor (if you believe in that kind of event) should be like that:

use Google Maps a lot and get used to it → try openstreetmap and be able to use it instantly → not miss any import feature that google offers → see that the data is nearly complete → pherhaps find even more cool features (like the export tab) that gogole does NOT have → understand that OSM is free and built by people like themselves → understand that they can help, too > report a few bugs via Open Street Bugs (which should not involve navigating to another domain) → pherhaps someday try to make minor edits him/herself, via potlach, possibly after watching a 5 minute video tutorial on how to do that → become a full contributor (GPS unit, JOSM, etc) sometime later.

Well, that’s my opinion on what is still missing to make OSM an alternative for the masses. I’m curious if you have any other ideas on that.

with best regards,
Brian Schimmel

I think this is a good coherent summary of major OSM usability issues. I remember a few months ago being completely perplexed by a map with lots of funny writing all over it: I’d accidentally gone to the Maplint layer! It was probably 3-4 weeks later that I realised what had happened.

My father is making use of OSM, but I noticed that he was printing from the web-browser not using the Export tab. I’m sure that he is not alone in expecting this functionality to be under an option “Print”. So a big PRINT option would be more obvious than EXPORT, and may well encourage more usage.

Your post is a good list of things that would be required to make more attractive for end users. However, is - at least currently - not intended for the end user. It’s primarily targeted at potential contributors, current contributors and developers. It’s supposed to show off our data so people think “cool, I can use this to create $some_innovative_application” that end users will then use. It’s not the end user application itself.

When this issue comes up on the mailing list, there is usually a strong opinion that the OpenStreetMap project should focus on providing data, not software - though, of course, many OSM contributors develop applications, too. Common arguments for this include

  • end user applications must be much more reliable than the services we offer
  • bandwidth and processing to serve all those end users is expensive
  • it further emphasizes the “one true map”, thus people will be even more inclined to map for the renderer, undermining the freedom of deciding what to map
  • the rendering needs of end users (rather simple, understandable) conflict with the ones of mappers (want to see everything in the map, even if it gets crowded)
  • we don’t want to prevent the creation of dedicated companies and open source projects using our data for end user applications (as these are much more suited for the task)
  • Google Maps clones are the least interesting way of using OpenStreetMap, you could do the same with Google’s Map Maker

There even have been suggestions to completely remove the map from our entry page and replace it with an overview of projects using OSM data.

Hi Tordanik

Is there already an overview page listing the these projects?

Dave F.

I don’t know any comprehensive list. We have a wiki category for Software, a list of sites using OSM maps for directions or the like (DE:OSM_Internet_Links, German), but nobody seems to have created an overview of major OSM projects/users yet. Some partial overviews were posted on the mailing lists some time ago, but I can’t find them right now.

are you looking for www.ö

Tordanik is spot on in saying “ is - at least currently - not intended for the end user”.

The main problem is that we don’t have enough developers: or, at least, not enough who want to work on the stuff you mention. Sure, I agree with lots of your suggestions, particularly the one about integrating OSB. But I don’t have any more spare time to do them. So unless you can pay to employ some developers, wave a magic wand to summon them from somewhere, or do it yourself, it ain’t gonna happen.