Release of OSMapTuner

Hi OSM-Community!

The research team of Salzburg Research announces the public release of OSMapTuner – a new mobile editor for OpenStreetMap data. OSMapTuner is the first mobile editor that focuses on improving tags and tag values. OSMapTuner is designed to support specific on-site editing use cases which are not supported well by current editors. OSMapTuner is a free App running on Android-based smartphones and Tablet PCs.

For more information go to: http://osmaptuner.salzburgresearch.at/

We appreciate your feedback, bug reporting etc. Please use the OpenStreetMap Forum.

And now: Start tuning the map with OSMapTuner!

The OSMapTuner team

From the FAQ:

So - what can I do with it? Edit tags is only a small part of OSM. Example: I’m standing in front of a POI, but it’s 10 m off on the map. Or it isn’t even on the map at all. I don’t see a good use case when you can edit only tags and nothing else. Moving and create new objects is one of the most important parts.

Btw: Did anybody try Potlatch on an Android tablet?

Well, I like the idea as I’ve actually been hoping for a mobile editor that says “no” to features that don’t work well on mobile devices, I don’t even want to draw ways or create relations with a phone, since I cannot e.g. draw ways accurately without a pointing device anyway. The only “geometry” feature I could imagine working reasonably well on a mobile editor would be placing POIs.

Thanks to aerial imagery, you can do most of the geometry editing geometry at home before you even survey the area. For example, I currently tend to draw building outlines from imagery, print the resulting map, then go out to validate the results of the tracing and write down data (such as house numbers) I later add as tags to the building outlines. Directly adding the tags on the tablet could make this workflow more efficient. So I’m definitely going to try this new app the next time I map outdoors.

Agreed. I’ve tried Vespucci, which can edit geometry, and didn’t find it usable on a smartphone (a tablet might change this). However definitely please add the ability to add and move isolated nodes, such as adding a missing restaurant, and correcting the location of a restaurant. So no editing of ways, just disconnected nodes.

Please seriously consider releasing the code as open source; a handful of the more FOSS/open-dedicated users will refuse to use it, others might try and replicate the functionality in another open source app (causing duplication of efforts), and who knows what improvements you might be missing out on by excluding other developers from contributing. And don’t worry about “prepar[ing] the source code for a release into the public domain”, this can be done later by yourselves or the community (we’re used to ugly code :).

Hey OSMapTuner-Team,

If you want to get even more feedback about your new app I would recommend to create an article about it in the OSM wiki, similar to all other software and apps that can deal with OSM data.

According to a fulltext search, OSMapTuner is not mentioned there with any word.

If doing so, have a look at othet app pages in the wiki how to use the Software2-Template.

Agree with Stephan, so I started this page for you, hoping you will keep them up to date and fill it with more informations for the community for that you worked on this tool :slight_smile:
http://wiki.openstreetmap.org/wiki/OSMapTuner

Without goof on your solution, I’m not sure if the currently way of communication (announcement after the work is done, giving the community only a binary package, providing informations only at third places, using the wrong channels, no feedback, …) will work with an open project. Please don’t misunderstand me, I appreciate every kind of results that is done by the scientific community, but to get a realy useful result, it has to be in full cooperation with the community.

P.S. Please tell your colleagues about http://wiki.openstreetmap.org/wiki/Research It’s hard for me to track the same institutions again and again and …

Thanks for the wiki page. We already filled the page with information. It was definitely one of the next steps for us to create and maintain the page.

Concerning your criticism: Well, with some of the issues you might be right (e.g. the binary package), but others are not really fair.

It’s not true that OSMapTuner was first announced after the work is done (BTW: it’s worth to mention that for us the work is by far not done with the release of the App). We worked on OSMapTuner for 3 years now, did a community workshop with the local OSM community in Austria, had a presentation at the SOTM Europe last year (http://sotm-eu.org/talk?50), had presentations at several conferences, e.g. AGIT 2011 (http://www.agit.at) and LBS 2011 (http://lbs2011.org). We submitted an abstract to FOSSGIS conference 2012 http://www.fossgis.de/konferenz/2012/ but did not get accepted. Since three years we are organizing the “OpenStreetMap Spezialforum” in the context of the annual AGIT conference with the goal to improve communication between the GIS and the OSM community.

Maybe you consider the channels above as the wrong channels. We are now trying to improve communication and to post information to the “right” channels. Although we are not OSM newbies (all of us are active OSM contributors), it’s not always easy for us to find the right channel to the community. We are glad about any suggestions.

That’s definitely not fair. Since we are very interested in valuable feedback, we offer several feedback channels (directly via Google Play, OSM forum, email, etc.) and provide answers in acceptable time. Although it’s not always possible to react within hours since OSMapTuner is only one of our numerous projects, we try our best and try to react within days.

Well, that’s a good point: What do you mean in full cooperation with the community? One of the answers is, that each of us feels as part of the community (I hope you agree with this). OSMapTuner is one of the results of our research work which we decided to contribute to the community. Data contributions are another issue (as already mentioned we are also active mappers in our spare time). As in any community, contributions may be liked or disliked. We are happy about the likes and try to improve our work to minimize the dislikes (although this is not an easy job as any community member has good reasons to like or dislike a contribution). So, OSMapTuner is maybe on the edge of two different communities: the scientific community and the OSM community. This may be challenging, but we are confident that such contributions can give new directions to both communities.

In your eyes, what would be a useful result? Don’t you agree, that any result which brings the OpenStreetMap project forward is a useful result? The release of OSMapTuner should pursue this goal (otherwise we would not have released the App). It’s very difficult to know in advance which direction brings the best results. With OSMapTuner we want to open new perspectives in mobile editing. It is designed to best support some of the mobile editing use cases. We don’t want to substitute any editor, but offer additional opportunities. That answers the question why OSMapTuner has only limited functionality. That’s not a bug, it’s a feature! OSMapTuner can be seen as a kind of experiment. With an easy to use interface we want to reduce the hurdle for OSM contributions and open the project to potential new contributors (in a similar way as http://wheelmap.org). As described on the project’s home page OSMapTuner is also perfectly suited to get to know the OSM project (check which data is already there, check tags and tag values, get to know the OSM wiki, aso.). Maybe it is not the best editor for “power mappers”. However, my preferred mapping style is to edit geometries from aerial imagery with JSOM and afterwards I go out and add the details (e.g. addresses) with OSMapTuner. One of the most valuable characteristics of the OSM project is that any mapper is free to establish an own mapping style. So, let’s have a try and look forward to the things happening. OSMapTuner succeeds if it helps to open new possibilities for contribution.

That’s definitely one of the issues we heavily discussed. By now we have decided to release OSMapTuner for free as binary package (except the changes to the mapsforge library which are publically available on our SVN). We are aware that some members of the OSM community refuse to use closed source tools to edit open data. However, on the other hand we think that most of the community members will be perfectly happy with a free tool and are not willing/able to contribute on a source code level. As research company we have to gain some return on investment (our work has to be financed somehow). Our current strategy is to re-use parts of the source code in commercial projects in order to gain some return on investment to ensure the further development of OSMapTuner in the future. However, we are also thinking about other strategies to finance our work. So, maybe in the future we will be able to release the source code of OSMapTuner to the public. In the meanwhile, anybody is free to decide whether the use of free but closed source software is acceptable for him/her to edit open data. For those not accepting closed source applications, there are several other open source editing options such as Vespucci, JOSM, aso. We understand OpenStreetMap as open project which gives us the opportunity to experiment with new ways of contribution/financing. Some of them will succeed and others fail. If closed source applications fail we might have to change our plans :wink:

We hope this helps to clarify some of the issues raised in previous comments.

So, keep tuning the map with OSMapTuner.

Karl Rehrl for the OSMapTuner Team

So only a field test to get some feedback, which is used later for commercial stuff. Thanks.

Hi Karl, I don’t want to argue and don’t blame on your team. I just tried to point out, why I personally believe that the current situation isn’t that good for both parties.

  1. after work was done:
    Sorry I missed this informations (or just forgot about it). But as I believe I’m very encouraged in the german community, this might be not a good sign.

2.right channels:
If you want to get feedback, you should go where your users are. Forums and mailinglists are great, but this means esp. to stay in the local channels (for me TALK-DE and German subforum). I know this will cost a lot of time for support, but don’t expect that a lot of people will find your post here.

  1. No feedback:
    Sorry but this was my impression. Web2.0 communication is very fast and people jump of, if they have got the impression, that their messages don’t get heard. 5 days is definitly to long in my eyes, but I see the conflict of getting all this stuff done within a usual day (subscriptions and registering timeslots for monitoring might help you).

  2. full cooperation:
    As I tried to pointed out, I think the result isn’t that useful as it might have become, if the community would had the chance to contribute to the idea. You wrote you introduced the idea months ago, so everything is fine. But esp. the lack of an add node function look a bit strange to me, as this is a definitly must have feature, for surveys in the wild. That’s what I understand as full cooperation, to take the community into account and not looking at them from far far away :wink:

  3. useful result:
    Yes, definitly. Sorry if I missed the tone and making you feel, that your work isn’t welcome here. But as said, I strongly believe, that the results would have become better, if you might investigated mobile tools that are already out there, it’s pros and cons and combine them in a new framework. OSMMapTuner isn’t that bad, but on the other hand it’s (currently) even not that good as it could be to add more details to OSM during the survey itself.

Here Feedback via Diaspora