[OSM-talk] Form to improve the OSMTracker tool.

[Simon Poole]

The thing is, it is -actually- no longer 2010 and the task of collecting 
data while moving faster than walking speed has been solved. Street-side 
imagery, that is with Mapillary, KartaView and so on, with 
post-recording ML supported analysis so that it is easy to find the 
interesting bits. This is safe, doesn't suffer from all the issues 
pressing any kind of buttons while moving has, and the major platforms 
are well supported in all major editors.

If there is need for development in this area then, IMHO, it should be 
for a multi-platform recording app, that is one app that supports 
uploading to Mapillary, KartaView, GeoVisio and so on

Simon

Am 06.08.2022 um 12:47 schrieb Ilya Zverev:
> Hi Jean, it's great the OSM Tracker would have some development.
>
> I've read the form, and it feels like instead of catering to OSM 
> Tracker's strengths, once again you want to make it like a dozen other 
> apps already existing on the market. Same POI collection and polluting 
> the volunteer notes space with traces and notes.
>
> OSM Tracker is unique, and while currently its usefulness is limited, 
> it can be made into an amazing tool for mappers. You just need to find 
> its niche and dig into it.
>
> As I see it, OSM Tracker is great on recording stuff you see on the 
> way, moving at a speed higher than of a pedestrian: on a bike, or in a 
> car. Otherwise you've got tons of editors that you can take your time 
> with, e.g. StreetComplete. These editor would be better than anything 
> you can hope to make OSM Tracker into.
>
> Now, having established that, look into prior art:
>
> * Surveying plugin for JOSM: 
> https://wiki.openstreetmap.org/wiki/JOSM/Plugins/Surveyor
> * NanoLog, a program for data collection at 60 kmh: 
> https://wiki.openstreetmap.org/wiki/NanoLog
> * OpenSurveyor, a failed app like OSM Tracker: 
> https://vimeo.com/73281384 (the design doc is in Russian, alas)
>
> I'd suggest starting with installing the app, then boarding a bus, and 
> seeing how much can you map at the speed. Which decisions you can and 
> cannot make. Then riding a bike with the app and seeing what you want 
> and won't map on the way. The choice of features changes drastically 
> when there's a deadline for all your decisions.
>
> You will find that sub-menus do not work, that deciding between 
> highway=primary and highway=motorway is not what you do, that most 
> fuel stations are already mapped, while marking parking spots is 
> useless, and that post-processing is unavoidable.
>
> And post-processing would be the most imporant — and tedious — step. 
> Most mapping parties I've help end up with a lot of filled walking 
> papers and GPX traces, but with no map. You need to think how to make 
> post-processing fun. Fast. Like, improving the accuracy of collected 
> waypoints. Making a JOSM plugin or a dedicated iD editor fork. Or a 
> tile layer from a central database to which all the collected traces 
> and points are sent. Choosing data to collect that does not require 
> looking elsewhere for confirmation or detailing. Doing away with audio 
> and photo notes.
>
> It's not 2010, forget sending data to email and uploading to OSM 
> notes. Make it better.
>
> Finally, go into the field again. Collect data. Ask people to collect 
> data: both experienced mappers and novices. Iterate. Make better 
> presets. Map. And map again. Working on a tool for surveying means you 
> need to be surveying every other day. Otherwise this would be another 
> academic work not applicable to real mapping, quickly forgotten.
>
> I hope this helps, and of course I can help test improved versions of 
> the app.
>
> Ilya
>
> On Sat, 6 Aug 2022, at 02:51, Jean Marco Rojas Umana wrote:
>>
>> Hello, I am Jean Marco Rojas, a student at the Technological 
>> Institute of Costa Rica (Public University), in the Computer 
>> Engineering career, I am currently doing professional practice in the 
>> Laboratorio Experimental. 
>> <https://www.tec.ac.cr/grupo-investigacion/laboratorio-experimental/>My 
>> practice consists of proposing and developing an improvement to the 
>> OSMTracker. We have created a very short form to validate the 
>> necessity to implement a functionality to upload tracked traces to a 
>> shared folder. The form has only 7 questions, most of them are “Yes” 
>> or “No” questions, so it will take you less than 5 minutes to 
>> complete. Thank you very much. Your contribution is valuable for 
>> improving the tool.
>>
>> Form link: https://forms.gle/yqgnCWT8E12Dh2L59 
>> <https://forms.gle/yqgnCWT8E12Dh2L59>
>>
>> _______________________________________________
>> talk mailing list
>> talk at openstreetmap.org
>> https://lists.openstreetmap.org/listinfo/talk
>>
>
>
> _______________________________________________
> talk mailing list
> talk at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk/attachments/20220806/bd23c189/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: OpenPGP_signature
Type: application/pgp-signature
Size: 495 bytes
Desc: OpenPGP digital signature
URL: <http://lists.openstreetmap.org/pipermail/talk/attachments/20220806/bd23c189/attachment.sig>