The Garmin Fenix 5x is probably the best watch available for creating data like this but there are handheld devices that’re suitable, albeit heavier.

My aim is to eventually have a map or maps with multiple tracklogs displayed so that I can see where I’ve hiked in Canada. This page is an experimental mess so far, and it doesn’t include any hikes to speak of.

If you’re not familiar with OpenStreetMaps, you’re missing out. It’s not only useful for publishing data but it’s freely available to handheld GPS devices, for example you can export maps into a Garmin watch. Using OpenStreetMaps and uploading files to your webserver is arguably the prettiest method of the three mentioned here, but it’s somewhat problematic due to its data usage. It is simple enough to show a few tracklogs and if you go on a vacation for a week or two, you could create a page “My Vacation in Hawaii”, and display a map relevant to said holiday.

You can do something similar with Google maps by going to “my maps” and importing a GPX file on each layer. There are severe limitations with this method too. You can only have up to so many layers and if you think you can circumvent this issue by merging multiple GPX files into one and then uploading that on the one layer, you’d be right–up to a point–but there’s a secondary limitation as well. The maximum file size of a GPX file you can upload is only 5MB. You might hike for 14 hours in a day and you’ll go well past the 5MB file size.

This was created using the google “my maps” page, and it’s easy enough to understand the information. My walks here were very short as my dog came with me, so I was able to include a lot without it going over the data limit.

This is with OpenStreetMaps using a KML file. For some reason, I cannot get the track to change colour.



Google Fusion is arguably the ugliest method but it’s technologically amazing. You can have loooooooads of tracklogs (I have no idea what the limitation is but I’ve tried a 140MB KML file with no troubles) and it works more or less in the same way as Google maps does when displayed on the website. The process towards getting said map to display is hugely different and I will give step by step instructions with pictures and whatnot at some point.

You need Google Earth, the Garmin Connect website, and the Google Fusion website. First you export your tracklog(s) as GPX file(s) from the Garmin website (you may need to set up a safe zone radius in privacy settings, create a second account, befriend yourself and export GPX files from that account to honour the safe zone in the GPX file), you import it (along with any additional GPX files) into Google Earth and export everything you’ve imported as one single KML file. Then simply upload that to the Google Fusion website and “share” it.


The problem with the above method is that it creates a million points on the map. With a lot of tweaking in Google Earth, you can make it look much better. But this is too time consuming if you have hundreds of files to import. I need to find a way to automate the process somewhat. The easiest is to import into Google Maps, then export the file but of course if you do this, you’re then limited by the 5MB file size again.

I have a couple of gripes… One is that when you export a GPX file from your own Garmin account, it does not honour the safe zone settings. You can circumvent this by creating a second account and befriending yourself but it’s a hassle.

My second complaint is that I can’t simply open a FIT file in basecamp, convert it to a KML file and upload straight to a Google Fusion table as there’ll be lots of points. Nonetheless, this is pretty fabulous stuff :).



What do you think? Which is the prettiest in your opinion? Were you aware Google Fusion exists? Google has quite a few amazing products, and they don’t always advertise things that well I think, but it is in a somewhat beta stage.

3 thoughts on “Map”

  1. Google Fusion Tables is amazing for larger datasets, and there seems to be a REST API also for partial automation for coders. With complex/big data Fusion Tables displays the maps much more fluidly than Google My Maps, as the latter tends to bog down and invoke a substantial delay on the first page load with heavier datasets (don’t know what an earth My Maps with its “newer” mapping engine is doing when there is lots of complex data to display, calculating the relations of the POIs or solving the Traveling Salesman Problem before displaying the info on-screen, perhaps?). I have a hunch that the mapping engine in Fusion Tables is the older one and similar to the the “previous” mapping engine of Google My Maps (ca. three or four years ago), since the icons mentioned in the Fusion Tables documentation did contain some of the “original” My Maps icons (from the time before the latest visual upgrade in Google My Maps) and the overall rendering feeling is/was quite similar to the previous version of Google My Maps, too, IIRC.

    However, even with the custom icons enabled and style options tweaked in Fusion Tables, I like to view my map data content on Google My Maps rather than on Fusion Tables – For me, it’s the presentation style, mainly. Also one key difference between these two Google mapping products is the POI details display style: Fusion Tables displays the POI info as a pop-up window and My Maps displays the POI info on a (collapsable) side panel. IMHO this makes My Maps better suited for viewing the content on a small screen (especially when there’s a lot of fields).

    For tracks, one way to circumvent the 5 MB file size limit in My Maps is to filter the track (to make it contain less nodes, gulp) in BaseCamp before exporting.

    To import tracks to Google Fusion Tables from Garmin BaseCamp, I’d suggest a GIS program, such as QGIS, as a stepping stone. Then the workflow would, for instance, be:
    1. BaseCamp: Export a track as a gpx-file.
    2. Import the gpx-file in a GIS program, such as QGIS (usually drag and drop works, too). If asked, which parts of the file are to be imported, track layer contains the path.
    3. Convert the gpx-file (or the selected layer) to kml and export it (if everything goes well, the outputted file will be formatted differently than the BaseCamp kml-version containing at least one element. Also EPSG:4326 is a good choice for CRS). This convert/export/”save as” procedure in some versions might be initiated rather unintuitively simply by right-clicking the track and selecting “Save as…”. The actual conversion/exporting usually doesn’t take much time.
    4. Upload the fresh kml-file to Fusion Tables. (the clear downside of this method is that most of the extra info apart from the coordinate string in the FIT-file will probably be lost.)

    All in all there seems to be at least some manual labor involved, when dealing with maps, tracks and different filetypes.

    1. Thank you! I’ve given it a go. It’s definitely the fastest method, but there’s no pointers at the start and end of the route! That’s a bummer.

Leave a Reply

Your email address will not be published. Required fields are marked *