I am taking a month-long trip to Tasmania, beginning on the 1st February, and as part of that I intend to take quite a few photos.

In addition to some new lens acquisitions I picked up a Gisteq Phototrackr Lite GPS dongle, which pings GPS satellites every 10 seconds and keeps a log of my geographic position throughout the day.

Because the bundled software for the Phototrackr is absolute rubbish, I decided to investigate ways in which I can extract and convert the raw data directly.  As it turns out, there is a perl script available to do this – I Track You for Linux (iTU4l for short).

Because the script is geared toward a linux environment and takes advantage of a few standard linux applications that aren’t available under windows, I figured – because the majority of my photo apps run under windows – I may have to set up a linux / windows dual boot of my eeepc 701.  So, I spent a bit of time nlite-ing a Windows XP SP3 iso down to a manageable size (480mb install) and tried installing and dual booting with various flavours of linux.  This, while successful, was kind of unwieldy.  Dual booting can be tedious at the best of times.  So, I resolved to try ditching windows entirely and find a linux-only solution.

In the end I went with the old faithful Ubuntu Netbook Remix.  Installation was simple, as is the case with Ubuntu by default.  The default install is a little bulky as it comes with a few productivity applications that I don’t really need.  Openoffice was the major one, and removing it ( sudo apt-get remove openoffice.org-base ) freed up 300mb right away.

I then installed Google’s Chromium (the base project for Chrome) via its Launchpad PPA, as Chromium is very snappy on the eeepc – much faster than Firefox – and takes up less screen real estate.

For RAW processing, the Ubuntu Netbook Remix comes with F-Spot by default, which is okay for managing, cataloging and organising images, but not good for batch conversion (in fact it has no batch conversion ability at all), so I installed ufraw (this stands for Unidentified Flying RAW! – somewhat random) via aptitude for this purpose.  ufraw is bundled with an automatic batch conversion application – named ufraw-batch, funnily enough – which converts Canon RAW into a variety of useful formats including jpg and 16 bit tiff.  One bonus of ufraw is that it comes with a very nice gui for finely tuned conversion of individual RAW images.  I haven’t played with this much, but I assume you can save a profile from it and use it as the basis of a batch conversion as well.

I downloaded iTU4l and its counterpart log conversion program sr2x, copied them to /usr/bin and chmodded them to 755 (making the scripts executable).  This worked pretty much out of the box for me – plugged in the dongle, ran the script with the detection parameter and then the extract + clear memory parameters, and then converted the resulting file to GPX.  It will also export to KML for import into Google Maps or Google Earth, various flavours of csv, and other formats.

To correlate the GPS log to my images, I installed gpscorrelate and also gpscorrelate-gui via aptitude.  The premise of this application is that you point it to a series of images, and also a GPS log, and it will check your photos’ timestamps and correlate them to GPS records in the log from the relevant times.  One caveat here is that it will not write geotagging data to RAW files – in fact it will crash – so it’s important to convert your images before correlating them via this app.

After correlation on some sample images, I tried uploading to both Flickr and Panoramio, which quite happily geotagged the uploads based on the EXIF data.  Mission accomplished.

To conclude, Ubuntu Netbook Remix, plus a few choice Linux applications, provided me with a basic suite to catalog my images, extract gps data from my dongle, correlate the gps data with my image timestamps, and upload the images to several online image websites.  Note that I have 2gb RAM on my eeepc 701 rather than the default 512mb, so if you try similar and haven’t swapped out your RAM yet, your mileage may vary.