Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

A couple of years ago I started working with drones and writing a bunch of custom software. One of my pet peeves in the drone industry is that a ton of stuff uses “reinvented wheels” that are drone-specific or product-specific instead of just riffing on open standards. Given that, I decided to spend a day or two and do a bit of a survey of the different open standards that are available around kind of the GIS/mapping world, the aviation world, etc.

At this point I didn’t even know about QGIS, but I was happy that instead of reinventing the wheel I already had well-defined file formats etc. to work with. QGIS came a little bit later when I had a bunch of data to analyze. My image metadata wasn’t exactly right for QGIS (and couldn’t have been done that way in real-time anyway), but a 3-line shell script using GDAL converted all of my images into something QGIS could load and draw directly overlaid on top of Google Maps. My flight plans were all done in annotated GeoJSON, which QGIS happily loaded as-is and helpfully split it into the layers I had used under the hood as my data encoding. There were a couple of files that were just plain-old CSV, and QGIS happily ingested those and drew them as points and lines and line segments as appropriate.

It’s phenomenally good. Some other people in this discussion have mentioned that the UI is a bit rough and I can’t disagree with that, although I feel the same way about the commercial tools too. It’s a power tool for sure, but it doesn’t take too long to figure out the quirks and build yourself a workflow.



Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: