thoughts on spatial tools and practices
I got an interesting e-mail the other day informing me about the “Roads to Rome” project. I don’t normally write items that arrive in my inbox, but this had two major hooks for me. First, I used to do a lot of routing analysis earlier in my career. Second, I am a Roman history buff. With these two factors in play, I couldn’t resist.
We do a lot of tiles for various customers at Zekiah. Tiling is as much art as science and sometimes things go wrong so we have a range of utilities that we use to perform various kids of QA. Because the caches can be large, we usually want to perform a visual QA on the static tiles before pushing them up to wherever they are going to live full-time.
So I’ve been playing with Leaflet a lot lately. It’s become my lightweight mapping library of choice. There’s a lot it doesn’t do so I keep OpenLayers and others in the rotation as well but Leaflet is direct and to the point so I use it when I can.
A while back, I stumbled onto the Leaflet.markercluster project on GitHub, which adds a clustering layer type. I wanted to try it so I revisited my old GISP heat map demo (Silverlight) and decided to rework it. I was happy to finally get a chance to strip out the plug-in, anyway.
A while back, I posted about some experimentation I did with Leaflet and CartoDB in the wake of FOSS4G in Denver. I recently had the chance to go back and update that sample with some spatial queries. At the time of the original post, CartoDB was still in beta and spatial queries didn’t seem to work, despite the fact that the back-end was driven by PostGIS.
Things have been kind of quiet on the blog lately due to things being busy at work. I call that a good problem to have. Since the beginning of the year, I’ve written a a lot of proposals for a mixture of potential customers. Interestingly, I’m seeing a lot more call for “GIS Analyst” work. One trend I’ve noticed, at least in the Federal sector, is that the time between proposal due dates and award announcements seems to be lengthening. That may be an indication of the ongoing flux in funding and organizations try to figure out how to fund their requirements. It will be interesting to see how it shakes out. Of course, it’s good that the opportunities are there in the first place.
One the technical side of things, I’ve been involved in a smattering of things that’s made it hard to roll up one good post. I’m pretty heavily involved in the PIM efforts that my colleague, Barry Schimpf, has been blogging about over on the Zekiah blog.
I’ve become a big fan of Leaflet for putting maps on the web. It gives me most of what I need without much of what I don’t and is fairly easily extended, as shown by the impressive work of Jason Sanford.
A while back, Dave Bouwman blogged about work he and the team at DTS Agile had done extending Leaflet to support ArcGIS Server layers. Given that there are a lot of ArcGIS Servers out there, this is a good thing to have. Thanks to section 4(f) of the Esri Web Services Terms of Use, it’s less useful for use with ArcGIS Online, but that’s probably the topic of another post.
One of the biggest sources of buzz at FOSS4G was CartoDB. It is a hosted solution from Vizzuality that uses PostGIS to allow you to store your spatial data online. I got a beta account a couple of weeks ago but life (i.e. paying work) kept getting in the way but I finally got to play with it recently.
One of the things that intrigued me is that, similar to Google Fusion Tables, CartoDB exposes a SQL interface through a RESTful API (I’m still not sure if the term “API” applies to REST but it’s a convenient shorthand). Essentially, CartoDB exposes PostgreSQL SQL and the spatial SQL extensions of PostGIS. Once your data is loaded, you can query it and return the results as either CartoDB’s JSON syntax, KML or GeoJSON.
With this information, I set out to build a simple application to query property data and display the results on a map in a browser. In addition to CartoDB, I elected to use the Leaflet Javascript library to accomplish the mapping (although I also experimented with OpenLayers). Displaying and styling GeoJSON in Leaflet is very straightforward and this task gave me and excuse to get a little more comfortable with it.