SpatialTau v2.9 – Geodatabase vs geodatabase

SpatialTau is my weekly newsletter that goes out every Wednesday. The archive shows up in my blog a month after the newsletter is published. If you’d like to subscribe, please do so here.


What does the word “geodatabase” mean to you?  I’m sure most of you answer that it’s Esri’s proprietary spatial data format, either the classic Microsoft Access Personal Geodatabase or the newer folder based File Geodatabase.  But really it also applies to any spatial database, Esri or not.  Spatialite is a geodatabase.  PostGIS is a geodatabase. CouchDB is a geodatabase.  I could go on of course.  But when we talk about geodatabases we sometimes get our wires crossed.

Just last week I was talking with a client and they were using the word geodatabase to describe both File Geodatabases and PostGIS.  Internally they knew what they were talking about but I went back and forth between understanding what exactly we were discussing.  After a good laugh we talked about the concept of spatial databases abstractly and wondered if the word “geodatabase” was the right one to use.  When I search with the term “geodatabase” with Google, my first 3 results are Esri, and then we see a Wikipedia article on spatial database.  Google image search is littered with Esri examples of their Geodatabase with almost no other database being described.  You can see above I refer to Esri’s product at “Geodatabase” (big G) and the generic term as “geodatabase” (little G) but when we talk it is very hard to see the difference.

So what right?  Does it really matter?  I’ve gone over this the past week since my discussion with the client and I’m of the mindset that Esri “owns” the geodatabase term.  If you search my blog you’ll notice I’m pretty good about making sure I use “Geodatabase” and “geodatabase” correctly but honestly I don’t see why I should.  Saying “spatial database” is so much more descriptive than geodatabase and honestly more understtandable to non-GIS users.  I think in the bigger picture of Spatial IT, referring to something correctly generically matters.  I’m going to use Geodatabase (big G) to refer to Esri’s spatial formats and spatial database to describe spatial databases.  I think the clarification matters and defines the difference between a file format and an actual spatial database.  Heck I can even say a Geodatabase is a spatial database and sleep well all night.

SpatialTau v2.8 – Buying Your Own Mapping Company

SpatialTau is my weekly newsletter that goes out every Wednesday. The archive shows up in my blog a month after the newsletter is published. If you’d like to subscribe, please do so here.


deCarta has a soft spot for many of us. We saw them hit highs with Yahoo! and Google and then lows when they were dumped. Just 2 weeks ago I talked with Marc Prioleau about deCarta on my Hangout. Last night though the news hit that Uber was buying deCarta.

Uber, the popular ride-sharing startup, is acquiring the mapping and search startup deCarta for an unspecified amount, Mashable has learned.

The deal, which closes later this week, is for deCarta’s technology and talent. Founded in 1996, the San Jose startup provides a software platform that focuses on location-based features, including mapping, local search and turn-by-turn navigation.

So deCarta is now owned by Uber. Uber told Mashable:
“A lot of the functionality that makes the Uber app so reliable, affordable and seamless is based on mapping technologies,” an Uber spokesperson told Mashable. “With the acquisition of deCarta, we will continue to fine-tune our products and services that rely on maps –- for example UberPOOL, the way we compute ETAs, and others – and make the Uber experience even better for our users.”

Makes total sense right? The whole point we use Uber and similar services is because they get a car to us in a couple minutes. Marc Prioleau wrote up his thoughts on the acquisition and came to the conclusion there is much to like about deCarta for Uber. I personally think deCarta helps the backend of Uber improve. Marc’s totally right about that (read what he wrote, it’s worth it). But on the consumer side, the one you and I see when we use the Uber.app I still don’t think they’ll replace Google Maps with deCarta. Marc has a theory that they may want to have a different look to their maps than what Google and Apple have but I honestly think consumers hate change and having Google as the visualization makes perfect sense for Uber.

We’ve seen companies like Mapbox do custom tile sets for clients but these projects are the exception to the rule. On iOS and Android, Apple and Google control the maps and most developers just use what is given them by default. The backend system though, those that improve how Uber operates clearly will be given more support. That said, it isn’t like Uber fails to get their cars to their customers.

I have no idea what Uber uses today, it could be some Google based application, some custom code created by them or a consultant or maybe even Esri. But having more staff that understand maps completely will only assist Uber in improving their service. Clearly every little thing helps Uber compete against taxis and Lyft so it’s probably money well spent. That and deCarta gets a nice landing spot. Wins all around!

SpatialTau v2.7 — Responsive Mapping and Eye Candy

SpatialTau is my weekly newsletter that goes out every Wednesday. The archive shows up in my blog a month after the newsletter is published. If you’d like to subscribe, please do so here.


I have to get some things off my chest but first just know that most people in Arizona drive around with a lasso in their truck for just such an occasion.

  1. Most people are very familiar with responsive web design.  The concept where a website automatically reformats itself to best fit on the size screen or type of device you have.  Awesome stuff right?  Rarely do I have to pinch and zoom on a website anymore.  Yet I’m still seeing tons of (mostly government) mapping applications fail completely on smartphones and tablets.  The mapping libraries are available to handle things, just need to get people to actually use it.
  2. Apple Maps released an update today.  Much of it is just better support to find things but they did go the eye candy route.  I don’t know about you but I find 3D mapping a huge pain in the rear on mobile devices. Heck I can’t even remember the last time I opened up Google Earth (quick check, nope I don’t have it installed on this MacBook).  Sure the idea that Big Ben has the correct time is pretty cool from a technologist perspective but the user in me just wants to have transit mapping and better traffic results.  That’s how you get me to use the map on my iPhone.
  3. Looks like Apple is also ready to release more details on their smart watch.  I’ve seen some really amazing things with the Android Wear products and one can only think that the Apple Watch will push the envelope more.  Personally I like the idea of my watch telling me which way to turn while walking in a city.  I suspect we’ll see lots more this summer with mapping and smart watches.
  4. Middleware is so dated, except in our line of business.  If there is nothing a GIS developer likes more is inserting multiple levels of severs between an application and a database.  Paul Ramsey said on my hangout that he thinks that PostGIS is about feature complete.  That’s why I think the magic moving forward will happen in the direct visualization of databases in the browser.
  5. Pitchers and catchers have reported.  Spring is here and the boys of summer are working hard.  Don’t worry about me though, it’s an odd year and that means the Giants won’t win the World Series.  Next year though be ready!

SpatialTau v2.6 – When Maps Got Slippy

SpatialTau is my weekly newsletter that goes out every Wednesday. The archive shows up in my blog a month after the newsletter is published. If you’d like to subscribe, please do so here.


It seems like just yesterday but 10 years ago Google Maps was born.
If you hopped in your DeLorean for a trip back to before 2005, you’d remember the days when we were all dependent on paper maps, print-outs, post-its and sometimes even a compass for directions! Getting from point A to B is something we do all day, every day—from finding the fastest way to get to work, to dropping the kids off on a carpool route, to meeting friends for drinks at a new spot—so it should be as easy as possible. That’s why we created Google Maps and why we’ve spent the last 10 years figuring out better ways for you to get around.

Re/code has a great article on the birth and evolution of Google Maps that I encourage you to read.  For us in the industry, the biggest thing we remember is the disruption of how we visual maps on the Internet.  I didn’t start blogging until May of that year (boy, almost 10 for me) but early on there was much discussion about Google Maps (and Google Earth).  Today, most of our mapping libraries mimic Google Maps either with their API or their tile structure or even the look and feel.  The days of panning and then waiting for a map to redraw or the re-center on a map click are over.  Tiling maps is as common as performing buffers on linear features.

“Google Maps like” is a phrase we see al the time on RFPs and marketing materials. Google Maps has so profoundly impacted our visualization work-flows that we almost delineate between BGM (Before Google Maps) and AGM (After Google Maps).  We compare all new mapping applications against Google Maps, in accuracy and in function.  Projects like OpenStreetMap are successful because Google Maps changed how we navigate and discover people and places.  Companies such as Mapbox and CartoDB exist because as a society we want to view information on maps quickly and easily.  Legacy GIS companies such as Esri have pivoted and become web-centric because Google Maps became the visualization method of GIS data.  Legacy GIS companies such as MapInfo and Intergraph have been pushed aside because they couldn’t change to work within this new dynamic.

Before Google Maps we created online maps in VB6, C#, Java and other complicated languages.  Now whole applications are built with nothing but JavaScript (for the best).  Mapping APIs all look and feel like Google Maps.  There are no weird silo methods to create and display mapping data.

​map.data.LoadGeoJson

​Just look at that.  You know what it means and what it does.  The impact of Google Maps is so complete we seem to forget it is even there.

Even outside of Spatial IT we see the impact of Google Maps.  How long will it take to get to work?  Where is the nearest bar?  When does the next bus arrive? How do I get to the airport?  What is the best place to get a taco near me?  These are questions we type in to Google and get map showing us information.  Even though our cars may have some proprietary Navteq navigation system, we prefer to use our smartphones to find out where we are going.

I’ve been thinking about this all week, my professional life has changed so much since 2005 because of one product.  This started as a simple USA centric car navigation application and has become Navteq/Yelp/Yellowpages/Fodor’s/Michelin Guide/Zagat/AAA all in one.  But with the Google Maps API, it becomes a GIS visualization tool that everyone can use.  I can connect it to PostGIS without much effort and display database information that would have take a complex Java/.NET middleware component to handle.

Google Maps is the most disruptive force on GIS that ended up being exactly what we all needed.  I can’t wait to see what we do in the next decade!