Book Review: GIS for Web Developers

I'm involved in running the local Ruby Meetup in Chicago, and one of the co-organizers got O'Reilly to send us some book for free. In return, I suppose the deal was to give them some exposure for their books, so I got to review this book.

I had actually ordered the book before I found out about the free copy, so now I have two. So it's on amazon now under wil822 seller if anyone wants to buy the first copy off of me.
With that, here we go:

---

GIS stands for Geographic Information Systems and these technologies existed previously, but were often in the domain of military, scientific, or government applications. Geospatial datasets were either public, but hard to decrypt. Or they were easy to view, but expensive. However, with the advent of Google Earth, NASA World Wind, yahoo maps, and google maps, GIS has come into mainstream awareness and attention.

“Earth materializes, rotating majestically in front of his face. Hiro reaches out and grabs it. He twists it around so he's looking at Oregon. Tells it to get rid of the clouds, and it does, giving him a crystalline view of the mountains and the seashore.” - SnowCrash by Neal Stephenson

Even with the explosion of what's available to us geospatially on the web, it would be naïve to think that the development of GIS has plateaued. There are several complementary technologies that are starting to mature that will develop it even further, such as open mobile phone, ubiquitous wireless connectivity, and open social networks. These are but some of the few reasons to look at GIS as a programmer, as it opens up more possibilities for innovation, creativity, and making useful things.

A word of warning to start: GIS for web developers is not a book telling you how to use Google or Yahoo maps APIs to embed maps in your web application. If you don't need much more than to plop down points on a map, then there's no need to know most of what's in GIS for web developers.

However, if you need to suck down geospatial data from different sources to better provide services for your users, then this book might have something for you. GIS has a host of terminology and difficulties that get abstracted away for you in a consumer application like Google or Yahoo maps. GIS for web developers starts with basic GIS terminology, takes you through spatial databases, washes you in OGC web services, and finally plops you down OGC clients. In addition, it focuses on free (as in free speech, not just free beer) open source solutions.

Like a tour guide for the novice, it starts by introducing various basics, and even tells you where to look for different datasets, how to interpret them, and the difficulties that might arise from it. To start, geospatial datasets aren't exactly widely publicized. They're available, but you'd have to know where to look. For example, the US Census Bureau provides a geocoding database called TIGER, and they also provide geospatial data on the demographics of each county in the United States. Many US government agencies provide geographic data, but you'd have to know where to look.

Even if you know where to look, the file formats of geospatial data are many and varied. The book preps you on various popular file formats and translators to help you along when you encounter this in your searches for geospatial data. In addition, it goes into difficulties that might arise even if you've got the file format straight. For geospatial data, there are many different ways to project a 3D sphere into a 2D plane. Because of that, you have to make sure all your data is in the same projection, or else they won't line up correctly. Throw in the fact that the earth isn't really a sphere, you'll be glad that you had this book to figure out how to make sure everything lines up.

Next, it takes you on a short tour of spatial databases. Up until now, I would have thought that querying, “All points of interest within 20 miles of here” would be something implemented at the application level. However, GIS for web developers shows us this is not the case. Apparently, PostgreSQL already has spatial capabilities built in, and one can have table attributes that indicate location and make queries based on distance.

The rest of the book is devoted to getting your feet wet about servers and client standards by the Open Geospatial Consortium. It shows you how to setup an OGC standard geospatial server implementation called GeoServer, so you can serve up geospatial data in a RESTful way. Then a brief romp through using OpenLayers to embed a map widget, before tying together the odds and ends at the end.

This book is a good start to exploring GIS, but it only scrapes the surface. It suggests the possibilities towards the end about going beyond putting push pins on the earth. There's still a lot of room for possibility to explore. A word on editing, however. On page 60 in chapter 4 on rasters, a whole paragraph was repeated. It was as if someone cut and pasted and hit paste one time too many. That should have been caught.

Overall, it's a good book in terms of its information content, given that you're not just looking to embed Google maps in your web application. It takes you through the terminology of GIS, database concerns, and then through standard OGC web services and clients. By the tech book standards, this one is brief, but that's intended, as this is an introduction. The writing itself isn't spectacular, but it's clear and not a drudgery to read through. If you can stand the occasional corny joke and you're looking to suck down geospatial datasets, then this is the book for you.