OpenGeo

As egregious as this self-horn-tooting is, I have to post about how enjoyable I have found using the OpenGeo Suite. It might sound odd, coming from an OpenGeo person, to “discover” the Suite at this late date but my personal reality is that I spend much of my time with PostGIS, and until recently only did short testing passes over the Suite.

But this last week, I have been revising a workshop that I will be giving with Steven Citron-Pousty at Where 2.0 this month, possibly at GeoWeb in the summer (and on my own at the Minnesota GIS/LIS conference in the fall). And included in that revision is using the OpenGeo Suite instead of vanilla GeoServer.

Among the things I’ve really enjoyed are:

• How easy the install process is. No messing around installing JDKs or JAI extension libraries.
• How clean the new UI is. Much more sensible names for things and less clicky workflow.
• The instant gratification of the Suite tools. Installer creates datastore and layers and associated styles in one go. Styler writes new SLD with a click click click.
• The little On/Off button in the Dashboard. For some reason I really like that.

The net effect of all the cleanups is I’ve removed about 6 screens of configuration boilerplate from the workshop. More time to talk concepts and scare the students with JavaScript examples!

I’m looking forward to this year’s Where 2.0 in San Jose coming up in three weeks! Where is always a different mix of folks from a usual GIS show, and the Silicon Valley vibe is something you can only get… well, in Silicon Valley.  I am going to be teaching a workshop with Steve Citron-Pousty on the open source geospatial stack, using our own OpenGeo Suite for a big part of the software we show.

If you’re coming to Where 2.0 and want to talk about OpenGeo in general or PostGIS in particular, please let me know! Either drop me an email or hit my Where 2.0 profile.

In preparing some data for our next round of training courses, I spent a fair amount of time today processing and cleaning the US Federal Electoral Commission (FEC) database for 2008. The FEC is extremely good about releasing their data, even though it looks like they have to dump it out of a very old database system.

I processed the three main files, and then converted the associated code tables into side tables, so the whole thing is pretty self-contained and hopefully self-explanatory. I had originally hoped to fully replicate something like the FundRace site from 2008, but since the FEC data only has zip-code as a location entity, that is not going to happen this time around. I assume the FundRace folks also had access to a nationwide telephone directory or some other way of taking name and zip-code and using that to leverage out an actual street address.

If you are interested in playing with the FEC data and don’t feel like spending a couple hours mucking about in Perl to get it into tables, I’ve placed a PostgreSQL dump file online. Candidates are linked to individuals via committees. The FEC model has a lot of complexity hiding in it, with some committees not associated with candidates, and so on, so using the data correctly will probably require a little care.

Update: The FundRace folks did in fact only use FEC data, the trick is, they used the original filings, rather than the FEC database dump. The original filings include a street address for each donation.

One of our missions at OpenGeo is to create an ecosystem where we can develop the best possible software while growing and investing in the community. We see this as a self-reinforcing cycle; as more people use our software, the more we are able to grow our business and devote more energy into improving the software, thus making it more appealing to a wider audience.

But software is only successful if users understand and use its features. And while one of our aims is to make software easy to use, sometimes you have questions that just can’t be answered by a documentation page. Even in this day of everything-over-the-web, there is still a need for answers from a real person, in a real-life classroom setting.

So, OpenGeo is now offering full-day training courses on the components of the OpenGeo Suite. First and foremost, we offer an Introduction to the OpenGeo Suite, which gives a full overview of PostGIS, GeoServer, GeoWebCache, OpenLayers, GeoExt, along with the applications GeoExplorer and Styler. We also have more in-depth full-day introductory courses on GeoServer, PostGIS, and OpenLayers.

These courses are open to all, and are designed to be accessible to a wide audience. Also, if you are one of our OpenGeo Suite Enterprise clients (Professional, Platform, or Strategic Editions), training days are already included in your package.

Below is our schedule for the next few months. All courses are held in our office in New York City, although we do offer on-site training as well.

• Thursday, March 4, 2010 – Introduction to the OpenGeo Suite
• Thursday, March 25, 2010 – Introduction to GeoServer
• Thursday, April 8, 2010 – Introduction to PostGIS
• Thursday, May 6, 2010 – Introduction to OpenLayers
• Thursday, June 3, 2010 – Introduction to the OpenGeo Suite
• Tuesday, June 22, 2010 – Introduction to PostGIS
• Wednesday, June 23, 2010 – Introduction to GeoServer
• Thursday, June 24, 2010 – Introduction to OpenLayers
• Thursday, July 8, 2010 – Introduction to the OpenGeo Suite
• Tuesday, July 20, 2010 – Introduction to PostGIS
• Wednesday, July 21, 2010 – Introduction to GeoServer
• Thursday, July 22, 2010 – Introduction to OpenLayers

We will be adding new and more advanced courses as demand dictates. We hope that our courses will be a valuable tool for any and all, whether you are a GIS professional, an enthusiast, a software developer, or a CEO.

Chris Holmes and I had a great conversation with Matt Aslett of the 451 Group (an IT business analysis organization) earlier this week.  The 451 Group has been a very useful resource in our exploration of open source business models over the past year, particularly their ominously named “CAOS Blog” (it stands for Commercial Adoption of Open Source). Anyhow it was interesting to hear what an expert in commercial open source thinks our organizational opportunities and risks are and Matt shared his thoughts yesterday on the CAOS Blog.

Paul Ramsey, OpenGeo’s spatial database wizard and frequent poster to this blog, was recently interviewed by Matt Ball for V1 magazine. It’s a good read, and we think you’ll enjoy it.

While the interview covers a wide range of topics, here’s a quote that encapsulates OpenGeo’s software to me probably better than any white paper or business plan could:

Building a web map app is something that a technically inclined forester should be able to do, it shouldn’t require someone from the computer science department. The Web is evolving to be a place where people can put a basic set of development tools, JavaScript and HTML, to a lot of different uses. Not inventing their own language…We’re just providing some geo flavor on top.

Read the entire interview…

On Thursday, the latest major release of PostGIS came out: version 1.5.  This release adds a long-wished-for feature to the open source spatial database—direct support for “geodetic” coordinates.

Geodetics are more commonly known as “lat/lon” coordinates. While you could load and work with lat/lon coordinates in earlier versions of PostGIS, the indexing and calculation code did not make any allowance for the fact that the coordinates were angular units, not cartesian units. As a result, objects that crossed the poles or datelines would not index properly, and calculations of areas, lengths and distances returned strange looking answers in “degrees” rather than meters.

With PostGIS 1.5, the new “geography” type is a 100% sphere-aware type, which can be indexed globally and returns answers in meters, using calculations on the spheroid for maximum correctness. It is built on top of a new disk storage and index format, which the existing “geometry” type will also transition to in version 2.0.

The development of the “geography” type was funded as a PostGIS core development task, by a company that chooses to remain anonymous.  Since the main geography development is complete the old task has been updated to a new version, outlining extra functions and performance additions that could be added to geography support.

We expect that the geography type will make it easier for new users to store their data in PostGIS (without having to learn about projections and coordinate systems before starting) and also allow global data managers to store and query international data sets for effectively.

What gets us up in the morning and into the office? Is it building an open source geospatial software empire, and eventually rolling around in a room full of money? No, that’s not the point.

Making money is a second-order concern for our organization that enables our first-order concern: putting useful tools in the hands of governments and NGOs. The revenue from professional services and support enables the continued development of the open source software, which we want to see as widely used as possible.

So, when we learn that the Amazonian Protection System is using PostGIS and Geoserver in their applications to manage and protect the rain forest, that makes us very happy.

And hearing how scientists and conservation groups have used Geoserver, PostGIS, and OpenLayers in building a decision support system for studying marine protected areas makes us sort of giddy.

And learning that the Great Lakes Commission is using Geoserver, PostGIS and OpenLayers in monitoring the environmental conditions of the Lakes make us quiver a little bit uncontrollably.

And then we have to sit down for a moment.

These are all folks who have used the tools that our team members build and maintain, to do good in the world. They don’t have to ask permission, and they don’t have to pay us, and the world is a little more understandable after they are finished than it was before. And that’s the point.

The US National Weather Service has a lot of weather to watch! They have sensors all over the continent, and in space, ranging from simple thermometers to orbiting satellites. When weather happens (and let’s be honest, weather is always happening, it’s happening on me right now) it happens fast — how does the NWS take in the whole situation at a glance and make decisions? On a map, of course.

The NWS Central Region Headquarters has built a test-bed for putting their situational information onto web maps, and the toolset they used is the OpenGeo suite of applications: PostGIS, Geoserver, OpenLayers. They also used raster-data standby GDAL for handling gridded data conversions.

A paper on their work (”Development of Web-based GIS Applications for Decision Support and Situational Awareness“) was presented by Brian Walawender at the  American Meteorological Association Annual Meeting this week.

This week, OpenGeo released version 1.0 of our OpenGeo Suite.

When we initially announced the OpenGeo Suite, it was a notion — a collection of individual software pieces we would professionally support as a whole. Now, it’s a product in the conventional sense — one download that provides all the pieces in a simple installer for Windows, Mac, or Linux:

• GeoServer — a geospatial data and map server;
• GeoWebCache — a map accelerator;
• OpenLayers/GeoExt — user interface libraries for building map applications;
• (New!) GeoExplorer — a browser-based map composer and publisher;
• (New!) Styler — a WYSIWIG editor for map styles (SLD);
• (New!) Recipe Book — code samples and documentation for building your own map applications;
• Full documentation for all components; and
• (New!) Dashboard — a unified administration panel for starting and managing the components of the OpenGeo Suite.

Up to this point, we have concentrated on clients already adept at downloading, integrating, and using the pieces of the Suite. With version 1.0, anybody can start publishing their data and building applications right out of the box.

This ease of entry aligns with the OpenGeo mission.  As an organization, we want to democratize mapping. That means offering tools available under non-discriminatory legal terms, like open source. It also means lowering barriers so that more people can use, build, and grow these tools.

Both novice and expert benefit from the seamless integration of the OpenGeo Suite.  By placing everything together in one place — software, documentation, examples, administration — the Suite offers a central resource to navigate, configure, and support the various pieces of your mapping application.

At a decision making level, we provide a corporate entity tightly bound to the software, providing support, expertise and training services to the community of users and administrators.

Version 1.0 is the first step in a long journey, but we know where we are going. Every day we ask ourselves: can we make our product easier to use? can we make it easier to learn? can we make it easier to try? We would love your feedback, so download the free 30 day trial and let us know what you think!

We’re looking forward to an exciting 2010, meeting those goals and growing our community.