Sensors and Systems
Breaking News
Maxar Secures NOAA Approval to Provide Non-Earth Imaging Services to Government and Commercial Customers
Rating12345WESTMINSTER, Colo.- Maxar Technologies (NYSE:MAXR) (TSX:MAXR), provider of comprehensive...
AEye Announces Groundbreaking Immersive Lidar Experience for Attendees at CES 2023
Rating12345DUBLIN, Calif.- AEye, Inc. (NASDAQ: LIDR), a global leader in...
WIMI Hologram Academy: Multi-Dimensional Holographic Vision Opens A New Chapter In Cyberspace Mapping
Rating12345HONG KONG – WIMI Hologram Academy, working in partnership...
  • Rating12345

Ramsey_Paul_TNPaul Ramsey has been an open geospatial advocate for some time and is deeply involved in PostGIS and MapServer development and project consulting. Paul was the keynote speaker at last year’s Free and Open Source Software for Geospatial Conference (FOSS4G), and teaches many workshops to help others develop solutions with this technology. He works for OpenGeo, a company that has just released the OpenGeo Suite, a productized stack of optimized open geospatial technologies. V1 editor Matt Ball spoke with Ramsey regarding this new suite of tools, and about the status of open geospatial tools in general.

Ramsey_PaulPaul Ramsey has been an open geospatial advocate for some time and is deeply involved in PostGIS and MapServer development and project consulting. Paul was the keynote speaker at last year’s Free and Open Source Software for Geospatial Conference (FOSS4G), and teaches many workshops to help others develop solutions with this technology. He works for OpenGeo, a company that has just released the OpenGeo Suite, a productized stack of optimized open geospatial technologies. V1 editor Matt Ball spoke with Ramsey regarding this new suite of tools, and about the status of open geospatial tools in general.

V1: I’m really interested in the mission-oriented work of OpenGeo. I see that the consulting work of OpenGeo supports a non-profit that aims to create a better managed urban environment.

Ramsey: OpenGeo is a part of the Open Planning Project. The goals of the Open Planning Project have always been around social engagement in urban planning and open government. We’ve taken those goals and superimposed them on what we do, which is to develop open source geospatial technology. We are managed as a separate entity.

We’ve formed ourselves as a social enterprise, meaning a business, but a business where the variable of maximization is not value of capital. In your traditional startup you dump some capital into it and you hope to make that pile of capital as big as possible. Our goal as a social enterprise is to take that starting capital and use it to grow as much social good as possible.

Our synthesis of how to do that best is to take our startup capital and work to be financially self sustaining. We’re working to build a business around open source tools that allow people to more democratically do mapping, but building a business with an aim to self sustain the development of that software so that we’re not tied to the vagaries of funding at the end of this process.

V1: What has been the motivation for the recently launched OpenGeo Suite of tools?

Ramsey: Our take on GIS tool space is that there’s some need for democratization, and there’s a place in the market for the kind of tools that we’re putting together. The old GIS tools have real access issues for folks who aren’t already part of the GIS priesthood. It’s a pretty expensive and daunting task to understand the ESRI ecosystem, and that’s why a lot of organizations have looked at GIS and said, “the hell with it.”

That flipped in about 2005 with the rise of consumer-oriented mapping tools, but the consumer APIs have their own limitations. They only do so much. We feel that there’s a place in the middle here for tools that are as open to non-geospatial developers to access as consumer tools, but not as limited in terms of the audience that they’re able and willing to serve as the consumer tools.

V1: Was the idea of the stack of tools already in your mindset when you started, to align these and make them simpler to use by stacking them?

Ramsey: There are a lot of developers that want to use open source tools and want to make maps. But when they enter the open source geospatial tool space they are subject the tyranny of choice. They have so many options.

If somebody wants to use open source tools to make a mapping application, there are more than a dozen different components and they can put them together in more than a dozen different combinations. You can use a python bit for your cache, an OpenLayers or Flex bit for the user interface, and so on.

The number of pieces, plus the number of ways you can combine them, means that the number of possible architectures is really high. That amount of choice is nice when you understand the terrain, but it can be paralyzing for new developers who just want to make an application.

One thing that the OpenGeo stack does is to cut down the choice. All the pieces that we picked we know are top notch, and because we’ve put together a team of people who are experts with them, we know we can make the pieces work together really well. We’ve cleared away the thicket of choice, and we’ve done the integration work that from a technological point of view makes everything work well together.

You asked if when starting out we thought about the stack, and we didn’t necessarily. The Open Planning Project’s experience with open geospatial software started with Geoserver and stayed that way for a long time. The main thing that we’ve taken from that single product that dates to 2001 is that open standards and interoperability are key.

Even when you build the stack together you have to build it in a way that people have the option to pull apart your stack in different places and bolt on different things. If you don’t do that, if you deliver a single black box, you’re substituting the problem of too many choices with a problem of no choices and no flexibility.

V1: There is a lot to cover in terms of what the open geospatial software suite of tools can do, and also what some of the sweet spots are for the technology. What are some of the best uses of open source geospatial products, and are there some limitations as you compare it to more established stack of geospatial tools?

Ramsey: Behind all the underlying technology of the OpenGeo toolkit is the original Open Geospatial Consortium’s (OGC) simple features specification. And that gives developers a heck of a lot of rope. It’s the specification that underlies the geospatial databases, but it also fits underneath the Java libraries that the middleware is built on, and increasingly fits under client-side objects, which are being instantiated in the browser.

With that representational understanding you can do things like full spatial queries of the sort that you can do with Oracle Spatial. You can run them on almost any layer of the stack. So there’s a great deal of spatial smarts, the kind of smarts that people use to do things like summarization operations, joining two spatial layers to create a new spatial layer, etc. There is a great deal of analytical power there.

It is not completely overlapping with the analytical power in something like ArcGIS or GRASS [Link to http://grass.itc.it/ ] for example. But I don’t know if that is a big deal. You can do compound spatial queries in your toolset, and that allows you to answer as much as 80 percent of the analytical questions that people come to GIS with. You’re really left with some really high end raster analytics that you don’t quite have available.

So, if you want to do viewshed analysis, it’s true that you can’t do that over the web with the OpenGeo toolset, but you can do most other forms of analytics.  The web toolset doesn’t need to do everything for everyone.

V1: My sense is that the OpenGeo Stack is a little more focused on services than say creating custom tools for specific purposes.

Ramsey: That would have been a fair characterization a couple of years ago when it was really just Geoserver. Geoserver was very much an engine that bolted on top of data sets and spat out OGC web services — WMS, WFS. Once you have those services, the question is what to do with them, and that’s where application building comes in.

Over the last couple of years the people that have joined the team have not been Geoserver developers, they’ve been developers in the world of OpenLayers and GeoExt. These are both web-based user interface frameworks. Our goal with the suite is to be able to make a tool that allows people to build high-end applications.

When I review my history in GIS consulting, I spent an absurd amount of time between 1995 and 2000 building variations on the same application using ArcView 3. The application being: view some layers on a map with a workflow process of five forms and a data entry point. That was a really common application to build, and due to the limitations of the toolset at the time, you could really only build it using desktop tools.

Now you can build it over the web so that you don’t have deployment issues, you don’t have licensing issues, you have just one point of build and one point of deployment. Over the web you can satisfy those little niche needs, which old customized desktop applications used to meet.

Some of these apps on the low end are being built using Bing or Google Maps, but there’s space for a toolkit that has more functionality to build vertical applications. That’s the place that we really want to move the OpenGeo Suite into.

We don’t think it should be very hard. Anyone that can do standard web development should be able to build these kinds of applications. It shouldn’t take a GIS 101 deegree. It should take some understanding of web technologies, but not a lot.

The OpenGeo Suite aims to be a toolset that gives you the software pieces to do that, and also the education pieces, such as examples and tutorials. It’s not a tool for end users, it’s a tool that web developers can use to build tools for end users.

V1: Are developers that are building these specialized tools for a particular client and a specific workflow then becoming specialized? For instance, are there groups of open source developers working specifically with natural resources or other specialized groups?

Ramsey: It’s really early days now, and right now most of the people that are doing this are the full-on technology experts. They’ve had to learn each element of the stack independently. There are a few shops that focus on municipal applications, but that’s still a pretty broad swath that’s not specialized down to a single vertical application area.

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, just using Java script and HTML. We’re just providing some geo flavor on top.

V1: In your own work are you doing specific applications for the urban planning set to serve the Open Planning Project?

Ramsey: We’re doing some interesting work at OpenGeo developing applications with the World Bank. They have some really specific goals around user-generated content and creating what they call a GeoNode, which can be deployed in multiple different places, and used for user-generated content that is then federated upwards to larger organizations. It sounds a lot like spatial data infrastructure (SDI), but cutting away a lot of old SDI rules that tend to be very data centric, while we’re trying to be more process centric.

The application area that has interested me the most is in transit. That’s a vertical where things are moving along quite quickly. My favorite example is Portland TriMet. They brought in an independent developer to tell them what’s new and possible, and he built them a cool beta site with OpenLayers, Geoserver and ExtJS. He built the site and then left, because he’s an independent developer and he had other contracts to do. The TriMet internal staff then picked it up to take it into production, and they called us to provide expert advise on customizing some of the tools, and bringing some enhancements into OpenLayers.

We did some consulting work for them, and when it came time to move to the managers they wanted the security of a support contract. It became one of our first enterprise support contracts before we productized the suite. It’s an arrangement where they have consistent access to expert resources outside their own very skilled internal staff.

It all came together for Portland TriMet, where we established a relationship and built tools in partnership. They in turn have fed back their knowledge of the transit space to us, which fits very will with the urban transportation goals of the Open Planning Project. So now we’re building a transit vertical, but we’re building it on the Open Planning side and feeding our OpenGeo spatial expertise back into our parent organization. They’re the ones that are doing specific transit projects, and we’re feeding our technology knowledge back to them.

So for now we’re focused on being experts in geoweb technologies, and try to meet the vertical experts halfway.

V1: Is there a growing interest in web-based technology as well as open source web based technology?

Ramsey: That linkage between interest in web-based technology and interest in open source technology is there because the two are growing in parallel. The sweet spot for open source geospatial projects in general is green fields, places where people haven’t made a decision already and haven’t been burned by any legacy systems – places like web applications. The opportunities can be brand new start up companies, which at this point are almost all web based, or they can be new projects within older organizations.

A new organization example is a company that I’ve been working with in Seattle called Zonar Systems. They do fleet tracking, and they’ve built an open source based system underneath their fleet tracking hardware to keep track of the history of where all the fleet vehicles have been over time.

Redfin is a real estate startup also in Seattle, they’ve built up an infrastructure for managing parcels and facilities, a very traditional GIS space, but because they’re a startup they didn’t have a lot of starting capital so they built their solution using open source tools.

New organizations don’t come with any legacy, so their question number one is whether there is an open source tool that can do it. And they only move on to question two if you get a “no” on question one. Question two being, “Is there some other proprietary tool that can do it?”

Even in existing big organizations, opportunity for open source arises when a new project is started. Its an opportunity to try open source even though they may have standardized on a proprietary solution. They don’t have to commit to open source or change the entire organization around it, but they can at least give open source a shot.

This is the route that happened about four years ago with the provincial government of British Columbia. They received a federal mandate to provide access to their web services using OGC WMS standards. The obvious default thing to do was to put a WMS connector on top of ArcIMS. But they used the mandate as a chance to experiment and test the alternatives and put together their WMS services using MapServer instead.

Open source just isn’t a dirty word anymore. Go back to 2000, and there were a surprising number of managers that would literally shy away… there was still that “dirty hippy” aura around open source. But at this point they do surveys of Fortune 500 CEOs about whether they’re using open source or have an open source strategy, and the responses have gone from 20 percent positive to 80 percent positive. The snide remark often made in the survey reviews is that the remaining 20 percent are using open source but their staff just hasn’t told them.

The thing that changes a conservative decision maker’s mind isn’t a great sales presentation, it’s knowing that other conservative decision makers have already made the decision. Once that wave starts rolling, it’s very difficult to stop.

At the leading edge, visionaries that see open source as a competitive advantage. The CTO of Zonar has an excellent rant about his use of the PostGIS database. He says that he hopes all his competitors use Oracle, because every dollar that they spend to support Oracle is a dollar they can’t spend on making their services better or bringing their price point down.

In the public sector, people are less taken with straight bottom line arguments and tend to see risk as the primary variable for consideration. And perception of risk only goes away when they see big risk-averse organizations that have evaluated open source and are using it. We hold up as examples the French National Mapping Agency, or EDINA a huge data holder in the UK, or Infoterra a large data seller in the UK. All of these people are using PostGIS for their applications. The UN, the World Bank and LandGate in Australia are further examples of large conservative organizations that are using Geoserver in production and have been doing so for some time.

So we’re working to knock down objections over time. We have reference clients, we’re an organization you can call to get support. If you want to know that there’s longevity and a technical road map, we have that too. Putting all this together, I don’t see any reason why organizations wouldn’t consider open source on the same basis as proprietary, and hopefully they would include both in their product evaluations.

Leave a Reply

Your email address will not be published. Required fields are marked *