Expanding the Reach of GIS through Web Mapping
- Details
- By Matt Ball | 28 September 2009 | 2408
New York City has had a pioneering web mapping portal since 2001 with the launch of the Open Accessible Space Information System (OASIS). The application has steadily evolved and continues to be at the technological forefront of web mapping. The New York City region is also blessed with one of the more detailed regional planning tools around with the Long Island Index interactive map. Both of these sites are the product of the CUNY Mapping Service and their partners. V1 editor Matt Ball spoke with Steven Romalewski, director of the CUNY Mapping Service at the CUNY Graduate Center, about the current state of web mapping, and the latest capabilities of web mapping tools.
New York City has had a pioneering web mapping portal since 2001 with the launch of the Open Accessible Space Information System (OASIS). The system got its start with funding from the Forest Service and a partnership of more than 40 organizations with the purpose of providing open space stewardship. The application has steadily evolved and continues to be at the technological forefront of web mapping. The New York City region is also blessed with one of the more detailed regional planning tools around with the Long Island Index interactive map. Both of these sites are the product of the CUNY Mapping Service and their partners. V1 editor Matt Ball spoke with Steven Romalewski, director of the CUNY Mapping Service at the CUNY Graduate Center, about the current state of web mapping, and the latest capabilities of web mapping tools.
V1: I’m excited about the advancements of Web mapping, particularly over the past year. It seems that there has been an explosion of tools and capabilities. Is that your take as well?
Romalewski: It’s amazing how quickly and how far the industry has come. When we started creating interactive mapping applications it was in the late 1990’s, and the tools were powerful, but pretty rudimentary. Just the fact that you had an interactive map online was an amazing thing.
Then, of course, when Google Maps came along that completely changed the scene. Now, it’s just commonplace that maps are online, they’re interactive, and you can view aerial imagery. Now, it’s seamless and totally streamlined, and I think it has really brought so many more creative and innovative people into the fold for developing and thinking about how to deploy interesting information on a map format on-line. I think you’re right that it has just exploded, and you know as they say, “the sky is the limit.”
V1: Most of the functionally that you’ve developed goes well beyond the standard base map mash-up that’s so established now. If you are a GIS person, the standard map website still looks rather rudimentary.
When I was running the New York Public Interest Research Group’s (NYPIRG) Community Mapping Assistance Project (CMAP) for about eight years, one of the reasons I moved on was because a lot of the maps that we were making were just dots on a map. A non-profit organization wanted to see where their service locations were located maybe in relation to city council districts. For them a simple customized map was perfectly fine, but from the perspective of what the potential was for spatial analysis, it was pretty limited and kind of boring.
I would try to tell the non-profit partners who we were working with that you could do so much more. Then all of a sudden Google Maps comes along and a million mash-ups happen and pushpins are everywhere (Laughter). I think there’s still somewhat of a demarcation between so-called, “consumer-oriented” interactive maps, where just seeing the locations of things is the goal, compared with more analytical projects.
Once you think a little bit more about geographic data that you’re working with and the potential to analyze information spatially and visualize it spatially, the different mash-ups that are out there that just show dot locations of things become very limited and repetitious. We’re interested to try to go much further beyond the simple mash-up, and really think about how we can layer in more information in a way that facilitates understanding of relationships and how the world works in more interesting ways.
That’s certainly what we tried to do with the Long Island Index mapping project, and that’s what we’re trying to do with the Open Accessible Space Information System (OASIS) website. We’re also in the process of developing a nationwide mapping application to help organizations that are doing outreach around the 2010 Census -- to make sure the undercount is minimized in metropolitan areas across the country.
The challenge really is to try to integrate information in useful analytical ways without overwhelming people. They don’t need to have the whole understanding of what GIS is, but they need to understand how different map layers can work together. We try to get to the “Ah-Ha” moments, to understand what the spatial relationships are. The visualization might not answer the questions, but it gets them thinking about the issues in a way they hadn’t before.
V1: There’s so much being done on the user interface side these days, and I’m certainly impressed with what you’ve done. It gets beyond the static map for a much more dynamic map, but it also hides some of the complexity.
Romalewski: That’s really the challenge. I think some of the so-called, “neo-geographers” talk in a disparaging way about people who think like GIS analysts. Most people don’t know what the term GIS means, let alone how to manipulate layers of information, turning layers on and off, and symbolizing things on a map.
We’re GIS experts and we want to create applications that leverage the power of GIS, but don’t have the user-complexity that comes with some GIS applications. To us, it’s obvious that you don’t design interactive maps in a complicated way. You want to design them in as straightforward a way as possible, but also in a way that puts some of the power of GIS in people’s hands, where they wouldn’t have had access to it before.
It’s definitely a challenge, but I think going back to your original question of explosion of new ways of displaying and visualizing map information online, the tools that are now available to developers of these applications are just amazing. I think a lot of that is because of the open-source movement, but also because a lot more creative and innovative people are thinking about this. They’re developing JavaScript frameworks that can do amazing things, and different ways of manipulating map information in a Web environment. Now, I think it’s eminently possible to get away from the complex GIS mindset and develop really powerful applications that are much easier to use.
V1: I’m impressed with the map design capabilities that are available online now. I’ve noted in something that you’ve written that you didn’t want to create just a mash-up because you wanted some control of the cartography, the look and feel of the map.
Romalewski: We certainly looked at the common online mapping platforms -- Google Maps, Microsoft’s Bing Maps, and Yahoo! Maps. They all are nice looking maps, but we didn’t want to just rely on their handy work, we needed to design some of our own symbolization and styling.
It’s still important when you’re creating maps and developing online GIS applications, to know some of the basics of geography and spatial concepts and to combine that with graphic design and cartographic techniques. Having that background really helps in terms of developing maps that can visually tell a compelling story and minimizes “map crap,” as John Krygier puts it, and minimizes the generalizations on that map to really focus people’s eyes on what the actual story is.
For example, with something like the Long Island Index map, the underlying street geography is not really the main part, the main part of it is visualizing the spatial relationships about demographics, population characteristics, and land use patterns. The streets, in some way, would just get in the way. Certainly some of the streets are there, but they’re minimized in the background and it’s just the major streets. If we had to rely on Google Maps for example, we would not be able to display data effectively, so that’s why we have to develop our own symbology.
V1: I’m really impressed by what you’ve created there. It’s one of the more interesting map websites that I’ve seen in terms of the amount of information that you’re able to deliver. What has been the feedback from the community? Are you seeing an explosion of use?
Romalewski: I have to say it hasn’t been an explosion of use. I think partly because the organization that we developed the maps for, the Long Island Index Project, has been more strategic in their outreach efforts. Soon, they’ll be reaching out to the library system across Long Island and a more aggressive media effort, both to get stories out about using the maps, but also to help reporters and editors use the maps as tools to help their reporting.
The market in Long Island is not as big as New York City, and it’s in some way a localized application, so there’s not this huge user-base, but the people who use the maps are very impressed. They very quickly understood that it’s a powerful tool for their work, whether it’s a reporter, or a community organization, or a transit advocacy group, or an organization like the Regional Plan Association (RPA), which is based in New York and is the major urban planning group in the region.
The RPA does a lot of planning work on Long Island in terms of thinking about downtown development, which in suburbia, you don’t think about much. They have been pushing the use of the Index and they’ve been able to use it as a platform to get information out and to visualize the relationship of housing and transit and other inputs for downtown development. When people use the maps and they see it, they’re really impressed.
We’ve integrated tools that quickly and dynamically change the transparency of layers. When I’ve presented that to organizations, there are just “Oos and aahs” in the crowd. People are amazed when they see that. It’s interesting to me because I know it’s a powerful tool and intuitively we knew that it would a useful way of displaying the information, but it’s a good affirmation that this is really helpful. Hearing that reaction from the crowd lets us know that we’re moving in the right direction.
V1: The Index is quite complex in terms of the amount of data that you have there; but it also is represented in a way that is fairly intuitive. Given the amount of options, there is probably a level where some of your users need a little hand-holding. Do you get requests for training?
Romalewski: We do, and we’ve put together some video tutorials about how to use some features of the map that may not be obvious or intuitive, and we’re exploring that more. That’s also a challenge to make the site accessible to anyone, but also to enable those interested to use some of the more advanced tools that we’ve built into it. We’re trying to borrow techniques from other web technologies and apply them to the map applications, such as hover-tips and little pop-up menus that provide contextual descriptions of what is happening or what you can do with this tool, and how you can visualize this layer in relation to another one.
That’s definitely been challenging with a site as involved, and as richly layered as the Long Island site, but I think as we get more feedback it’s going to be even easier to use for people at all levels.
V1: I appreciate that you don’t just stick up a web map and say you’re done, particular in such an involved and dynamic use case as this. There’s probably always new data to add or new functionality.
Romalewski: Definitely, and we took a page from some of the web-designers that are pushing the idea of not developing a detailed requirements document and then following that to the letter. The process that we use for the Long Island site was and still is “agile development,” and much more interactive.
We develop a bare bones site and see what people want, and then begin adding features and data. We also design the sites to be expandable, so we’ll be incorporating data in the next couple of weeks about more detailed age categories in the population section, and other information about transit and usage, and also other locations. We talked to someone today about integrating data about stimulus funding projects and health care facilities.
We designed the site with the strategic needs of the Long Island project in mind. We could have put a million different Census data sets on the website and made it all more complicated and challenging to use than it is now, but we wanted to focus the data to strategically help or supplement the ideas and proposals that the Long Island Index Project has been talking about such as downtown development and more dense housing.
V1: How long have you been working on the CUNY mapping service?
Romalewski: The CMAP project at NYPIRG started in 1996, and that continued until about 2005. In 2005, we spent time thinking about how we could expand and move beyond just the non-profit world. I had some professional relationships with people at the City University of New York, and John Mollenkopf who is the director of the Center for Urban Research does a lot of demographic analysis and immigration research. We had worked together on a number of projects, and it seemed like it would be a really good combination of the demographics analysis and academic resources that his Center provides, combined with our mapping skills in an urban planning context.
We decided to move to the Center at the CUNY Graduate Center in January 2006. The Center for Urban Research already had a project called the CUNY Data Service, which is a Census Bureau affiliate that analyzes Census data on a full-time basis. So to parallel that project we decided to call what we’re doing the CUNY Mapping Service.
V1: Do you continue to do outreach to non-profits and groups that want to use your service?
Romalewski: Yes, but one nice thing about being at the Graduate Center is that while we’ve been able to continue providing GIS services to non-profits, we’ve also been able to work on much more interesting spatial analysis projects with other city researchers and through John’s relationships in the foundation world. He has a number of projects with foundations such as the Ford Foundation and the MacArthur Foundation, looking at the low-wage workforce on a nationwide basis, and immigration issues in metropolitan areas and nationwide. We’ve been able to work on bigger, more interesting projects in more innovative ways. We’ve been able to really leverage the university’s IT resources, in terms of developing these online mapping applications.
V1: Do you have a pool of students that you harness to help with your work?
Romalewski: We’ve been able to work with some really great PhD students at the Graduate Center at CUNY. CUNY is a big institution, with twenty-three campuses throughout the five boroughs. Some of them are community colleges, some four-year undergraduate colleges, some of which also have masters programs and the Graduate Center is the PhD granting institution within CUNY. And there is an environmental and earth sciences PhD program that works closely with two geography departments at CUNY at the Hunter College campus and the Leman College campus in the Bronx. We have been able to work with some really great students that not only know GIS very well, but know GIS within the context of environmental and demographic research. They’re really good at statistical analysis and pushing the envelope as far as using open-source tools, as well as proprietary software.
I’ve also been really lucky to work with great staff colleagues at the Graduate Center, especially at the Center for Urban Research but also throughout the school. And in particular on our mapping projects, I work directly with an amazing application developer, David Burgoon. He combines professional programming skills with a solid knowledge of GIS – a rare thing, I’ve found – so he’s able to intuitively develop impressive online mapping tools with seemingly minimal effort. And Christy Spielman, a graphic designer and GIS expert, works with us now on a contractual basis but I’ve worked with Christy for about 10 years, first at CMAP and then on the OASIS project and now at CUNY. She helped create the original OASIS website, has been instrumental in the project over the years, and now is helping to guide the updated version and work closely with our partner groups along those lines.
V1: What are you most excited about in terms of your work right now?
Romalewski: I’m most excited about our experiences in the intersection of proprietary and open-source tools, and mixing and matching those to get the best of both worlds, and integrating that to develop really powerful and easy-to-use applications. There are three online mapping projects we’re involved with now that I’m really excited about. There is the ongoing Long Island Index project, we just rolled out a complete transformation of the OASIS project that we started back in 2000 with the U.S.D.A Forest Service and a nationwide mapping service for groups that are working on outreach around the 2010 Census.
The OASIS project is a partnership with ESRI and a number of other organizations. We created an online mapping application that included aerial photography and detailed land-use data for the five boroughs of New York City, and it was very innovative and ground-breaking at the time, and very quickly eclipsed technologically by things like Google Maps and others. Now, we’ve rolled out a totally new version that uses ESRI technology and Microsoft technology on the backend and some open-source frameworks on the front-end that I think is going to make for a very powerful application.
The nationwide mapping service for groups that are working on outreach around the 2010 census is designed to highlight areas in metropolitan regions that are considered by the Census Bureau to be hard to count. The application allows groups to use maps to target their outreach work and to zoom in close into detailed information about individual communities and census tracts. It includes demographic characteristics of those areas so groups can shift their outreach message appropriately to the local audience. It also incorporates other data-layers to be able to visualize relationships among different variables like poverty, and immigration, and race, and ethnicity, and language spoken at home, and things like that.
V1: It sound like there’s a lot of the focus on social outreach as well as the environment in your work.
Romalewski: We try to develop applications that are useful in a broad set of experiences and a broad set of areas of interest. A good example is the OASIS project. It started with the focus on bringing together information about open-space resources in the city -- mainly parks, but also community gardens that are not maintained necessarily by government agencies but are set up by private citizens and local non-profits.
Maps were built to help inform the public about the importance of open-space resources, but you really can’t do that in a vacuum, you also need to be able to show how these open spaces relate to transportation patterns, and housing, and land-use more generally. We mapped proximity to schools to community gardens in order to bring kids there to have them learn about the environment. We also added historic landmarks, and all sorts of data representing what we call “civic infrastructure.” It’s really become a much more broad-based, cross-issue, and comprehensive application that provides a wealth of information richly layered about the city that you’re not going to get in one place anywhere else online.
V1: Is the new OASIS site now available online?
Romalewski: The old archive site is still there at www.oasisnyc.net, but now we have a new beta version that uses ArcGIS Server, OpenLayers, and the Ext JavaScript framework that you can access via “OASIS 2.0 beta” links on the site.
We have a lot of layers there, including zoning, transit routes, housing projects, schools, and a whole set of environmental information layers. A couple of intriguing things we include are historical aerial imagery that uses transparency to fade from one layer to the other, so on the OASIS site, you can fade from 2006 aerial imagery to back to 1996 aerial imagery. It’s a seamless transition from one image to another, so you can really see how an area has changed. It’s very powerful. Then we’re also working in partnership with the Wildlife Conservation Society to incorporate data from the Mannahatta Project.
Mannahatta is a GIS effort to take old maps from the 1700’s and develop an impressive ecological analysis and create digital representations of what Manhattan probably looked when Henry Hudson sailed up the river in 1609. We incorporate several data layers that they have produced, including the original shoreline of Manhattan and the Native American trails throughout Manhattan, and perhaps most impressively, a photo-realistic image of what Manhattan looked like. You can go from 2006 back to 1609 and see the vegetation and the waterways that existed on your block. It’s really cool.
We view OASIS as an important and unique entry point into these types of issues. You’re not going to get everything, but if you are looking at the OASIS website and you are at a particular spot on the map, you can see a wealth of things and then there are a lot of pointers to other places, where you can get much more detailed information about that location.
V1: These kinds of applications excite me because they get back to the true promise of GIS for the stewardship of the planet. That was the focus of your work when you began, but now, with all this data accessible, you can really paint a picture and let people explore it on their own.
Romalewski: It really opens people’s eyes. On a more explicit basis, we are working as well through OASIS on a partnership with the Forest Service, where they have developed a stewardship mapping project called “STEW-MAP,” where they’ve surveyed a couple of thousand local environmental stewards in New York, and based on information that these groups have given back to them, they have mapped their areas of interest or “turfs” as they call them. So you can not only zoom on OASIS and look at your block, and see if there is a garden nearby, and what it looked like in 1609, but you can also find if there are any local groups that are involved in environmental stewardship and help them out, volunteer with them, learn more about what they’re doing, and get much more connected to the stewardship of your local community and by extension, the stewardship of the planet.