Archive for the 'GeoBlog' Category

How I think about GIS

GIS is my profession. What is GIS? I suppose about half the people who I talk to know what it is. It can stand for either Geographic Information Systems or Geographic Information Science. Although I once read it can also stand for Geographic Information Services. Geographic Information Systems collect, store, analyze, present, and disseminate geographic data. Geographic Information Science is the science behind all of this.

Even if someone doesn’t know what GIS is they have still used a GIS. Everybody has used Google Maps or Google Earth. Everyone uses a GPS navigation system to get somewhere. Geographic data is being presented and disseminated to you so you can make better decisions. Many times user location data is being collected automatically by the application to improve performance. User location data is very valuable because it can also be used for advertising as well as be used to improve algorithms for other services. A tremendous amount of work has gone into creating all of these maps. Much of the technology used to make these applications and analysis is intelligently hidden from the user.

Before GIS there were only maps. Computers revolutionized maps and the result was GIS. By understanding the transformation undergone by geographic information as it has moved from paper map to machine, you can understand GIS. Roger Tomlinson, also known as the father of GIS, coined the term GIS in the 1960s. It was in 1962 he presented a paper entitled, “Computer Mapping: An Introduction to the Use of Electronic Computers in the Storage, Compilation and Assessment of Natural and Economic Data for the Evaluation of Marginal Lands” at the National Land Capability Inventory Seminar in Ottawa. To provide a frame of reference, the first interactive game (SpaceWar!) was created in 1962. Computers wouldn’t even become mainstream for another two decades. Computers are still evolving, and so is GIS. I personally think the most interesting part of GIS right now is Web-based GIS.

Some important topics in GIScience include data collection, spatial data structures, spatial statistics, spatial analysis, visualization and data discovery. Here is a list of some examples:

Data collection: satellites, sensors, digitizing, crowdsourcing

Spatial data structures: spatial databases, topology, rasters, and vectors

Spatial analysis: spatial overlays, simulation and modeling, cellular automata, network analysis, classification, viewsheds, and watersheds

Spatial statistics and relationships (type of analysis): interpolation, spatial regression, and spatial autocorrelation

Visualization and data discovery: interactive maps, geographic information retrieval and spatial search

GIS professionals use Google Maps and Google Earth because they are the best tools for what they do. They use other software as well depending on the task at hand. ESRI’s ArcGIS software has the biggest market share, but there are other types of proprietary GIS software. It is also possible to do almost any GIS task using completely free and open source GIS software. Finally, you don’t even have to be a GIS professional to contribute edits to OpenStreetMap, the free editable map of the world.

I had to include at least one map in my GIS post. Here is a screen of my Walking Dead Twitter Map for my WebGIS class. Halloween tiles courtesy of Mapbox

I had to include at least one map in my GIS post. Here is a screen of my Walking Dead Twitter Map for my WebGIS class. Halloween tiles courtesy of Mapbox

Credits: Many ideas for this post were taken from Twenty years of progress: GIScience in 2010 by Michael F. Goodchild.

Share

iBeacon

iBeacon is a piece of Bluetooth location based tech that was introduced in iOS 7 and is also found in the latest version of Android. It is relatively simple in concept. A Bluetooth transmitter emits a signal and your phone can pick it up. That’s basically it. But please don’t stop reading. Big things have small beginnings.

I attended an iBeacon Meetup recently to learn more about it. It was at Canvas.co in DC, a very cool place for start-ups and the like. Three presentations were given about iBeacon. The first one was the most comprehensive, given by Radius Networks, a whole company that seems to be built around working with iBeacon. The presenter did a great job talking about the potential of iBeacon, its capabilities, and limitations.

The event had a bunch of iBeacons around the room and an App that we could download and start detecting iBeacons. iBeacons have a range of about 50 meters and are just bluetooth transmitters. They can come in many forms such as a battery powered dongle, usb hub, laptop, or even your iPhone itself. All they do is emit a three-part identifier every second or more. There is also a signal strength value that enables a rough estimate of distance from phone. iBeacons don’t transfer any actual information.

iBeacon usb hub

iBeacon usb hub

Using a unique identifier iPhones can detect iBeacons of a certain type. The user will have to have a specific application on their phone to see the right iBeacons. The iPhone can monitor for the right iBeacons and when one is found it can launch the application for 5 seconds. Enough time for instance, to display a notification on your phone. The second presenter gave a neat demo of a food stand having an app. When the customer walked into the store the app would open with the latest menu, the customer could order with their phone and be notified when their order was ready.

The most important thing I learned about iBeacon was that iBeacon is not good for providing accurate indoor locations. This is what I would be most interested in, I have done some research in micro-geographical spaces. I think this is a rather technological challenge, because I think you need some rather precise clocks to accurately measure signals between the transmitter and receiver in such close areas. I’m no expert in this area however.

Nevertheless I think iBeacon is still a cool feature and I can see it taking off in the next year. It would definitely be useful for making purchases in physical locations and interacting with displays. I think some creative people will find uses for it in games as well.

Share

Designing Accessible Maps

When making maps or any type of graphic I never thought of how the product would be perceived to people who have color vision impairment. I didn’t know how big of an issue it was. After reading Benhard and Nathaniel’s Color Design for the Color Vision Impaired paper I now know it affects a significant amount of people, especially men. Therefore I agree it is preferable to create maps that can be understood by people who have color vision impairment.

I think that this type of design consideration will not always be followed because of the increased overhead it requires. Maybe some designers will not follow these best practices if they are not enforced. Other reasons for designers not making maps for the vision impaired include ignorance of the situation or some designers will not want to sacrifice their original design to cater to people who have color vision impairment.

I believe that the best remedy would be to cure people of their color-blindness by injecting trillions of copies of a genetically engineered virus into their retina. DeWeerdt’s Seeing Red paper explained how this has been tested on squirrel monkeys and enabled their cone cells to produce red-sensitive pigment. Why should us as cartographers alter our maps just accommodate less than 10 percent of the population? Of course I’m not being serious. Fortunately there are guidelines out there on how to make your maps better for those who have color vision impairment.

For every type of user, spectral color schemas, otherwise known as rainbow color schemas are not recommended for quantitative or ordered data. This is because there is no inherent magnitude message in a rainbow. In Spectral Schemes: Controversial Color Use on Maps, Brewer believes that spectral color schemes should be encouraged for diverging quantitative data. I agree somewhat, more so if a diverging spectral scheme is matched with diverging data. I think spectral schemes have an advantage of the amount of different hues they use, I think that this increased ‘spectral resolution’ can bring out details in the dataset. The disadvantage is that I don’t have any association of magnitude between green and blue. Reading Brewer’s paper I have come to appreciate the diverging two-tone spectral scheme:

Brewer gives some recommendations for people who have color vision impairment. Skipping over yellow-greens should accommodate map-readers with red-green visual impairment. I think this is a fairly straightforward rule I can remember. I also stumbled upon Colorbrewer2.org, a site that helps you pick appropriate color schemes. Copyrighted by Cynthia Brewer, Mark Harrower and The Pennsylvania State University. Clever play on words Cynthia! In the Bernhard paper a piece of software called Color Oracle is mentioned. I installed this and think it is great. With just one button click you can quickly see how people with Deuteranopia, Protanopia, or Tritanopia see. This can be a very useful tool for cartographers creating barrier-free maps!

Share

DARPA Innovation House

This is the post where I get to brag about what I’m doing. I’m the lead developer for one of the teams selected for the DARPA House Innovation Project. This is an eight week program that started on Sept 17th. This came at a great time for me as I wrapped up my summer internship as well as completed all my thesis requirements in August.

Let me share some of the DARPA Innohouse press kit:

http://www.darpa.mil/NewsEvents/Releases/2012/07/10.aspx

http://www.engadget.com/2012/07/10/darpa-innovation-house-project-imaging-data/

http://www.wired.com/dangerroom/2012/07/hackathon-guinea-pig/

http://www.washingtonpost.com/blogs/innovations/post/darpa-george-mason-seek-trailblazers-for-eight-week-innovation-house-study-program/2012/07/10/gJQAv06naW_blog.html

It feels great to be working for the agency that created the internet and was critical in the development of UAVs. I first took notice about the project in July reading engadget.com, a technology blog that I read practically daily. I decided to put together a proposal and a team and apply; and now here I am!

Our team is called Looking Glass. We are creating software that can extract meaningful information that is relevant to the user and present it to them in a meaningful way. We have reshaped our concept several times since beginning our project and are approaching this problem from different angles. I have started a new blog for our team at teamlg.org/blog. I’ll work on linking our team posts for Looking Glass to my personal blog site.

DARPA is experimenting on the feasibility of effective software design and development in a short-fuse, ‘crucible-style’ living and working environment. I have heard some chatter around the office that we might be lab rats in some type of social experiment.

There are lots of resources at our disposal some amazing mentors to help guide us. This is important as the objectives are not well defined on purpose, it seems like they really don’t want to risk constraining us. So far I have been working many hours and learning a ton. I feel like I need to work twice as hard in order to stand out amongst the other teams. At the same time this project is more of a collaborative effort and everyone has been really friendly and supportive so far. There are six teams total, and I am surrounded by a lot of talent; from nuclear physicists to digital signal experts to artists.

Share

TomTom can help you find your way

This past weekend I helped my dad pick-out a gps device. My parents are going to Florida for vacation next month and he thought it might come in handy. We ended up purchasing the TomTom One system from Circuit City for $250.

I was impressed with the quick set-up time. It came with a relatively strong charge out of the box. The TomTom asked you basic questions such as your language preference, what time it was, and your home address. Everything was inputted via the touchscreen, which was great, and then it automatically synced with the GPS satellite. We tested it by letting it guide us back to our house.

It comes pre-loaded with many POIs (points of interest). I was pleasantly surprised when it showed all of the nearby restaurants on the menu. Even my friend Peter’s restaurant El Estribo showed up.

picture-3.png

If you pay extra for TomTom Plus services you can have extra features such as live traffic and weather reports. A requirement for these services is to have a bluetooth connection through your internet-enabled cellphone. Whoah! We don’t have a Bluetooth or internet-enabled cellphone, those are expensive! But it is interesting to think what we can do if we ever get one. Another interesting feature I noticed was TomTom friends, with TomTom friends you can see wherever your friends are at if they also have the service. I doubt very many people have this service but it will probably be more ubiquitous in the future, for the good and the bad.

The prices of gps devices have really come down, they used to cost twice as much a year ago. I am seeing them more in people’s cars. I will definitely consider purchasing one for my own car in the future.

Share