Cairns Blog: Cellular Government

I recently spent some time thinking and writing about a really interesting project called City Scan. The project puts digital cameras and other devices in the hands of residents to document the problems with physical conditions, such as pot holes or broken park benches, that they see in their neighborhoods. The information residents gather is mapped and used to foster a better-informed dialogue between citizens and their local government. The project is a brilliant idea, but the technology it employs is outdated. In the following piece I discuss modernization and other aspects of the project in more detail.

Connecticut Policy and Economic Council & City Scan

The Connecticut Policy and Economic Council (“CPEC”) operates the City Scan project (the “Project”). The Project’s goal is to gather and organize information regarding physical neighborhood conditions and to coordinate city services to address them. It promotes resident involvement in problem identification and problem solving while enhancing government accountability. The Project has been successful, but, as technology has become cheaper and better over the years, it needs to be updated. There is room both to reduce the cost of project adoption and to better serve the underlying goals. The Project has a great fundamental model, but long-term success depends on these modernizations.

The model of the Project is its combination of technological components: mobile input devices, information storage, and information processing. For mobile input devices, the Project uses hand-held computers that run Microsoft Pocket PC operating system, Sony Mavica digital cameras that store photos on 3.5” floppy diskettes, and a Global Positioning System (“GPS”). These devices are used on-the-go by residents to gather information regarding neighborhood conditions, literally putting the task of identifying these types of problems in their hands. To store and organize the information residents gather, the Project uses traditional computer databases. Lastly, this raw information is processed into a more meaningful, easier-to-understand form by Microsoft Access 2000 and ESRI’s ArcGIS software. Microsoft Access creates reports that summarize the information in the form of statistics and are more traditional in form. The ArcGIS software creates maps, plotting the geographic location of each problematic condition on area maps. The reports and maps are tools used by local government, city officials, and community residents to address neighborhood problems and improve delivery of city services.

The practices of the Project are also a part of its model. The input devices are distributed in some cases on “project days” and in other cases for longer, but still finite, periods of time. Project days are supervised events where volunteers target specific areas and “scan” as many problematic conditions as time allows. Where the devices have been distributed for longer periods of time, such as to high-school students in a part-time summer job program, scanning is not as structured, but has the added benefit of providing before-and-after documentation. In either case, the information collected is transferred from each device to a desktop computer where it is “synced,” or the hand-held computer entries, digital photos, and GPS data are combined into a single file documenting a neighborhood condition. The data entries are then transferred to the central database where, at a later time, the processing software is applied and the output reports and maps are produced. This model, the combination of the Project’s technological components and its practices, has been successful at reaching the goals noted above, but it has become outdated. Understanding what needs to be changed, requires taking a closer look at how and why the Project works as it exists today.

The mobile devices used in the Project allow residents to enter text describing the nature and location of a problem, to take a digital photo of it, and, using GPS, provide the exact latitude and longitude for those problems in areas without street addresses, such as parks. These devices are capable of providing great detail, which is important to support meaningful discussions later. They are, generally speaking, relatively expensive and not widely used. Ensuring that the resources will be there to purchase these devices and to train residents unfamiliar with their use tends to require centralizing ownership. In fact, this has typically been the case with the Project. The devices have been purchased or leased in small numbers by the city or organizing group who then retains possession of them. Centralized possession has the distinct advantage of creating authority in the possessor to choose when and where scans take place, thus, ensuring that all areas or, at least, the areas of concern are actually scanned. Similarly, insofar as the possessory group is supervising device distribution and training, the authority is created to control what information is gathered. Generally, the supervision that is possible under centralized possession enables very effective quality control over the input of information. The net result guarantees that desired information will be gathered from all areas of a city at a minimum level of detail.

The first proposal for improving this model is to choose cell phones as the input device, taking advantage of their existing widespread possession. This change would both reduce the cost of project adoption insofar as the cost of providing input devices is eliminated and better serves democratic interests to the extent participation is opened up to the public at large. Further, choosing an input device that residents already use eliminates whatever costs may arise in familiarizing the public with use of the system. Where residents already own the input device and already know how to use it, the initial cost of participation is minimized creating a potential for a great increase in participation. Increased participation is desirable, because it produces a more representative and much larger dataset, both of which generally tend to increase its value.

It is important to ask whether using cell phones fits within the standards of the Project’s model. At first glance, choosing cell phones appears to come at the cost of detail, because they are not capable of text, photo, and GPS as are the devices used in the Project. As noted above, this type of detail is important to support meaningful discussions later. The level of detail may indeed suffer at the level of individual records, because the quality of photos taken by cell phones is relatively low and virtually none provide GPS functionality necessary for locations without street addresses. This shortcoming should be corrected by the increase in participation, which generally tends to increase the value of the dataset. Additionally, a cell phone has the added functionality of allowing an individual to call and speak with an expert operator, or even an automated expert system. Relying on an operator or automated system could further improve the level of detail by simply instructing the operator to ask for specific information. The net effect of the loss of individual detail and the gain of larger datasets should be a level of detail at least on par with the existing system.

Decentralizing possession of the input devices destroys the authority to control when and where a project is deployed. This is indeed true, but, in the same way the increase in participation corrects for the lower level of detail, the increase in participation should cause a natural spread of scanning across all areas of interest. To the extent that needs go unmet, it would be a simple matter to inform the public of any shortcoming and request special attention to these needs.

One potential criticism of decentralizing possession of the input devices is that the loss of authority to supervise creates the opportunity for gaming of the system. Here again the overall increase in participation should compensate. To the extent the system cannot distinguish and disregard such participation, the volume of good faith participation would likely subsume all but organized gaming efforts.

The second proposal for updating the Project’s model is to change the administration of the database and approach to information-processing. Access to the database and information processing in the existing model are the exclusive domain of CPEC employees or other experts. The benefit of relying on experts to carry out these tasks is that they can produce consistent and high-quality output reports and maps, which are, similar to ensuring that the information collected in the first place is high quality, important to supporting meaningful discussion later. But, technology has made the public equally capable of producing consistent and high-quality output.

The sophisticated tools necessary to process information have become more widely available and simpler to use. Updating the model of the Project to account for this would mean granting the public “open access” to the database. Here, “open access” means granting access to it over the internet and publishing the documentation on its technical specifications. Shifting information processing to the public will further reduce the cost of project adoption, because the cost of purchasing licenses for the information-processing software—Microsoft Access 2000 and ESRI’s ArcGIS—and the cost of paying experts to use the software become unnecessary. Freely available or open source alternatives, such as Google Maps (http://maps.google.com/) or OpenReports (http://www.oreports.com/) could be used instead. In fact, there is no reason the public and the city cannot both use these free alternatives.

Open access allows a project to enjoy the benefits of new technology as it becomes available, because there is no longer a reason to be entrenched in commercial software packages. Today, Google Maps is a favorite free service, but it will likely be eclipsed at some point by something better. When it is, the public will begin experimenting with it and, if it makes the data collected by a project easier-to-understand, it will be naturally adopted. In this way, open access allows a project to bootstrap itself to the cutting edge of technology development, ensuring that the benefits of recent innovations are quickly realized. Further, as more cities adopt projects relying on such products as Google Maps or OpenReports, a new cutting edge will emerge—one oriented around the specific needs of this type of project.

As more people experiment with new ways to process this information, the visibility of a project will increase, drawing more participation. Participation will increase both at the input side and at the information-processing side. As described earlier, the value of the data is improved by increased participation in input, but increased participation in information processing is important for a different reason. At the moment, the model of the Project produces information in two forms, reports and maps, but the public is a diverse group. The form of information best for one person may be very different from the best form for another. Increased participation at the end of information processing will allow people to create and experiment with a variety of forms, satisfying the interest in providing equal opportunity to partake in the discussions that the information is supposed to support. Equal access to this information is not an explicit goal of the existing Project, but the significance of the principle to democracy demands note. Over time, a project would support increasingly rich dialogue and, as more cities develop similar projects, government performance of one city may be compared against others.

The model of the Project is simple and effective, but technologically out of date. Allowing the public to use technology to gather and organize information regarding physical neighborhood conditions enables better coordination of city services addressing neighborhood problems and promotes resident involvement in problem identification and problem solving. Modern technology could reduce the costs to the city of adopting such a project and provide other benefits. Future projects would benefit from choosing widely available devices for information input, increasing participation and the overall quality of information, and granting open access to the database, promoting more informed dialogue.