Exploring Local
Mike Dobson of TeleMapics on Local Search and All Things Geospatial

A Street and Road Information Clearinghouse

May 6th, 2010 by MDob

Well, Mr. Swanson never called back, so the world will have to wait to see whether Foxboro attempts to start the ball rolling on the road “clearinghouse” concept in Massachusetts. I guess I will have to go it alone on that topic.

The basic premise, carried forward from my last blog, is that a local jurisdiction “should” know more about the streets and roads for which it is responsible than the governmental unit that contains it. So, the city of Laguna Hills should know more about its streets, than agencies representing Orange County, or those of the state of California, or the federal government of the United States. Now, there are obviously some exceptions. For example, counties should know more about county roads than any other jurisdiction (even the city governments through which these roads pass) and states should know more details about state roads than any other level of government. In addition, states may be uniquely qualified to know more details about interstates in their jurisdiction than the federal government, since states are often responsible for the construction and the spend that creates additions and augmentations to these roadways.

Let’s not get confused by the fact that various jurisdictions may create maps or databases that may include some streets and roads, which they know little about. State highway departments often fall into this class, as they provide online maps of the state that faithfully report state highways and interstates, but often have egregious errors when representing local streets and roads. Perhaps this distance decay in reporting is an offshoot of Tobler’s First Law of Geography that “Everything is related to everything else, but near things are more related to each other.” If municipalities have verified change information about their street and road infrastructure, what better source would exist to propagate that information to mapping companies and others who create, update and manage spatial databases, particularly those used for navigation?

It goes without saying that these “local” jurisdictions should know more details about the streets and roads for which they have governmental responsibility and know of these changes that alter their transportation network sooner than, say, that knowledge is discovered by NAVTEQ, TeleAtlas, Google or OSM. In fact, since most roadwork depends on budgets and budgets must be approved and publicly represented, it is likely that many road and street changes are knowable far in advance of the construction that alters their physical structure. I bring out this issue, because field operations are those designed to monitor what is, not what will be. Consequently, databases based on observations, such as OSM, are often limited by what people can observe, rather than what they could know if they looked in the right place.

Wait a minute – if all this local knowledge stuff is true, why not just build a roads database based on the contributions of these administrative units? Well, I think that strategy is, in part, what GDT was attempting to accomplish in the 1990s when it was creating its geodatabase of the United States (Dynamap). GDT, for example, merged its existing database with the first version of TIGER and then exhibited a preference for the use of data mining, rather than field research to enhance Dynamap.

I suspect that GDT learned that: 1) many jurisdictions don’t do a good job of inventorying their streets and roads, 2) information representing the same street segment may vary between agencies within a jurisdiction, 3) numerous jurisdictions are not computerized, and 4) others jurisdictions simply cannot afford to update their spatial or information databases in a timely manner. When GDT later contemplated offering a navigation capable database, as opposed to a simple geospatial database, I think the company’s management began to realize that independent “eyes on the road” and field measurement were a required part of map updating, at least if you want to build a navigation-grade spatial database.

So, it is my opinion that building a navigation database requires a variety of redundant inputs; a database built solely on reports from interested municipalities could not be used to create a viable navigable database. Today, inputs from states and municipalities are being used by Google, NAVTEQ, TeleAtlas and others as part of the normal update process. Unfortunately, what seems to be lacking is that navigation database provider’s do not have a perspective on what is going to happen to the transportation network and when the alteration actually happens. While these types of changes can be discovered in due time through field operations or UGC-based systems, it just seems to take too long for these groups to discover, describe and integrate these changes into their databases.

Our recent series on the problem in Providence, Rhode Island, although good for a laugh, is a textbook example of how the lags in the map updating system can provide data that is not fit for the intended purpose. Below is a photo of the now infamous I-195 realignment in Providence that was snapped by a friend on Tuesday of this week, as his airplane began its descent into Boston. Look – there are cars on the new segment and the old one no longer connects to I -95 on the west. Below that is view of the same intersection on NAVTEQ’s website as of 1735 (PDT) today (May 6, 2010) and I am sure we are all glad to note that, at least on the map, traffic is flowing at normal speeds on the roadbed that is being deconstructed. Notice that traffic is moving at normal speeds even where the roadbed no longer exists. (I have wanted one of those vehicles ever since I saw George Jetson sporting about in one many years ago).

Yep, it's Providence and there is the new highway.

NAVTEQ just doesn't believe me.

George Jetson, his family and the space car I covet

So, if municipalities knew of road and street changes in their jurisdiction that would influence navigation and navigation systems and communicated those to an official clearinghouse, these data could be available, often in advance of the change, to any interested party. The repository of change information, would in all likelihood not be of the quality to allow any commercial creator of navigation databases to adapt it without further research and processing. However, it could and would provide more rapid map updating than is available today.

Maybe this stuff is not of interest to NAVTEQ, since their business model seems to be built on in-car systems whose owners rarely buy new updated navigation media after they have purchased their car. TomTom on the other hand, should be interested in this method, since its PNDs are updated more often by its users and it is now making some inroads with Renault and car companies in outfitting vehicles with TomTom units that are dash-dockable. On the other hand, TomTom’s actions appear to indicate that it believes that UGC (MapShare + probe vehicles) may be a better answer to map updating than relying on professional field observation. Indeed, they may be convinced that UGC reveals these types of changes more quickly than “authoritative” sources such as local government.

So how would this work? (The short version)

My suggestion is to create a nationwide clearing house that operates a secure, online system provisioned to allow certified sources to check in map data and/or other information about roads changes and proposed road changes in the areas of their jurisdiction. Non-certified parties would not be able to submit or alter information. Any interested party would be able to download these data and use them without limitation.
The data would be proffered without guarantees of accuracy or fitness for a particular use. Instead, the platform would serve as an early warning system, allowing interested parties to focus on road and street changes that they might not become aware of until far in the future.

Those of you still awake will notice that I have proposed a clearinghouse even though some providers of “local” road information are not very good at reporting these changes. Well, that is true, but in this case, something appears to be better than nothing. As to why someone did not suggest a clearinghouse in the past, well, nobody was providing navigable databases on a scale that made a difference. With online routing, PNDS and a surge in popularity of in-car systems, updating maps to reflect current reality has become a much more prominent problem.

What do you think? Let’s have some comments.

By the way, I did not have time to get to my prediction about the ascendency of local mapping firms and may do that next time. However, I have this idea about maps and cyber war that may trump it. In addition, I have been doing a lot of research on addresses and addressing and want to discuss that as well. Tune in next time to see where we end up. Did I mention my iPAD 3G arrived last Friday – more toys!

Click for our contact Information

Bookmark and Share

Posted in Authority and mapping, crowdsourced map data, Data Sources, Google, map updating, Mapping, Navteq, openstreetmap, OSM, Personal Navigation, routing and navigation, TeleAtlas

One Response

  1. Anon

    Many municipalities charge for data, so why give it away for free in a clearing house. Assuming there was a clearing house, there would should be a standardizing of data posted, at least that would be the intelligent thing to do. Standardizing would probably require retooling, training and infrastructure support at all government levels. Before the government agency moved on this a feasibility study($) would have to be done to ensure there’s ROI before committing. At a time when governments (local, state, etc.) are struggling with budget deficits, adding additional overhead is probably not a welcome mat that’s going to be thrown down any time soon.

    Thanks for your comment.

    I suspect that public safety could be a factor that would encourage municipalities to find a method of communicating change information.

    In addition, I think that a lightweight protocol could be used to communciate a minumum amount of information to call attention to the new condition.

    Next. inititiatives like the FDGC 50 State Initiative (designed to promote the National Spatial Data Infrastructure), as well as the need for communities to participate in the LUCA program and BAS programs of the U.S. Census may help develop competencies in digital mapping and GIS that were previously lacking.

    Thanks again,

    Mike