Podcast: Exploring NGAC and NSDI with Guest Host David Smith
By Joe Francica and Adena Schutzberg
February 10, 2009
Classified Ads:
Take advantage of a special year-end sale on SPOTMaps, the 2.5 meter, seamless, color mosaic made to fit your area of interest. Save 25% off all SPOTMaps through November 10th, when you mention this ad! Click here for detailsGGuest host David Smith of Synergist Technologies joins Executive Editor Adena Schutzberg to explore last week's National Geospatial Advisory Committee meeting and the discussions regarding a National Spatial Data Infrastructure and how these issues might play out in the federal stimulus bill.
 |
 |
 |
 |
|||
|
||||||
|
 |
 |
 |
|||
Show Notes
This year ESRI is offering two code challenges in conjunction with the ESRI Developer Summit, the ArcGIS Server Mashup Challenge and the ArcGIS Mobile Challenge. Enter your original code samples by March 6th for a chance to win a cash prize. The challenges are open to all qualified developers. For more information visit the the Code Challenge website.
|
Your Comments Post a comment All comments provided in this section are those of the individual who has created the post. These are not the opinions of Directions Media, its editors, staff or owners unless otherwise noted. Directions Media retains the right to edit or delete any comments posted herein.
|
|
||||||
| An interesting discussion; good to hear a guest perspective. The core issue at hand is "what benefits the greatest community" of users? The NSDI (as recounted by the NGAC, and geospatial cognoscenti that have offered up solutions in the last month) is an aggregation program...a massive undertaking to build a comprehensive, cohesive, and integrated map. Well, it didn't, doesn't, and won't work. The thousands of 'owners' of GIS data at the town, city, county and regional level have virtually no interest whatsoever in contributing to a national program with no payback. It costs real money and staff time to package and deliver the data - no matter whether it's through a Service-Oriented Architecture (salvation!!) or by good old sneaker-net and snail-mail on disks. Then there's the ongoing integration and assimilation process at the Federal level...who will do that? The 137 people left standing in the USGS who know anything about geospatial production work? (sorry folks, not meant as a slight, just making a point) Not likely; look at the USGS' share in the just-passed Senate stimulus package - the Senate CUT 40% of the original amount for USGS that was in the House version...from $235 to $135 million. Okay, so pay the locals to host the data! say the stimulators. Exactly. But here's where the story gets more complicated. For SOA to work correctly (salvation!!) one needs a catalog of the contents. Simple, say the stimulators...we have the Geospatial One Stop. Well, GOS is dead end. They have tens, maybe even hundreds, of regular monthly visitors. And the vast proportion of them are probably bots from Microsoft and Google and Mapquest trolling for new content. In fact, GOS is so roundly ignored the NSGIC folks went out and built their own to try and get the job done (RAMONA, - http://www.gisinventory.net ). It's working better, but even RAMONA could use a major infusion of assistance to be really useful. In the event you're still reading this, it's likely you're thinking this is all negativity and no solution. Not so. The solution is to NOT try and assimilate and produce the Mother Of All Databases, but instead fund local users with sufficient resources so they can store and manage their geospatial resources in a secure, private, and NOT SHARED space inside a State- or Regional-level data center. Yes. Private. Not shared. Protected against acquisition by the Federal Geographic Whoever groups. Protected so MY CITIZENS INFORMATION REMAINS PRIVATE. Well, that's certainly not productive, say the stimulators and SOA (salvation!!) converts and evangelists. How does that serve the public? The fact of the matter is virtually every local/regional GIS program has its own unique set of legal and bureaucratic hurdles that are required before something produced with taxpayer funds can be published and/or shared. This is the hard part...the data owner and data publisher need to negotiate and agree on what can, and what can't, be used. Multiply that by thousands of programs and XXX number of data products per GIS and pretty soon the project starts to look darn complicated. This is one reason why this hasn't happened...it's too darn complicated. The answer is to build a strongly secure and openly available data center where each program can keep a clean, active-active copy of each and every file they manage. A mirror of the local program. And to FUND an ONGOING program to keep that alive...not offer up 'seed' money that dries up in 3 years after the Fed in charge moves on to other things. Participation in that storehouse would require that the system is made available to emergency workers when an emergency occurs. Authorized, competent, GIS workers who need good quality data to do their work right. This would be a HUGE leap forward from where we are today. Once the concept of co-location begins to settle comfortably on folks, they might (just maybe) agree to list 'some' of the data for discovery through a national catalog and publish through a (salvation!!) SOA node on some standards-based network paid for by Federal monies. Or maybe they should just give a copy to Google to burn onto that globe. Today we have virtually no reciprocity of in-depth geospatial products at a national level. Some states are trying (trying...) with their limited funds to store and serve some basic products through clearinghouses. That's good work, but not nearly the grand vision folks forsee. And not the detailed products that could be available through an array of secure data stores. My bottom line - stop the National-this and SOA-that and start serving the citizens where the stimulus will do the most good. Right there at every city, county and regional GIS, at the micro level. Where the work really gets done. And give these folks something that's useful. How much would all this cost? Simple math: 3,140 counties 3,140 city GIS's worth counting (we'll make that number up, it's likely to be different, or could fall under County products) 3,140 Regional GIS's (same as the above number) 392 State-level GIS's (56 state & territories, each with 7 large GIS programs) 9,812 total number of submitting organizations $250,000 Annual support paid to each submitter $1,000,000 Annual support paid to each of the 56 Data Centers, each of which also receives shared costs from each submitter to operations/maintenance Total system-wide annual cost: $2.5 Billion That's twice what ESRI requested in their opening paper. And on the relative measure in an $838 Billion package, it ain't much. $2.5 billion isn't a lot of money to fund high-quality local employment, strong local investment in IT support, meaningful local contributions to useful aerial photography, and overall improvement of the local GIS programs that can most use the money. I might add this will be money spent in local tax rolls, which further bumps the needle. Now THAT is stimulus and sensible investment in our geospatial information infrastructure. |
||||||
|
||||||
| Steve - I believe that your statement that "The NSDI (as recounted by the NGAC, and geospatial cognoscenti that have offered up solutions in the last month) is an aggregation program...a massive undertaking to build a comprehensive, cohesive, and integrated map." is not correct in terms of the NSDI 2.0 proposal. That group is suggesting a services approach in which geospatial content remains with the authoring organization (the owner) and is provided to users using open and standard interface, protocols, and encodings. As an example, if you look at the Canadian Geospatial Data Infrastructure (CGDI), you will notice that the government provides the guidance, reference implementations, technical architecture, seed dollars, and policy infrastructure. Since the architecture is standards based, any organization can (and they do) at any time provide access to whatever content and possibly related services they wish to as part of the CGDI. The costs are considerably less than you suggest - by orders of magnitude. http://cgdi.gc.ca/en/index.html . There are a number of ROI studies available from the CGDI community. From the CGDI webpage: "Sharing data from its source reduces costs and improves decision making The CGDI's success depends not only on data owners and custodians adhering to CGDI data standards, but also on them sharing their data with others online. Sharing data directly from the source ensures that users get up-to-date, accurate, and authoritative information quickly and easily. For example, if an organization that collects, say, air-quality data sends a copy of its database to another organization, as time passes, this copied data will become out of date. That's because the air-quality data provider will often update its database, but these changes will seldom be reflected in the copied database. In that case, the copied database becomes increasingly less useful to decision-makers. By keeping data “closest to its source” and allowing others to access this data online—key principles of the CGDI—data owners and custodians ensure, however, that everyone always works with the most up-to-date, accurate data. This characteristic lessens the cost and effort of managing data and improves decision making—the CGDI's primary objective." Further, your suggestions that "The answer is to build a strongly secure and openly available data center where each program can keep a clean, active-active copy of each and every file they manage." is totally contrary to just about every successful SDI activity that I am aware of. Check out this report http://sdi.jrc.ec.europa.eu/ws/Advanced_Regional_SDIs/arsdi_report.pdf. There are similar activities in many Asian, Latin American, and African countries. Finally, your suggested approach does not deal at all with how to work with real time content generation and processing, such as from in-situ and dynamic sensors. All of the above examples are serving extremely large communities of interest! Cheers |
||||||
