Why isn't MLS using the U.S. National Grid?

Discussions around using and interfacing with the Church MLS program.
skiptaylor
Member
Posts: 190
Joined: Fri Mar 23, 2007 10:13 pm
Location: Lancaster CA

Why isn't MLS using the U.S. National Grid?

Postby skiptaylor » Mon Oct 06, 2008 9:48 am

I've been attempting to work with the built-in GEO Codes fields in MLS and figuring out what to put in there. This can be a daunting task, and has resulted in a short crash course on different types of mapping and ideas that others have used to GEO Code their Ward or Stake.

I have come to the conclusion that there has been no standard adopted by the Church. Perhaps it's time we start to think of implementing a Church standard for this? I think that a likely candidate is the United States National Grid system.

Right now, each Ward has it's own GEO Coding idea, then each Stake has it's own. That is a form of mass confusion in my mind. Should an emergency happen, just how long will it take to become familiar with the Ward's GEO Coding system? Would you be able to tell someone in another Ward or an emergency service just how your GEO Codes work so help needed at a particular location could get there? Perhaps we should stock different colored flares? :D

In all my recent reading on this, I have run across several national as well as worldwide grid systems. I recently learned the United States has it's own National Grid system called (oddly enough) the United States National Grid. It has the ability to pinpoint a location that's easily understood. A good explanation can be found in the poster at http://www.fgdc.gov/standards/projects/FGDC-standards-projects/usng/Skagit-County-USNG-Poster.pdf (The poster is large, and probably best to try to read online vice printing it) This illustrates the use and potential for the USNG. From what I've read, but not researched yet, many of the newer under $100 GPS units have the ability to work with the USNG as well as the traditional grids/coordinate systems.

The USNG's project page is at http://www.fgdc.gov/standards/projects/FGDC-standards-projects/usng/index_html and appears to have been completed in 2001!

There are a lot of information sites, one shows an example of the system using Google's Mapping API at http://www.fidnet.com/~jlmoore/usng/ - further information and a pretty good explanation, as well as programming code/examples, is at http://www.fidnet.com/~jlmoore/usng/help_usng.html More info is also at http://mississippi.deltastate.edu/

I'll make this short, but once you have the area and region down (basically prefixes), the following grouping numbers would have the following accuracy.

4 digits - 2306 - locates a point with a precision of 1,000-meters (a neighborhood size area).
6 digits - 234064 - locates a point with a precision of 100-meters (a soccer field size area).
8 digits - 23480647 - locates a point with a precision of 10-meters (the size of a modest home).
10 digits - 2348306479 - locates a point with a precision of 1-meter (within a parking spot).

I'd like others thoughts on getting this implemented in MLS.
Skip Taylor :eek:
Lancaster CA.

User avatar
kd7mha
Member
Posts: 156
Joined: Thu Mar 13, 2008 1:27 pm
Location: Logan, Utah

Geo codes

Postby kd7mha » Mon Oct 06, 2008 11:01 am

The GEO codes in MLS were never intended for precision mapping but rather to divide up the ward into segments to be used for boundry realignment.
This topic has however been discussed elsewhere on this forum, do a search for geo codes and you will find several threads on this subject

skiptaylor
Member
Posts: 190
Joined: Fri Mar 23, 2007 10:13 pm
Location: Lancaster CA

Postby skiptaylor » Mon Oct 06, 2008 11:08 am

I guess I'm not making myself clear.

I understand the use of GEO Codes as intended in MLS. That's not what I'm getting at.

What I meant to say (but I guess I didn't) is why can't MLS calculate the GEO Code for us, based either on address or the National Grid. With the USNG, if it shows on a map, it can be calculated for the grid address and we could input the grid address for those without an address to calculate from.

It just seems in my mind to mesh better not only with emergency services when needed, but also with stake boundary realignments etc. Would save a lot of work that's duplication of effort is my thought.
Skip Taylor :eek:
Lancaster CA.

User avatar
aebrown
Community Administrator
Posts: 14693
Joined: Tue Nov 27, 2007 8:48 pm
Location: Sandy, Utah

Postby aebrown » Mon Oct 06, 2008 11:22 am

avskip wrote:I guess I'm not making myself clear.

I understand the use of GEO Codes as intended in MLS. That's not what I'm getting at.

What I meant to say (but I guess I didn't) is why can't MLS calculate the GEO Code for us, based either on address or the National Grid. With the USNG, if it shows on a map, it can be calculated for the grid address and we could input the grid address for those without an address to calculate from.

It just seems in my mind to mesh better not only with emergency services when needed, but also with stake boundary realignments etc. Would save a lot of work that's duplication of effort is my thought.


I don't see how a grid is helpful for boundary realignment and many of the other uses of GEO Codes. You would be hard pressed to find any ward in the Church where the boundaries are laid out on a grid, and there are good reasons for this. There are natural boundaries (streets, rivers, subdivisions, school districts, etc.) that may all be helpful for dividing wards (or for laying out fast offering routes or some of the other uses for GEO Codes). In Utah, for example, a grid system might well put neighbors who share a back fence in the same GEO Code, but if they live on different streets, they should probably not be in the same code. Similar analogies would occur on a larger scale in other areas.

GEO Codes have essential uses that cannot be replaced by nor generated by a grid system.

However, you raise some good points about emergency services. Others on this forum have proposed that it would be very helpful for mapping purposes to have a latitude/longitude field in MLS. There is definite merit in this proposal. It seems to me that the National Grid system that this is essentially the same information, but using a different system. Lat/Lon seems more flexible to me, as it can be used as input to mapping systems, but I am not familiar with USNG. Could you comment on the relative merits of USNG vs. lat/long?

jbh001
Community Moderators
Posts: 854
Joined: Thu Mar 13, 2008 5:17 pm
Location: Las Vegas, NV

Postby jbh001 » Mon Oct 06, 2008 11:23 am

Thanks for the clarification. So for an over-simplified example, you could potentially tell MLS somewhere that all USNG codes between, say 2400 and 2500 get assigned to Ward GEO code 01. or for more presicion, USNG codes between 2400000000 and 2500000000 get assigned to Ward GEO code 01.

I like that idea, becuase it helps automate and simplify the process.

RossEvans
Senior Member
Posts: 1346
Joined: Wed Jun 11, 2008 8:52 pm
Location: Austin TX
Contact:

Postby RossEvans » Mon Oct 06, 2008 12:19 pm

avskip wrote:What I meant to say (but I guess I didn't) is why can't MLS calculate the GEO Code for us, based either on address or the National Grid. With the USNG, if it shows on a map, it can be calculated for the grid address and we could input the grid address for those without an address to calculate from.


In fact, Church developers are known to be working on a solution to geocode all households by lat/lon coordinates somehow, which someday will be captured in new fields in the membership record. My expectation is that when that solution is rolled out, it will support online member mapping on a website -- presumably the LUWS to take advantage of authenticated logins -- and also give us some capability to export those coordinates.

However, that is not the same thing as calculating and populating the ward or stake "GEO Code" fields. These fields are intended to contain arbitrary codes representing geographic areas defined locally. They do not represent point coordinates, but rather area polygons someone has drawn on a map.

The optimal choice of such polygons varies greatly with local circumstances. Sometimes local units can piggyback on the work of external sources such as the U.S. Census to define such jigsaw pieces in useful ways that make sense on the ground; other units are obliged to define their own. What works for dense wards in the Wasatch Front, urban/suburban wards elsewhere, or rural wards likely would be different solutions. Units outside the United States will have their own circumstances. Fixed grids are probably the least desirable choice

Deriving the two "GEO Code" fields automatically from lat/lon cordinates is theoretically possible, but would probably requre MLS or some other host system to support machine-readable definitions of such polygons, presumably imported in common file formats such as GIS shapefiles (.shp) or KML files. I doubt that capability, which begins to extend into basic GIS functionality, will be forthcoming soon, but perhaps I will be pleasantly surprised. It would probably be more feasible to implement such functionality in a central site rather than incorporated into the distributed MLS desktop software.

FYI, see also this recent thread.

skiptaylor
Member
Posts: 190
Joined: Fri Mar 23, 2007 10:13 pm
Location: Lancaster CA

Postby skiptaylor » Mon Oct 06, 2008 12:26 pm

Alan_Brown wrote:However, you raise some good points about emergency services. Others on this forum have proposed that it would be very helpful for mapping purposes to have a latitude/longitude field in MLS. There is definite merit in this proposal. It seems to me that the National Grid system that this is essentially the same information, but using a different system. Lat/Lon seems more flexible to me, as it can be used as input to mapping systems, but I am not familiar with USNG. Could you comment on the relative merits of USNG vs. lat/long?

I would say simplicity is the USNG's strength.

The way the USNG is laid out, it's in squares (grids). To find a location on a map, it's a matter of something similar to going to map grid A1 but with more numbers depending on accuracy needed. For this, Lat/Long would be great, but a LOT slower.

If you look at the one page PDF at http://www.floridadisaster.org/gis/usng/Documents/Reading_USNG_FL.pdf it will tell you a LOT more faster and better than I can. I think you will see the simplicity of the USNG vs. Lat/Long for map usage for yourself. It may take a minute to sink in, but when the light comes on you will like it.
Skip Taylor :eek:
Lancaster CA.

skiptaylor
Member
Posts: 190
Joined: Fri Mar 23, 2007 10:13 pm
Location: Lancaster CA

Postby skiptaylor » Mon Oct 06, 2008 12:28 pm

jbh001 wrote:Thanks for the clarification. So for an over-simplified example, you could potentially tell MLS somewhere that all USNG codes between, say 2400 and 2500 get assigned to Ward GEO code 01. or for more presicion, USNG codes between 2400000000 and 2500000000 get assigned to Ward GEO code 01.

I like that idea, becuase it helps automate and simplify the process.

The USNG can be CALCULATED from Lat/Long coordinates which can be calc'd from an address. I am unaware of the actual math, but there are examples on how it's done. So potentially, as long as there's an address, MLS can tell itself.

For those without an address, you would want to input the USNG manually.
Skip Taylor :eek:
Lancaster CA.

skiptaylor
Member
Posts: 190
Joined: Fri Mar 23, 2007 10:13 pm
Location: Lancaster CA

Postby skiptaylor » Mon Oct 06, 2008 12:45 pm

boomerbubba wrote:Fixed grids are probably the least desirable choice

I guess I've lived in a flat area too long (38 yrs). For the major part, all roads here are one mile apart east/west and up to 16 divisions within a mile north/south. The basic area is a one mile square with subdivisions within the square. So one gets used to thinking in grids with all that.
Skip Taylor :eek:
Lancaster CA.

RossEvans
Senior Member
Posts: 1346
Joined: Wed Jun 11, 2008 8:52 pm
Location: Austin TX
Contact:

Postby RossEvans » Mon Oct 06, 2008 1:37 pm

avskip wrote:I guess I've lived in a flat area too long (38 yrs). For the major part, all roads here are one mile apart east/west and up to 16 divisions within a mile north/south. The basic area is a one mile square with subdivisions within the square. So one gets used to thinking in grids with all that.


Flat, straight roads are more the exception than the rule in geography, not to mention complications like rivers, school district boundaries, Census boundaries, postal zones etc., that are siginificant for various purposes. And even if the roads all run due north or due east, there is nothing that makes them coincide exactly with the grid lines of the USNG. That grid system has is uses, but defining ward boundaries for all of us is probably not among them.

Several years ago my former stake used a straight-line grid system -- this one happened to be defined by a commercial mapping company based on a local city grid-- as a crude definition to populate the "GEO Code" field. But at best, when calculating proposed boundary changes, this field yielded only a rough apporoximation of where the boundary lines actually got drawn. That is why the general recommendation is to define jigsaw pieces that make sense on the ground locally. I believe my stake now uses Census tracts or block groups, which are anything but square. (I also don't think anyone bothers to enter them manually into MLS; the stake clerks use a real GIS system instead.)


Return to “MLS Support, Help, and Feedback”

Who is online

Users browsing this forum: No registered users and 1 guest