Publishing location based data in Easting and Northing, Longitude and latitude, or Addresses?

+2 votes
16 views
asked Aug 4, 2015 in Open Science by Gram (185 points)

I am working on publishing a listing of shelters and hostels, I would like to expose my data to anyone who would like to use it. One of the things I would like to do is allow the shelters and hostels to be plotted on a map.

There are three methods of posting locational based data:

Addresses

This is the traditional street and Avenue system that the postage system uses.

Easting and Northing

These are (usually) used for horizontal and vertical from the a specified datum position. For example, a mountain close to where I live is located at 50° 52′ 10″ N, 115° 39′ 3″ W.

Longitude and Latitude

This is a common grid system that uses two numbers (a coordinate) to pinpoint a location. For example, the same mountain can be fount at 50.869444, -115.650833.

My thoughts so far

The addresses system good for projects in a city but if the location is away from a road (like some hike-to hostels) it does not work so well. My data is already in Easing/Northing, however with everyone using GPS and cell phones now it sounds like longitude and latitude may be more popular?

Question

What type of location system should I chose if I want as many parties as possible to use my data in their work? It's hard to know how the data could be used, but I suspect consumers of my data to be interested in travel, real estate, poverty and government.



This post has been migrated from the Open Science private beta at StackExchange (A51.SE)
commented Aug 18, 2015 by Gram (185 points)
I didn't want to have the risk of incomplete data, that being that someone would just enter the address and think they didn't need a coordinate due to the address or the reverse.

This post has been migrated from the Open Science private beta at StackExchange (A51.SE)
commented Aug 18, 2015 by Michael (140 points)
Why not both address and coordinates (whichever you choose)?

This post has been migrated from the Open Science private beta at StackExchange (A51.SE)
commented Aug 18, 2015 by Gram (185 points)
Yes, they are almost the same thing, however from the viewpoint of someone who wants to use the data I wonder if they would have a preference. To convert one to the other takes some time to implement in a programmable matter. :)

This post has been migrated from the Open Science private beta at StackExchange (A51.SE)
commented Aug 18, 2015 by Simon W (155 points)
You're probably aware of this, hence I haven't included it in my answer, but note that the last two methods that you give are basically the same - they're both giving latitude and longitude, but one gives it in degrees, minutes and seconds, with N and W to give direction, while the other gives it in degrees with a sign convention to give direction.

This post has been migrated from the Open Science private beta at StackExchange (A51.SE)

1 Answer

+5 votes
answered Aug 4, 2015 by Simon W (155 points)
 
Best answer

This is very problem-dependent, but if the most likely use of the data is for positions to be plotted on a map (rather than, for example, sending mail to those places) then the data should be given as positions, rather than street addresses. This way they are unambiguous, wheras addresses may change over time, and the locations that correspond to them may depend on what lookup tool is used to convert addresses to locations.

If you're giving location data, then the most important thing is not the exact format that you give it in, but that it is clear what the format is. Coordinates should be accompanied by metadata explaining what the coordinate system is. For example,

  • If using lat/lon, you should ideally note what geoid is referred to (usually WGS84 these days, since that's what GPS uses, but best to specify if you know).

  • If it's something more complicated (e.g. UTM coordinates) then that needs to be clear and the appropriate metadata given.

  • If vertical elevations are given, it's important to know what datum is used (i.e. where is zero elevation).

With regard to your specific dataset (and without knowing the details of the likely uses) I would recommend giving position data rather than address data, and (less importantly) expressing that in degrees rather than degrees / minutes / seconds, because in many cases it will be easier for software to parse.



This post has been migrated from the Open Science private beta at StackExchange (A51.SE)
commented Aug 18, 2015 by Gram (185 points)
Thank you for the explanation, you have provided insight that I did not consider. From what you have explained it makes more sense to use a coordinate system as I suspect it will be the most useful in the applications and articles that may consume my work.

This post has been migrated from the Open Science private beta at StackExchange (A51.SE)

Welcome to Open Science Q&A, where you can ask questions and receive answers from other members of the community.

If you participated in the Open Science beta at StackExchange, please reclaim your user account now – it's already here!

e-mail the webmaster

...