Should I geotag the location of the subject or the photographer?

Question

If I take a photo of a distant object with a zoom lens, and then geotag it manually, would it be more useful to tag the location of the distant object (what is seen in the photo) or the location from where the photograph was taken?

For example, if I take a photograph of Alcatraz from Coit Tower in San Francisco, should I tag that photo with the location of Alcatraz or Coit Tower?

Answer 1

Most of the time you have more use of the position from where you took the photo.

If you know from where the picture was taken, you can often from the photo see exactly which direction the camera was pointed. If you know the position of the subject, you might be able to see approximately which direction it was taken from, but seldom the exact position.

Of course, if you know that you will only ever be interrested in the location of the subject, you can discard any other information. Geotagging is always a bit of a compromise; it would be nice to have an exact three dimensional vector from the position of the camera to the position of the subject, but we only tag one point, and only in two dimensions.

Answer 2

There's no "right" answer to this question; that said I would take a look at why you geotag your photos. Depending on the answer I could make arguments either way:

• If you geotag photos to keep a record of where you've been then it probably makes sense to tag them with the location of the camera. Using your example, as you view your geotagged pictures you'll be able to look back and say "Ah, yes, I was at Coit Tower and here are the images I created while there."
• If you geotag photos as a descriptor of the photo subject then the geotag should be applied to the location of the subject. Again with your example, you could geotag the photo with the location of Alacatraz and then if you upload to Flickr, someone could see your image as they browse photos of Alcatraz.

Your choice depends on how you plan to use the information in the future.

Answer 3

Think of large subjects, especially mountains. You're going to get wildly different viewpoints depending on where you stand (and thus what angle you take your photo at).

If you or someone else wanted to do something from the same perspective (ex: recreate the photo), they'd need to know the photographer's location, not the location of the subject.

Geotagging with the subject's location means that you effectively lose data. There can be a large number (infinite, really) of different locations from which to photograph a single subject. It's probably easier to figure the subject location from the photographer's location than vice versa.

In a perfect world,

1. The metadata would include the direction the camera is pointing (azimuth), vertical (inclination) angle, and even the distance to the subject
2. You'd have two sets of metadata for geotagging: photo location, subject location. (The latter could even be calculated if you have the direction/inclination/distance. Barring that though, you could set it manually.)

Answer 4

http://www.sno.phy.queensu.ca/~phil/exiftool/TagNames/GPS.html shows a list of GPS-Exif-Tags.

Besides [GPSLatitude, GPSLongitude] I find [GPSDestLatitude, GPSDestLongitude] and [GPSDestBearing, GPSDestDistance].

So it is possible to tag both locations - but which software supports this?