What is the correct geolocation?

[Paul Barrett]

Apologies if this is a bit off topic but the audience here is very likely to have some well informed views on this

Q.  What is the correct geolocation for a photo, the position of the camera or the subject?

They may be quite different.  For example if I take a photo of a mountain I may be miles away,

I think it has to be the subject's location but GPS enabled cameras will make it the camera's location.  That means some work is required to correct it.  My photo library apps can handle that work but am I being pedantic about this?  Should I just let it go and bow to the inevitable?

I would appreciate your thoughts on this issue please folks.

Regards

Paul

[ryerman]

What exactly are the GPS co-ordinates of a mountain?
What are the co-ordinates if the subject is a star?, earth's moon?, an airplane in flight?, a building that has a side 1 mile long?

[StarGeek]

I believe that the location of the camera is the correct location.

The exif standard has two sets of gps locations.  See GPS tags, tags ids 1-6 and 13-1a.  Unfortunately, it seems that the EXIF spec doesn't clearly state which tags to be used for what purpose.  But the Metadata Working Group Guidelines For Handling Image Metadata (location section starts pg 45) says that tags 1-6 are for the location created and 13-19 are location shown.

My opinion is that it should be the location created so you can go back to the same spot to see the same scene later, to try and recreate the shot if you would like to.   If you were to put in the location of the subject only, that would be extremely imprecise.  If the subject was a mountain and you put that as the location, then it would be extremely hard to duplicate that shot, because the location could be any number of miles away in any direction.

Additionally, the subject of the photo could be very subjective.  While you may be taking a picture of your friend Bob, someone else might focus Godzilla emerging from the volcano erupting in the distance and yet another persion might think it should be Chuck Norris beating on ninjas 50 feet away on the right.   All three worthy subjects but at very different locations. 

[StarGeek]

What are the co-ordinates if the subject is a star?

Well there is an altitude tag for gps tags. 

Excuse me while I'm off to calculate the exact geolocation of Wolf 359 this morning at 5:36:23Z.  Exactly how many feet are there in a Light Year? 

Heh, sorry, for some reason I found the idea really amusing.

[Paul Barrett]

Stargeek

If I have amused you then my job here is done 

- Paul

PS all opinions respected

[Phil Harvey]

Actually, the new IPTC for XMP extension adds GPS tags for both LocationCreated and LocationShown.  Presumably the former is for the camera and the latter is the subject.  Celestial coordinates are still problematic however.

- Phil

[Alan Clifford]

  Exactly how many feet are there in a Light Year? 

Heh, sorry, for some reason I found the idea really amusing.

Just thought I'd see if google could answer that:

https://lmgtfy.com/?q=how+many+feet+are+there+in+a+Light+Year

[Paul Barrett]

OK, so accepting that the subject is subjective and I have to be conscious of Godzilla in the background, I think that as the owner of the photo I have  the con and can choose MY subject.

But the bigger issue is that we have an emerging standard that differentiates between camera position and subject location (which is  great) but apps that don't use the existing tags consistently.  What could possibly go wrong with the new standard? 

Will vendors use the opportunity to get their houses in order, will they just ignore the new standard or will they just f*** it up some more?

Watch this space (if we can agree the coordinates for it that is). 

Thanks for the insights folks.

- Paul

[José Oliver-Didier]

You may find this amusing/facinating also:

Take a look at this site/service: http://nova.astrometry.net

You can upload a file or send a URL of a photo of the night sky and it will identify the objects (stars, galaxies...) in it.

[José Oliver-Didier]

And of course there is an XMP schema for astronomy photos, hehe.

XMP schema for astronomy https://www.virtualastronomy.org/avm_metadata.php
EXIFTool AVM Extensions http://avm.ipac.caltech.edu/page/6