well - "you're looking at all geo-tagged photos in london" is a bit misleading / ambiguous.  wouldn't it be better to pull out the geo data and then use the semantic data [words] of places etc? that would at least allow for some value on the photos themselves. they could then say with greater confidence where the image 'is'.

having said that it's great how it can take in multiple geo-data 'types' e.g. postcode! 
http://www.flickr.com/map/N8+uk
but how does flickr know that this photo was taken in N8 for e.g.?
http://www.flickr.com/photos/festivalhall/249180992/

sure the kings head pub is in crouch end which is in N8 but this isn't explicitly stated on the page.  is flickr cross referencing crouch end with postcode data?

sorry - i'm dumb. ;-)

JB


On 11/10/06, Andrew Disley < [EMAIL PROTECTED] > wrote:

On 10 Nov 2006, at 16:09, James Boardwell wrote:

is there any pref given to the data?  i.e. geodata / celldata over tags: lodon, greenpark? is it possible to tell?  

From the page itself:

"You're looking at
All geotagged photos in London."




--
voice: 0787 8952479
words: http://www.technogoggles.com
more words: http://stickrattling.com/blogthing
photos: http://www.flickr.com/photos/mashed_potatoe/

Reply via email to