## EXIF & IPTC, photograph metadata @ Adrian Tritschler · Friday, Jul 7, 2006 · 2 minute read · Update at Jul 7, 2006 ·

I’ve been importing my photos into Adobe Photoshop Album over the last few years, entering titles and tagging the images. The titles go into the images in the EXIF header, but the tags and other information is held in Adobe’s proprietary database. I can extract the EXIF:ImageDescription with python or perl, and some of the other image viewers will display it … some, but not all. My latest experiments have been with Google’s picasa, or more specifically, the beta version from http://picasaweb.google.com/, which allows geocoding and has various tie-ins with GoogleEarth. Unfortunately it seems that Picasa uses the IPTC:Caption-Abstract as the source of its title, so all the information I’ve entered via Photoshop Album is ignored.

A few quick searches and then, Phil Harvey’s came exiftool to the rescue! Read and write every single possible type of metadata, at least every possible type I’m interested in at the moment.

exiftool -IPTC:Country-PrimaryLocationCode=AUS *JPG
exiftool -IPTC:Country-PrimaryLocationName=Australia *JPG


I’m not really sure what to do with the postcodes (zipcodes), I think I’ll put them into IPTC:Sub-Location, at least until someone sends me a nasty-gram telling me the correct field to use.

PSAtools was next (the author no longer maintains it, but I found a copy archived elsewhere. Dump out the Photoshop Album catalogue into CSV and XML text files so I can play with it to my heart’s content.

My website, an agglomerative mess, probably half-eaten by a gru

# …The Owner

There’s not much more I can add to who I am.

# …The Site

Vanity site? Technology experiment? Learning tool? Blog? Journal? Diary? Photo album? I could tell you, but then I’d have to kill you…

I experiment. I play. I write and I take pictures. Some of the site is organised around topics, other parts are organized by date, then there’s always the cross-references between them.

Its all been here a fairly long time. Like the papers on my desk, or the books on the bedside table, the pile just grew… and it all grew without much plan or structure. I try not to break URLs, so historical oddities abound.

Long ago it started as a learning experiment with a few static HTML pages, then I added a bit of server-side includes and some very ugly PHP. A hand-built journal/blog on top of that PHP, then a few experiments in moving to various static publishing systems. I’ve never wanted a database-based blogging engine, so over the years I’ve tried PHP, nanoblogger, emacs-muse, silkpage and docbook before settling on Emacs Org mode for writing and jekyll for publishing. But the itch remained… I never really liked jekyll and the ruby underneath always seemed so much black magic. So now the latest incarnation is Org mode and hugo.

# …The ISP

• Hosted by @cos