You may know that Flickr is one of the largest repositories of freely
usable public domain and Creative Commons photos in the world, hosting
collections contributed by libraries, national archives, foundations,
museums, galleries, and individual users (I've uploaded more
than 10,000 CC-BY-SA images of my own). However, with its latest
redesign, Flickr has made is very difficult to copy the images it has
been entrusted with, and nearly impossible to correctly attribute them
in accord with their license terms.
Today, we're fixing that. A little, anyway.
Years ago, Boing Boing reader Cory Dodt (no relation, obviously) created
a script called "attributr" that took the structured license data
in each Flickr image page and created a snippet of text that set out
the permalink for the image, its creator's name, and the license it
was released under, along with a link to the license -- for example:
Late last month, Yahoo updated the Flickr pages for each image in a
way that removed all the structured Creative Commons license data,
breaking the script. And CC users who tried to make use of the images
Flickr is privileged to host found that replicating the attribution
text by hand was nearly impossible. The page used scripts that
intercepted copy-to-clipboard shortcuts and also broke text-selection,
so that it was nearly impossible to copy the name of image or the name
of its creator to your clipboard unless you found it in the page's
In addition to removing the structured, machine-readable CC metadata,
the new Flickr pages also don't make use of the standard CC license
logos, familiar due to hundreds of millions (possible billions, by
now) of webpages that use those logos to signal in familiar terms that
the material on the pages can be shared.
Finally, the links for downloading the high-resolution versions of CC
licensed images that had been entrusted to Flickr have been buried
under a cryptic ellipsis.
Cory Dodt has replicated his Flickr script, with the caveat that it is
now extraordinarily fragile, because it scrapes the Flickr source, and
any future changes will break it. Accordingly, it's hosted on github
so that it can be maintained when (inevitably) it breaks.
But this is bigger than one script. I have a long history with Flickr.
I served as an advisor to Stewart Butterfield and Caterina Fake when
they were working on a game called Game Neverending, and Stewart
offered to prioritize GNE's photo-sharing feature to help me court a
woman I'd started dating overseas. The feature was so successful they
folded up GNE and renamed it Flickr, and now I am married to that
woman. I owe my marriage to Flickr, and Flickr exists in part because
of my marriage.
Over the years, I continued to advise and cheer on Flickr as it became
a powerhouse for CC-licensed images, one of the most important
photographic resources on the planet.
I understand Yahoo's desire to update Flickr in the face of
competitive pressure from other image-sharing services, and I
celebrate its design experiments. But the current iteration -- either
through negligence or deliberate intent -- terribly undermines the
Flickr Commons and does not do justice to the trust and generosity of
its many institutional and individual contributors.
I hope that Yahoo will take speedy action to remediate this issue.
Honestly, restoring the CC metadata -- a feature that's been part of
Flickr since the day it added CC licenses -- is trivial. Using real CC
logos will have virtually no visual impact on the pages. And putting
the download link back where it belongs, visible and located near the
CC license, is a minor change. Ending the interception of control-C on the page
so that users can copy the names of the images and their posters would
And if Yahoo doesn't do this, I think it's time to start thinking
about hosting CC images somewhere else -- perhaps Wikimedia Commons,
or a purpose-built site. Flickr still has its admirable API for
bulk-ingesting CC images and rehosting them in a way that facilitates
the sharing that their copyright holders and custodians expressly
desired for them.