Spell with Flickr
Published by Jeremy Douglass August 3rd, 2005 in search, Text Art.
Spell with Flickr takes a text string and renders it out of photos of individual letters. In order to do this, it draws on the specialized groups within the Flickr photo-sharing community who are creating tagged collection (or “pools”) of letters, numbers, and punctuation - all with an eye towards enabling just this kind of projects.
Kastner, the developer of Spell with Flickr, has provided both the main source code and the js wrapper (spell) source code. This open attitude fits the ethos of his program, whose other free and open component is the human data stream filtered through the Flickr API. The practice of tagging individual letters began before the program and will presumably continue to grow its outputs, even though the code does not change.
Unlike most search art, which generally leverages huge datasets, Spell with Flickr is a custom collection. Any Flickr member contributing photos to the pool (for example, images of W, R, and T ) could reasonably expect to see their contributions in the output within a few iterations.
Like installation art pieces such as George Legrady’s Pockets Full of Memories, the character of the output is shaped by what people choose to upload into the system. Unlike Pockets, here the idea of commemorating commonplace fragments into a database isn’t disruptive, as it is in the gallery space - it is an ongoing, semi-normative practice in the Flickr community which is already its own end, and onto which Spell latches.
Spell with Flickr has a simple API - any text string following the URL is taken as input. (While the program plays with word-images, the URL, metaatem.net, is itself a wordplay - a mirroring of the word ‘meta’)
WRT:
http://www.metaatem.net/words/wrt
The Lady or the Tiger:
http://www.metaatem.net/words/the%20lady%20or%20the%20tiger
Especially nice is the ability to rerandomize individual letters, tweaking the final results.
Do photos of individual characters qualify as digital character art? Part of their interest is that they exist on one boundary between encoded character and and rich image, where one becomes the other.
Flickr Fiction:
Another use of Flickr could be electronic storytelling, along the lines of the Flickr Google-Map annotation. What gets lost in the discussion of Scott’s post is that these automated systems of annotation could be used as more broadly for storytelling?
Flickr Fiction could go far beyond maps.
Why not create fictions by annotating any picture? You’ve got images, with text, participation (Writer-Response), and system for distribution and sharing. Any picture could be annotated. One might ask who is writing the text? What is the relationship between the text or image? How is the image sliced and diced? These are the creative questions.
Here’s a rough draft of a “Lady or the Tiger” benchmark using Flickr annotation.
Lady or the Tiger
I don’t mean to confuse this with Flicktion, which is writing fictions that accompany photos in Flickr. This example from Andrewlos shows how that works. Or perhaps i’m suggesting annotated Flicktion.
Of course, adding messages on hotspots, as others on GTxA point out, is not new to Flickr. But this is an easy-to-use collaborative writing medium using images. Conceivably, collaboration could occur in the comments as well, but the notes on the image seem to change the image in a way that the comments below the picture do not. They seem more like blogging.
I like the idea of using Flickr annotation to tell stories - while embedding text in mouseovers has been used in a lot of different kinds of net art and new media, I wonder if “hotspots” is the right word - I always assumed those were launch-points for some sort of entity (new window, audio or video clip, etc) while the hovering text box that is revealed by pointing is more tenuous and ephemeral, appearing after a short delay and then disappearing as soon as the mouse leaves scope. Is there a formal term? In Ted Nelson-speak, it seems like that would be “hovertext” or “pointtext”…
Do you have a reference link to the GTxA thread you mentioned?