Das eez kaput! Sometime around 2002 I spaced the entire database table that mapped individual entries to categories. Such is life. What follows is a random sampling of entries that were associated with the category. Over time, the entries will be updated and then it will be even more confusing. Wander around, though, it's still a fun way to find stuff.
That is : a bare-bones daemon that sits around and sets your desktop background every day, pulling down an image from the growing legions of photographers who are posting their work on-pline. It needs just enough of a GUI to enter the relevant data for a payment service (probably PayPal), the URI for one or more photobloggers and frequency that you want to grab a new image from a given photographer.
There are some technical impediments in the payment widget that would need to be overcome. Mostly, I think, it's just idiot-proofing things on the server-side for photographers who don't understand, or care, about the security issues. But it's not like this part hasn't already been done.
And the client has, sort of, already been written thanks to the fact that Morbus did all the hard work and then set it free. That, to me, has always been the beauty of Amphetadesk. Try as I might I can't really get excited about aggregators but the thing about Amphetadesk is that it demonstrates the basic framework a tool that sits around in the background, periodically fetches stuff off the Network and then does stuff with it. And it
out of the box. Across platforms. On OS 9, no less. With a simple
for plugging in user configs.
Yes, it can be a bit slow and if I were to write something from the ground up I might write it in wxPython. But Amphetadesk is there and in my magic lala-land it just seems sort of rude not to take advantage of everything it has to offer.
It is also possible that the economics of this kind of service remain prohibitive. I haven't actually bothered to run any numbers.
Off the top of my head, though, it is clear that any given image would need to be priced according to the amount it costs the photographer, in bandwidth, to send to a subscriber. That shouldn't cost too much but it does need to keep pace with whatever a hosting provider decides to do when they discover that they are serving a 1900x1200 pixel image four hundred million times a day. (Note to self: add code/configs to the client preventing charges in excess of n .)
The other immediate problem is how to control licensing of the images which, given their size, may be repurposed for a variety of purposes without a photographer's consent. I don't have a ready answer for this. It is easy to say :
Sue the bastards!
but the having the ability to take legal action and having the means to take legal action are two different things. On the other hand, it seems that given proper attribution the benefits, in terms of exposure, of having a third party re-use a photographer's image might offset so-called proper renumeration. If the third-party happens to be an art director at, say, Time-Warner, well then there's always stenography and a 10 percent commission for an eager lawyer.
There are a bunch of other details, for sure, but I think it could actually work. It could earn photographers (painters, print-makers, yadda yadda yadda) a little extra money, a lot more exposure and just generally offer a little more serendipity in people's lives which has always seemed like the real promise of the Network to me.
So there's the idea. Please, feel free to run with it.