A Farewell to Amazon Light
Seven years ago (!) I launched an app I called Amazon Light, a very early adopter of Amazon.com's brand new "Web Services API". It was designed to be an alternative, lightweight interface to Amazon.com's vast library of stuff. It got a brief, if exciting burst of attention, leading to a decent amount of income from affiliate sales, at least two cease and desist notices (one from Google for looking "too googly", one from Amazon for linking to iTunes & Netflix from a page featuring an Amazon item), and one SXSWi Web Award. On top of all that, it was great fun, I learned a tremendous amount, gave some code and learning back to a community, and picked up a little bit of money.
In the last seven years, Amazon's Web Services group and charter has grown vastly, and the original API has been relegated to just one of many features - even retitled a less-exciting "Product Advertising API" in the Associates arena. I've rolled with many changes over the years, tweaking my code to match changes in the API, but this August, I'm going to let the train roll on without me, and Amazon Light (all four versions) will shut down.
Why? Because, as of August, Amazon will require API developers to authenticate every call, failing unauthenticated calls. Basically, it means I'd have to hide my secret access key (so client-side hacking is out), mangle, sort and hash the request parameters, and make certain no double-escaping happens, etc. on every request. I'm not sure why this is now required but it is, and it's the straw that broke the camel's back.
Now, before you say "WTF? That sounds like a pretty simple thing to do", let me elaborate, 1) all of my code is written in ASP classic (before .Net), a poor language I barely remember, and have no interest in continuing to work with. 2) The amount of earnings I get from Amazon Light has dwindled to less than $100 per year. 3) I really do not have the time (or inclination) to do this these days. Seriously.
I wish Amazon every success with their many programs, I have benefited well over the years, but it's time to let this particular project go. Apologies to all three of you who still use the service, and many thanks to anyone else who ever did use it over the past seven years.
7.16.2009 @ 10:16 AM
View blog reactions
1 Comments +