On pp. 478-479 of Chapter 16 of my mashup book on online storage APIs, I show how to how to reproduce the calculation of an API signature in Amazon S3 in Python and PHP. Recently, because I now want to access S3 from within the browser, I have figured out how to do the same example using JavaScript. Specifically, the following is a Chickenfoot script that generates the correct signature.   (Note that the code depends on Paul Johnson's sha1.js library. (See his instructions on how to use the library. I had to consult the page to learn about the b64pad variable).):

// reproduce results on pp. 478-479 of
// Pro Web 2.0 Mashups

// http://docs.amazonwebservices.com/AmazonS3/2006-03-01/RESTAuthentication.html
// http://pajhome.org.uk/crypt/md5/instructions.html

include("D:\\Document\\JavaScriptLib\\sha1.js");

var AWSAccessKeyId = "0PN5J17HBGZHT7JJ3X82";
var AWSSecretAccessKey = "uV3F3YluFJax1cknvbcGwgjvx4QpvB+leU8dUj2o";
var Expires = '1175139620';
var HTTPVerb = "GET"
var ContentMD5 = ""
var ContentType = ""
var CanonicalizedAmzHeaders = ""
var CanonicalizedResource = "/johnsmith/photos/puppy.jpg"
var string_to_sign = HTTPVerb + "\n" + ContentMD5 + "\n" + ContentType + "\n" +
  Expires + "\n" + CanonicalizedAmzHeaders + CanonicalizedResource
output(string_to_sign);

b64pad = "="; // needed for "strict RFC compliance"
var sig = b64_hmac_sha1(AWSSecretAccessKey , string_to_sign);
output ("|"+sig+"|");

I have uploaded two more first drafts:

• Chapter 15: "Online calendars and Event Aggregators"
• Chapter 16: "Online storage"

They are a bit rough but I'd definitely value feedback from any valiant readers out there!

I've been captivated by the potential of Amazon S3 (Amazon's "Simple Storage Service"), which is described in the following way:

Amazon S3 provides a simple web services interface that can be used to store and retrieve any amount of data, at any time, from anywhere on the web. It gives any developer access to the same highly scalable, reliable, fast, inexpensive data storage infrastructure that Amazon uses to run its own global network of web sites.

With S3 (combined with Amazon EC2 -- the elastic computing cluster), I keep thinking that I have the raw ingredients for a relatively inexpensive supercomputer. Now what to do with that computing power -- that's the subject of another post.

At a basic level, Chapter 16 is meant to be a tutorial of Amazon S3 and rival/parallel/comparable services. What are other services that I'd like to cover (if I manage to have enough time to write up)? On the list of possible services to cover are:

• Box.net with Box Enabled - Box.net Web Services API
• Streamload with its API
• Omnidrive and the OmniDrive API
• openomy and its API

First place to start: learn to program Amazon S3. One thing I'm reading is Amazon Web Services Developer Connection : Monster Muck Mashup - Mass Video Conversion Using AWS, which touches on not only S3 but also Amazon EC2 and the Simple Queue Service.

In my post Mashup Guide :: Amazon URL structures, I mentioned that I will pose my questions concerning the Amazon web services in the appropriate forums. I'm looking at the list of forums Amazon Web Services Developer Connection : Developer Connection but I'm uncertain whether any of them is appropriate yet. Along the way I did find the Amazon EC2 Firefox Extension. I plan to come back to looking at that plugin, in combination with S3 Firefox Organizer(S3Fox), as ways to jumpstart one's exploration of Amazon S3 and EC2.