Logic behind Amazon Web Services APIs?

Hi! I've been following this list for quite some time, this is my first post.

I was looking at the AWS documentation and noticed that they have HTTP APIs (sometimes refering to them as REST APIs and all that).

Now, the weird part is that it's not even method tunneling through POST payload and/or URI queries, but

a) They're using GET for everything [http://docs.amazonwebservices.com/AmazonSimpleDB/latest/DeveloperGuide/APIUsage.html]
(I found a thread regarding this - "SimpleDB: welcome to 1999" [http://tech.groups.yahoo.com/group/rest-discuss/message/9994] It's from 2007. So, I guess nothing's changed.)

AND
b) Method tunneling through a custom HTTP header [http://docs.amazonwebservices.com/amazondynamodb/latest/developerguide/MakingHTTPRequests.html]
e.g:

POST / HTTP/1.1
host: dynamodb.us-east-1.amazonaws.com
x-amz-date: Mon, 16 Jan 2012 17:49:52 GMT
x-amz-target: DynamoDB_20111205.CreateTable



Is there a reason for this design?
What are the benefits?

Maybe AWS are routing requests according to header value before the whole message is received so that you don't have to parse the message to get the operation method (as when using XML-RPC). Is that even a 'thing'?

But why use HTTP if you're gonna misuse it? 

What do you think?
You know, am I missing something?

 


Thanks!

Dimitar A.