Shopping Feeds Management

Re: Google - A proposal for making AJAX-based sites crawlable (source)

After reading through Google's current proposal we have decided to share our thoughts.
Here is what we think how it could be implemented, and demonstrate it on the working example.

Let's have an example (please provide your text and click button to update state):

current state:

the state box above keeps current ajax state which is updated on page load. Of course its updated with JavaScript functions.
The problem here is how to update it when headless browser is asking for it.

Here is how we assume it should be done:

  1. Headless browser is asking for http://feedsmanagement.com/proposals/crawlable.ajax#state=HiIAmHeadlessBrowser
  2. Server is getting full URL with fragmented part #state=HiIAmHeadlessBrowser since only headless browsers can deliver it (i.e. CURL).
  3. Server is parsing fragmented part # (splits it by & and then = which will create $_hGET{HASH GET} array with key=>value pairs parsed from # fragmented data string in url) - PHP hGET script below.
  4. Server uses $_hGET values to recretate requested state and generate static page (which will be working even if standard browser vendors will decide to send full url #state=HiIAmHeadlessBrowser with #fragmented part once a session which could prevent cloacking since differentiation between spider and user would be hard.)

This way is much easier to implement, since it keeps consistent url format across standard and headless browsers.
Search engines are storing, displaying and accessing urls as they appear in the user browser.

To check it in real live, please visit this url with headless browser [CURL] (i.e. http://feedsmanagement.com/proposals/crawlable.ajax#state=HiIAmHeadlessBrowser) with your state to see that state box and input field (above) will be updated according to your state value.

hGET.php installation (it's really simple script and can be easly implemented on any other platform):

hGET.js methods:



contact: @feedsmanagement