Problem with page indexing

[erebrus]

Hi all,

I think I'm missing some detail trying to get google to index my gwt website.

If I try the address with _escaped_fragment= in the browser (e.g. http://foo/?_escaped_fragment=bar corresponding to http://foo/#!bar) I see my html friendly snapshot. However, when I try fetching http://foo/#!bar in the webmaster tools, I still get the original JS page, not the snapshot, which I'm assuming means that the google bot isn't really understanding that he should get the _escaped_fragment_= version not the #! version.

I have included <meta name="fragment" content="!"> on the original html.

What could I be doing wrong? Any tips on where to look would be very helpful!

Thanks