Hey John, >From what I can tell, you need to be careful with that approach (although it's pretty darn cool). Google get's very nervous when it finds out that the googlebot is going one place but users are going another since this is something that evil, spam-y websites do. It's tough to get a definitive answer, but my decision from reading the google bot forums was that it was too dangerous to try this approach.
I'm totally biased of course, but you might check out the interview I did with InfoQ that includes a free chapter of my book. The chapter just happens to be the one about SEO with GWT, so you might find it useful. It's available at http://www.infoq.com/articles/progwt My solution was to serialize data into the page itself instead of RPC and then output in <noscript> tags. The process is described in the chapter available above. gl, -Jeff On Sep 28, 11:41 pm, John Fowler <[EMAIL PROTECTED]> wrote: > Please advise concerning my approach for enabling googlebot to index > my GWT site:http://www.SudokuComplete.com/ > > First, I needed to create a static, flattened, html-only view of my > website. To do this, I created a simple .Net program that uses the IE > WebBrowser control to navigate to my GWT site. The program takes a > URL with a history token, let's the browser run through the GWT > javascript, and then saves a copy of the resulting DOM html to a file > with a name based on the history token. This file is essentially an > html-only view of the site as-of the history token. I use the program > to save static versions of each of the major history tokens on the > site. These files are the ones I want googlebot to index. And to > provide navigation for googlebot, I add links between all the files at > the end of each of them. > > Next, I created my "UserBotRouter", a .Net HttpModule (analagous to a > J2EE Web Filter) that analizes the incoming requests to the website. > It checks the UserAgent header to see if the request is being made by > a bot or a standard browser. If it is a bot, then it routes the > request to the appropriate static html page (created in step #1). If, > however, a standard (non-bot) browser requests one of the static > pages, then my module sends an HTTP redirect to send the user's > browser to the corresponding GWT page including the respective history > token. In this manner, I am able to route users to the GWT pages, and > bots to the static HTML pages. > > Am I missing anything? Does this sound like a workable approach? > > Can GWT build something like this into their compiler? That is, if > GWT compiles different versions for the different browsers, why not > create a set of standard "bot" pages. In the module XML file the > developer could specify the tokens for which GWT should create html > pages. GWT would use an approach similar to mine above to create > static html files for the tokens. Then, in these static html files, > some javascript could redirect the browser to the corresponding GWT > url. This javascript redirect would affect actual user's browsers, > whereas bots would continue to read the page as-is, following links to > the other static pages. > > Please let me know what you think, > > john... --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Google Web Toolkit" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/Google-Web-Toolkit?hl=en -~----------~----~----~----~------~----~------~--~---
