Hi Bryan,
I understand your frustration. Unfortunately, due to the extremely
competitive nature of web search, we here at Google can't say a lot about
the Google bot or the roadmap for future improvements. Indexing JavaScript
apps is a general problem not particular to GWT. Obviously, this is one of
those problems everyone in the web apps & search community needs to keep
coming back to in order to find better ways to solve it.
Just to give you an idea of the complexity involved, the first "page" of
JavaScript for GWT basically runs a big switch statement that loads a
different script depending on which browser is running (which browser should
the googlebot run. Which bugs should it emulate?). It doesn't actually
create the DOM until after the body of the document is finished loading
(when does it know to start looking at the DOM?). Your app might be
perfectly happy for the bot to index just the front page, but that is still
going to leave a huge swath of unhappy app developers. Another page might
present something on the first page that is not very indicative of the
content, like: "this browser is not supported" or "Login or create an
account" or "choose your region" using images before continuing. A page
might have tabs or a menu with content that doesn't actually get attached to
the DOM until after the tabs are clicked and has a message ("click on the
menu to ...").
Here's some spin for you: I think the message from the search side of
search engines isn't "Don't use JavaScript". Instead, the message is to
provide a page of HTML that faithfully describes your app and/or its content
when the search engine crawls your page. I know its more work, but think
about how that might actually be an opportunity for Web 2.0 authors.
-Eric.
On Sat, Jan 24, 2009 at 2:59 AM, bryanb <webbt...@gmail.com> wrote:
>
> That's the point of my query/question, Why can't the Google bot
> understand Javascript ? As I said originally, using Firebug I can see
> what the Javascript has rendered to the DOM, so there's no good reason
> the Google bot can;t do the same. Granted, it cannot follow links or
> any of the possibly unlimited execution paths in the Javascript, but
> it should be able to render the initial state of the page, and
> consequently index stuff on that page. Likewise if there is a site map
> with history tags, it should be able to render the initial state of
> each of those pages and index accordingly. The initial state is really
> all you want indexed anyway - if I do a Google search for "fubar", I
> reasonably expect the URLs returned to point to a page with "fubar"
> on it somewhere i.e. for a GWT app the initial state of that page.
>
> It just seems a bit strange that one part of Google has created a tool
> for making really usable web sites, but the search side of Google says
> "don''t use Javascript" if you want to be indexed.
> >
>
--
Eric Z. Ayers - GWT Team - Atlanta, GA USA
http://code.google.com/webtoolkit/
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Google Web Toolkit" group.
To post to this group, send email to Google-Web-Toolkit@googlegroups.com
To unsubscribe from this group, send email to
google-web-toolkit+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/Google-Web-Toolkit?hl=en
-~----------~----~----~----~------~----~------~--~---
