Forum Moderators: Robert Charlton & goodroi
With Google have no real way to compare them, it seems to me there is no way for google to algorytmically decide whether the <noscript> content is in fact a relevant html version of the flash, or really just a heap of hidden text.
One site I am looking at at the moment, has added paragraph of relevant text, to every flash heading in its navigation. No WAY you could do this in html and get away with it.
Is <noscript> the new, google approved dumping ground for hidden text?
There is an even better approach, IMO. Create some html equivalent content for the Flash. Serve THAT content by default, all held in a single wrapper div. Then use javascript to test for Flash support and if support is there, then use the DOM to overwrite the div that holds the html content with the Flash file. There is no need for a noscript section in this way.
There is a temptation to hide content in this way that is not actually in the Flash file. I know of some problems that happened when Google discovered this. how they discoverd it I cannot be sure (seems pretty intense for the algo to uncover), but there is a lot of competitive squealing going on these days.
It might not be the <noscript> causing this. If the flash is really well done and visitors like the site and keep coming back, google might be looking at traffic items such as page views, return visits, bounce rates, etc.... It could be the traffic making the site rank better. If people really like it, google will bump it up in the serps.
Matt Cutts has given several pretty clear indications that Google doesn't trust toolbar or traffic data enough to use it for rankings. It's "noisy" and it's prone to manipulation. My guess is that Google is using such data as a check on other aspects of its algo, but not as a prime algo element.
The suggestion that frequent visitors could cause rankings is in this case also illogical. Even if the visitors caused some kind of popularity factor to kick in, there's a relevance element that would be lacking (as there's no visible text content), and there would be also be no initial search rankings for the popularity algo to boost.
It's much more likely that the <noscript> html text content that nippi cites is indeed what's causing the rankings, most likely that and inbound links. As tedster points out, though, it's not a good idea to misrepresent the content of a Flash page. Competitors are very likely to report you.
If the Flash page does have text content, then it's quite legit to include that text in a default html version of the page.
The problem with much Flash content is that it is not generally conceived around text content... it's primarily visual... and it would be virtually impossible for search engines to interpret it, even if they could somehow "see" it.
[edited by: Robert_Charlton at 7:17 pm (utc) on June 7, 2007]
