I´ve got some questions concerning the lately released
Flash Search Engine SDK.
Well, what I know so far is it allows spiders from Google or
Yahoo to read out the text used in .swf files.
But how does it really work? For example:
If you avoid using flash on a website the normal way to get
an acceptable page rank is to use things like title, keywords which
are repeatedly used in continuous text, h1/h2 tags, file names,
link names, ...
But which of those things apply on the .swf files as well?
Are link names recognized as related to their target and by
Do keywords defined in the head apply on the .swf files?
Do the spiders see dynamic generated text inside the .swf
files (text you read out of external files, xml files, ...)
I hope you can give me some useful information on how it
works exactly because I couldn´t find any useful confirmed
information on the internet so far.
The spiders see ALL text on all pages. They actually run a
specialized version of the player, and they go through and catch
any text. They don't catch text that is invisible, off the display
list, or off the stage, and they only catch text that is still text
(so no broken apart text). They do see the dynamically-generated
text, but the specialized player does not allow the loading of
files, so if it is XML-generated, they don't catch it. A caveat
there, the google spider will mark the XML file that you try to
load, and will crawl and cache it separately.
Also, the text does not have size or color information, so
weighting information visually does not work.
Google has not released anything (even to the Adobe devs
working on this) about how the information is getting weighted.