For some time now, Google has been working to improve the indexing of Flash files. They have now released some information about the new algorithm they have developed by integrating Adobe’s Flash Player Technology for this purpose.In an official announcement made last week, Google said that they have improved the indexing of textual content on all types of SWF files, including gadgets like menus, buttons, self-contained Flash Websites and so on.
Google will now be able to index all the text that is present in Flash files. The search engine will also use the text indexed from Flash files to generate snippets, and words in the Flash files will be searchable as query terms by users searching on Google.
Over and above this, Google will also be able to identify URLs embedded within Flash files and crawl these links. However, so far, they have been unable to index non-textual content such as images and videos on FLV files. Even if the Flash files contain only images, it will not be possible for the Google spider to recognise them, and hence they will not be indexed.
Anchor text in Flash buttons will also not get indexed, which poses quite a major problem from a search engine optimisation point of view.
The algorithm developed for this purpose will work by clicking buttons, entering input, then ‘remembering’ all the text that it comes across and finally indexing it. This algorithm has been built in using Adobe’s new Searchable SWF library.
All textual Flash content will automatically be indexed by Google. If webmasters do not wish to get some content indexed, they can just replace the text with an image, thereby ensuring that the file remains invisible to the algorithm, for now.
This indexing facility still has some technical limitations, which Google is working on; For instance:
- If a file is loaded from an external source, such as HTML, XML etc. Google will index it separately but will not be able to attach it as part of the Flash file
- Flash indexing is available in most languages on the net, but so far, Google is unable to index bidirectional languages such as Hebrew and Arabic.