Monthly Archives: November 2007

As you know I'm always looking for ways to help Flash developers make content accessible to search engines. Today I received a link to a site claiming to have the answer to that age old question, "How to SEO Flash"?

The site claims to have a sample of "SEO SWFAddress 2.0" code that "provides a better separation between the content and the presentation." Better than what I'm not sure! Either way, the urls in the "SEO sample" still contain #anchors. Googlebot ignores #anchors in URLs and I'm really hoping "SWFObject 2.0" isn't based around such a myth!

A good example of this is Google's cache of "SEO Sample" portfolio 2.
Here is what the user sees:

As you can see the two pages are different and that is called cloaking!

- Sample from SWFAddress 2.0 Website

As an SEO I'm always looking for ways to help Flash developers make Flash sites more "search engine friendly". I recently came across an article on the Adobe Developer Connection that sounded interesting at first. As I kept reading, I was surprised by what they call a "solution" for the "one URL per page" issue as it relates to sites in Flash. To solve this deep-linking issue, Adobe proposes a method for directly linking to content that's "buried". The technique they suggest uses a variation of RESTful URLs in Flash. REST or "representational state transfer", basically uses one or more distinct URL/s linking directly to different content or different states of content within web-based applications.

The technique uses a "frame anchor" located in the URL to specify one specific frame in the main timeline. As a result, the playhead jumps to that specific frame in Flash and users with Flash enabled see content associated therein by the Flash developer.

"The syntax for writing a URL to point to a particular anchor location in HTML is to use the pound sign (#) followed by the designated name for the anchor, as in the following examples:

* #section1: a URL that points to the anchor named "section1" in the current HTML page

* some_html_page.html#appendix: a URL that opens the URL some_html_page.html and then scrolls to the anchor named "appendix""

from: "Deep-linking to frames in Flash Websites"

This solution may work for "deep linking" in Flash but it's yet another nightmare when it comes to making Flash sites search engine friendly. Bottom line, Googlebot ignores #anchors but browsers do not. So when users link to "any URL dot com" /home.html#/about.html from their blog, PageRank, "link juice" and/or relevancy intended for about.html is instead given to home.html.

I've been researching search engines, search technology and Google for a number of years now. About 6 years ago, I happened upon documents which appear to be the original first drafts of "The Anatomy of a Large-Scale Hypertextual Web Search Engine" by Larry Page and Sergey Brin. As you may or may not know, it's in this paper originally published in 1998 that Larry Page and Sergey Brin propose a prototype search engine called Google.

These documents are interesting, obscure, possibly important from a historical perspective but, either way other folks should know they exist. To date, there is little if any mention of these documents as far as I can find.

Last modified on Tuesday November 25, 1997, is titled "The Anatomy of a Search Engine". It's in the later, "stable" version of this paper that Larry Page and his friend Sergey Brin present Google.

In terms of last modification time, the next draft by Sergey Brin seems to have been posted 7 hours later. was last modified at 12:49:15·GM the same day.

Google: The Anatomy of a Large Scale Web Search Engine

Google: The Anatomy of a Large Scale Web Search Engine

The Anatomy of a Large Scale Web Search Engine

The Anatomy of a Large Scale Web Search Engine

The Anatomy of a Large Scale Web Search Engine