Forum Moderators: Robert Charlton & goodroi

Message Too Old, No Replies

Google & CSS Image Replacement

         

Komodo_Tale

9:43 pm on Mar 15, 2006 (gmt 0)

10+ Year Member



I'd like to discuss the use of CSS Fahrner Image Replacement (FIR), Leahy/Langridge Image Replacement (LIR), Rundle’s Text-Indent Method, and the Cover-up Method and how they specifically apply to Google rankings.

If you visit Matt Cutts site you will se an article titled SEO Advice: check your own site. In this article a website is demonstratively using CSS Image Replacement to cover blatant keyword spam.

On the other extreme, if you visit the CSS Zen Garden you will see some stunning examples of using Image Replacement for legitimate style purposes.

My question is where can one ethically draw the line or is there a line?

For example, if I replace <H1>Fall Color Report</H1> with a top of the page image that stylishly declares Fall Color Report, my own personal logic tells me that I am giving Google an honest representation of what the human user is seeing. Is my perception correct?

Does Google recognize legitimate uses for CSS Image replacement?

GoogleGuy: if you are willing, your input would be most welcome.

[edited by: tedster at 9:52 pm (utc) on Mar. 15, 2006]

tedster

10:08 pm on Mar 15, 2006 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member



Last year at the New Orleans PubCon I brought up this topic with two different Google engineers at two different times. At that time, it was not a topic that was apparently on the radar for G.

One engineer (who was not initially familiar with Fahner Image Replacement) was adamant that even if the image and the text were an exact match, this would be still considerd a breach of the Google guidelines. The second (who was familiar with FIR and its legitimate use in making content accessible) felt that, as long as there was an exact match all would be well.

It seemed to me at the time that this would be/should be a judgement call, based on a hand check -- and I am assuming that it isn't currently an algorithmic decision since it would require widespread spidering of external css files. The ability to use OCR to see the image text could certainly be automated.

However, as to an official Google position on the topic of legitimate uses of FIR and its relatives, I haven't heard one. SiFR -- the Flash replacement -- also comes to mind. ABCNews was using that and not suffering -- not that this is any guarantee, of course.

smokeybarnable

11:55 pm on Mar 15, 2006 (gmt 0)

10+ Year Member



yeah i used to use css to mimic what my image said but got tired of getting penalised so now I must settle for an ugly page. I know of no way around this after trying different methods for six years...I finally gave up.

tedster

12:08 am on Mar 16, 2006 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member



Smokey, did you get an official word from Google that you were penalized?

smokeybarnable

5:18 pm on Mar 16, 2006 (gmt 0)

10+ Year Member



no i didn't but it was obvious.

smokeybarnable

5:25 pm on Mar 16, 2006 (gmt 0)

10+ Year Member



thats a good idea about ocr...I was wondering if there was any possible way for spiders to understand images with text.