39 replies to this topic

[keli]

I had a client ask me the following:

I use a CSS technique that allows me to replace text with images, using only CSS, so the end result is that graphical browsers see an image but anything else sees just text. The SE's will only see the text and will search those accordingly. So visual aid devices like screen readers will read the text as if the images weren't there, and so do SE's.

and then asked if I agreed. He sent me these links to explain this process a bit more:

http://www.alistapar...m/articles/fir/
http://www.kryogenix...de/browser/lir/

Because I had not heard of this technique before I wanted to ask around to see what others thought. From what I've read and understand, this technique shouldn't hurt him nor help him.

What do y'all think?


Thanks,
Keli

[jman]

Hmmm. To take a "Google-esque" perspective, it seems like a good method to provide increased usability for the sight impaired (the technique's apparent derivation) and thus should not be deprecated. However, it's also clearly capable of being used to spam search engines to some degree. This conflict could well go the way of the altruistic, however since there are other techniques (e.g. alt tag and others) to ascribe text info to a picture that are probably far more common, they might also start looking in CSS code for words like "hidden".

It seems like Google (and others) have enough work to do without parsing CSS as well, especially if it's in an external file, however a while ago I removed a "hidden" style from my CSS just in case. (I was using it to notify users that don't have CSS turned on that new browsers were available.)

But the main issue is why this would be helpful (SEO-wise) in the first place. This technique inserts an image as background in the style element of choice, thus you are simply exchanging the opportunity to have the search engines key in on the image filename and alt tag for the text you are replacing it with. Is this a good trade? Nobody but Google knows for sure. I suppose you could argue that a lot of text could be placed there, more than would be scored if put in an image alt tag. Still, seems very little gain for a lot of effort. Plus, what happens if the image doesn't load for some reason? Your regular users are stuck. (Although no differently than if the page were formatted normally.)

[Scottie]

How could you use this to decieve a search engine? I'm missing it. All you are doing is delivering images via stylesheet instead of HTML.

I don't see any problem with it at all. Alt attributes are for describing images, not influencing search engines. AFAIK, they are not even indexed unless they are ascribed to an image that is also a link.

If you don't want Google caching your images in their image search feature, or people right-clicking and saving your images (they can still get them, though) this is a good way to go.

From an SEO standpoint, it does make the delivered page smaller (less code) and many people think there is a benefit in that.

[jman]

How could you use this to decieve a search engine?

Same way you could include hidden text through other methods. E.g., same color as background. Could hide entire paragraphs (even with links) on the page in ways not detected and deprecated by current ranking algorithms. Although, even if it weren't detected, the author would have to be very careful to write it in a style that doesn't look like spam. E.g. "spam my spam your spam everyone's spam spam spam Monty Python spam" would probably get detected.

Note: I am NOT advocating this.

And Scottie, I before E except after C. (Wouldn't bother chiding you, but saw your post about copywriting in another forum.)

[Scottie]

And Scottie, I before E except after C. (Wouldn't bother chiding you, but saw your post about copywriting in another forum.)

I hate that rule.

[Scottie]

Same way you could include hidden text through other methods. E.g., same color as background. Could hide entire paragraphs (even with links) on the page in ways not detected and deprecated by current ranking algorithms. Although, even if it weren't detected, the author would have to be very careful to write it in a style that doesn't look like spam.

FWIW- she wasn't asking how to hide text or use images to obscure other things on the page.

There are plenty of techniques that can be used to try and fool search engines. But it is ok to use tools and techniques to build pages the way they were meant to be used!

[jman]

FWIW- she wasn't asking how to hide text or use images to obscure other things on the page.

From what I've read and understand, this technique shouldn't hurt him nor help him.

I was pointing out that there's a chance that it could be viewed as a spamming technique and thus hurt if it warranted a penalty.

I suspect your grammar will be near-perfect for the rest of the day.

[Scottie]

I suspect your grammar will be near-perfect for the rest of the day.

I don't know about that.

[Ron Carnell]

The basic idea behind FIR is to give our <img> tags the equivalent of a <noscript> or <noframes> alternative. Basically, the image is on TOP of the text, completely obscuring the text if someone has a graphical browser. A text-only browser, however, won't see the image, ergo will see the text.

Just as with <noscript> and <noframes> the FIR technique can be used for both good or evil. To the best of my knowledge, no current search engine (including Google) gives less weight to the usually-hidden text than to any other text on the page. My crystal ball is a little fuzzy about the future, though.

[domokun]

The basic idea behind FIR is to give our <img> tags the equivalent of a <noscript> or <noframes> alternative. Basically, the image is on TOP of the text, completely obscuring the text if someone has a graphical browser

im not sure thats true ron.
my understanding, and i have used the technique before, is to display the graphic, and dont display the text, if the browser is capable of reading style sheets (most modern browsers) and to display the text, and not the graphic, if the browser cannot/does not use style sheets.

its the same end effect however.

i see nothing wrong with this technique, and agree with ron on this point. MOST techniques can be abused and the FIR is no exception. i think the FIR is SEO friendly because i intend to use it correctly.

[qwerty]

domokun, I believe both you and Ron are correct. The method uses CSS to put an image on top of text. If a user can't read the CSS, the text shows up. But if a user can't display images, the effect is the same -- they get the text as well.

[Scottie]

I would just advise people doing this to be careful about the text- make sure it's stuff you actually want people to see!

I had to browse with images off a few weeks ago (working on a pre-paid dial-up service and didn't want to waste time downloading images at 26.6kps) and some of the sites I visited were completely non-functional without their images.

I think most people discount that (who browses with images off?) but you never know who will need to access the site with limitations.

[keli]

What about the ability to follow and read those links that show up as images? I've always liked my clients to have text links, not graphical ones. Will this method interfer with the abitliy for spiders to follow the links?


Thanks,

[Scottie]

Hey Keli-

I'm not sure I understand the question. You want to load image links from the stylesheet?

[keli]

Basically I just want to know if this method, that a client wants to use, is spiderable. I usually suggest text links.


Thanks,