20 replies to this topic

[joker]

Has anyone had a problem with search engine spiders crawling sites using cascading style sheets?

As long as the code is sound, I don't think I would expect a problem, but I would like to know for sure that spiders see the <h1> and <h1 class="header"> or <h1 style="...."> as the same thing.

[K.S. Katz]

I personally use external CSS and never had any problems with the spiders crawling. In my log files the Googlebot and other spiders pull the .css file all the time and it doesn't seem to cause any issues. I recommend using them because it reduces the file size and the pages load faster.

[Jill]

No problems with CSS.

Jill

[SearchRank]

I don't think they have any problem with them at all and would think that they would prefer them as K.S. noted, it does allow for much cleaner coder - no font attributes on every paragraph. I do think it is wise to place CSS in a text file and then call for it in the html, especially if you have a huge old mess of CSS code. This also makes it easier to make changes - modify one file and the change is implemented throughout the site.

[Vertster]

On a similar note...

I remember there was a mention (I think by GG) that as far as Google is concerned, you should not exclude your CSS file in your robots.txt. I would presume that this is because they *want* to crawl it. Probably to run it through some kind of algorythm that determines if you are hiding text...

[Jill]

Personally, I think it's none of G's business if you exclude your CSS file. I was shocked to hear them recommend that. (They did, however, at one of the SES conferences.)

[Randy]

I doubt G wants to tell you not to exclude your css file Jill, but they have either be able to see it or make a decision whether to exclude the listings for any site which keeps them from seeing it.

The issue is that with CSS is that it's powerful enough that you could actively hide entire layers (cloaking) or set your H1 tags to be 0.1em, which would effectively be invisible (deception).

As to the main question... All of my sites use an external CSS file and none have had any SE issues.

The fun of using an external css file is that when you find someone stealing your stuff, but forgetting to change the hardcoded, full URL css reference it's laughable simple to really trash their site. You'd be amazed at how many people have no clue what that "link rel" line means or does.

ps Yes, I've been known to do that...even have a stock background image ready and waiting. It's a really simple thing to drop that into their background, hide their text layers and put in others, etc. It's an easy way let everybody know that the site is stolen and where to go for the real thing.

[eatapeach]

that sounds like an excellent plan, randy.

so far I only have used css imbedded in the pages, so the external use is still new to me. but the way you describe changing the stylesheet on stolen pages raised a question or two.

if you change the stylesheet for one page doesn't that affect all the pages that use that stylesheet?

if so, how do you change the link rel info on the other pages that include that stylesheet to include a different one?

i've already found one page that was stolen from one of my sites but i couldn't do much about it. it will be nice to have a way to solve that problem should it occur in the future.

take it easy,
michael

[K.S. Katz]

Create multiple .css files and give them different names. (i.e. main.css, sub.css, etc)

[Randy]

Hi Michael,

Yes, changing something in your CSS will change every page which links to it.

When I catch someone using my layout and still linking to the css file on my server, I'll just rename my css file and then a search and replace on all of my site files. Upload all of that and I'm ready to start having some fun. Make sure you leave the old css file on the server though. Don't wanna tip them off too soon.

Once your site is using the new css file you can safely tinker with the old css file to your hearts content since the only thing it'll affect is the offending site.

There are all kinds of neat things you can do, depending upon how much fun you want to have. Change it just a little to mess up their site. Move everything around on the page. Absolutely position stuff 10,000 pixels from the top of the page. Make every layer invisble so that the page appears to be blank.

If you really want to get radical and have sufficient privleges on your server, you can even configure Apache to process .css files through the server's php module before delivering them to the browser. In which case you can pop a line or two of code in your old css file and redirect all of the traffic from the offender to your site or any other site to which you'd like to send the traffic. That's a bit much, but I do have to admit doing it once and redirecting all of their traffic over to their 404 Error Page just for giggles.

<Edit>No I'm not really evil...just have a weird sense of humor.</Edit>

[qwerty]

I like this idea, Randy. The closest I've come to doing that is renaming an image somebody was hotlinking and sticking in a series of EBay ads and replacing it with this little message.

[air-dog]

Hi All,

>if you change the stylesheet for one page doesn't that affect all the pages that use that stylesheet?

eatapeach, instead of using a .css file - create a external style sheet in perl, php etc.(server permitting) then you can query the external stylesheet from each individual page using the normal <LINK REL tag

You could even programme in a back up plan, such as the one Randy described, using the referrer functions

[torka]

replacing it with this little message.

Hey, qwerty, mind if I stea... err, "borrow" that image?

--Torka

[qwerty]

Feel free, Torka. I'm halfway hoping for an opportunity to use Randy's CSS trick, myself

[Clintorius]

On a similar note...

I remember there was a mention (I think by GG) that as far as Google is concerned, you should not exclude your CSS file in your robots.txt.  I would presume that this is because they *want* to crawl it.  Probably to run it through some kind of algorythm that determines if you are hiding text...

If G wanted to they could read your .css files at any time. There is no law against ignoring files listed in robot.txt files. Instructing the robot to follow a link in HTML with .css extension no matter the content of robot.txt is no big deal.

So do not worry about it - unless you are about to hide anything