Forum Moderators: open

Message Too Old, No Replies

What do we know about Google Spam Filters?

         

IanTurner

1:01 pm on Jan 29, 2002 (gmt 0)

WebmasterWorld Administrator 10+ Year Member Top Contributors Of The Month



Now we are aware that Google has SPAM filters - what do we know about them?

They appear to fall into two categories manual and automated.

Manual filters are penalties on sites which are reported to Google as spammers (my guess)

Auto filters appear to be part of the update process.

Auto filter penalties appear to be white page rank or greyed out page rank in the case of duplicate domains.

Auto filter penalties appear to be triggered by External CSS files.

Has anyone any other indications of what causes auto filter penalties?

wardbekker

1:05 pm on Jan 29, 2002 (gmt 0)

10+ Year Member



"triggered by External CSS files"

What's wrong with that?

IanTurner

1:12 pm on Jan 29, 2002 (gmt 0)

WebmasterWorld Administrator 10+ Year Member Top Contributors Of The Month



My research indicates that External CSS files may cause a Google SPAM filter and white page rank problem.

Looking at top ranking sites over a number of fields I can find no signs of external CSS if you can give me examples where this is true then let me know.

Damian

1:50 pm on Jan 29, 2002 (gmt 0)

10+ Year Member



Searching for 'mp3', about half the top sites seem to be using external css...

ciml

11:29 pm on Jan 31, 2002 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member



Ian, speaking for myself, what I know about Google's spam filtering could be written on a postage stamp.

Human assigned penalties (often for hidden text/links and/or "doorways") seem to get the grey toolbar and no cache. At least judging by the URLs of some people who've been told by Google that they've been blacklisted.

Some people have had white PR problems due to a glitch, but a large proportion of the people who have white PR that didn't recover in the last update (when the glitch was presumably fixed) have been involved in link building.Google says:

Do not: Participate in link exchanges for the sole purpose of increasing your ranking in search engines.

This time round, the proportion of sites with SEO input seems high. Last time plenty of very innocent people had the duplicate glitch related PR0. Google also says:

Be very careful about allowing an individual consultant or company to 'optimize' your web site. Chances are they will engage in some of our "Don'ts" and end up hurting your site.

I can't help wonder if maybe, just maybe, they're getting quite good at catching us out. The problem is that some of the affected sites just happen to be collections of related, stand-alone content or services.

We all know that serious spammers (multi domain with duplicate content and hidden text and links) are still getting through, but a few people have suggested to me that there's less spam than a couple of months ago.

I've been sent quite a few PR0 URLs, and neither JavaScript nor external CSS are strong themes. If they're involved in this then there must be other elements too.

If anyone has a very clean and simple site with PR0, no doorways and no link-building (but with links from pages in Google) then they may be able to spot what's going on.

I think I might have an affected page. Both it and the only page that it links to use the word guestbook. I see nothing else that differentiates that page from the rest of the site.

Poison words? Poison links? I don't know, but I hope someone finds it soon. I need sleep!

Calum

john316

1:59 am on Feb 1, 2002 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member



<<Auto filter penalties appear to be triggered by External CSS files. <<

I've never seen the googlebot ask for an external css file, I think the more likely culprit is this:

h1{font-size:the smaller the better;}

in the page (not external).

Also calling out the font size inside the <h1> tag.