Jagger3 Busting CSS Spam with Collateral Damage?

[greeneagle]

Matt Cutts issued an unusually stern warning about using CSS Spam techniques, coinciding with the Jagger Update (strangely enough) Oct 19, 2005:
http://www.mattcutts.com/blog/seo-mi...wise-comments/

Here is link a to the article in Threadwatch entitled; “Google Engineer Hammered over CSS Spam Comments”:
http://www.threadwatch.org/node/4313

I am seeing Jagger3 update busting CSS Spamers:

Look what happened on the way to the “wood pile”!

I believe many are seeing collateral damage.

Ken

[nashville]

Have you noticed and traffic or rankings change from a couple weeks ago? Better? Worse? Same?

[greeneagle]

Hello nashville,

Jagger Traffic and PR changes throughout the Jagger Update are being discussed in multiple Jagger threads in this Forum.

I would like to keep this one on topic:
Google's Renewed Interest in CSS Spam and whether or not that is a big play in the Jagger update and in collateral casualties.

Thanks,
Ken

[MarcGrobman]

Thanks for bringing this up Greeneagle. I'm curious to see where it leads. Have you seen any other examples with Jagger3?

Sygon seems like a simple case of banning. That wasn't really even css spam as much as good old fashioned hidden text. And lots of it. These days a lot of sites just leave that stuff visible and, though ugly, it just seems grey hat. Why do you think this is a new CSS thing? Because styles are used so it might be harder for search engine to detect if the text matches the background programatically? That text wasn't hidden, it is right there on the screen.

[greeneagle]

MarcGrobman,

http://www. sygon. com/styles.css
BODY Background color = #003366
hidden links = class “key” with #003366 colored text and hover, matching the background

They needed the links to all their hidden pages in the site for the SE's but didn't want the visitors to see them so they CSS'd out of view.

I think it was interesting that Matt came out so strong addressing these kind of issues with almost perfect Jagger timing.

Ken

[MarcGrobman]

Yea I see the matching text and background. I wonder if it is algo or just banning. It is a real obvious case of hidden text. I'm no expert, but my understanding is that there are more sneaky, less detectable ways to use css. I wonder how these are doing right now....

[greeneagle]

Marc

They were banned, delisted and lost all PR after Jagger3 started. They were still holding stead at PR5 and high SERP through Jagger2. In fact they had even increased SERP for some keyphrases after Jagger2.

Even if this "pans out" as a component... It's just one nugget in a complex rollout.

Ken

[Faglork]

Originally Posted by greeneagle

I am seeing Jagger3 update busting CSS Spamers:

Yeah, if you put Goog's nose right onto them. Otherwise, not.

As I said in another post: If Google were to penalize all those "minor issues", then many good no-spam sites would fall out, and the real spammers who know their CSS would stay in. This is not in Google's interest (as I believe).

We have seen that Google is - in many cases - not able to detect even the most blatant spam. Do you think Google catches the small fish ("there is a 4 percent chance that this might be spam ... lets kick it out") because it can't get the big ones?

I don't buy that.

Cheers,
faglork

[greeneagle]

Matt was quite specific when hinting that they may have been working on different CSS Spam detection algs:

“If you If you’re straight-out using CSS to hide text, don’t be surprised if that is called spam. I’m not saying that mouseovers or DHTML text or have-a-logo-but-also-have-text is spam; I answered that last one at a conference when I said “imagine how it would look to a visitor, a competitor, or someone checking out a spam report. If you show your company’s name and it’s Expo Markers instead of an Expo Markers logo, you should be fine. If the text you decide to show is ‘Expo Markers cheap online discount buy online Expo Markers sale …’ then I would be more cautious, because that can look bad.”

Now that I look back, my Jagger casualty Site has several points that may be collateral damage in a Google CSS Spam filtering scenario. There is nothing hidden anywhere, but let’s take for example: same sized font in consecutive header tags.

Ken

[greeneagle]

Faglork:

"Yeah, if you put Goog's nose right onto them. Otherwise, not."

That Site was discussed at WPW at least a year ago. In that time, I don't know if anyone Spam reported them.

I certainly didn't.

I can't imagine GOOGLE rushing to bust that Site discussed at WPW, when they haven't busted some of the others.

All I know for sure is it was coincidental with timing between Jagger 2 and 3.

Ken

[Faglork]

Originally Posted by greeneagle

There is nothing hidden anywhere, but let’s take for example: same sized font in consecutive header tags.

Perfect example. Penalizing *that* would be just ridiculous. You can *optically* indicate levels by color, font, indent, background, you-name-it-we-use-it. This should be up to the Designer.

If Google would take

h1,h2,h3 {font-size: 1em}
h1 {font-family: courier}
h2 {font-family: verdana}
h3 {font-family: tahoma}

as indication for spam, they would be completely nuts.

Your Matt Cutts quote gives a hint what the Goog is up to. They apparently know that you can't simply judge by the use of certain CSS elements, there have to be other factors present before something "qualifies" as spam.

A sort of trigger operating at various levels would IMO be appropriate. Levels, because you need to have the chance to get out with a "zero" rating: A simple addition of "spam indicators" won't do justice.

But I think they know that.

faglork

[greeneagle]

There is an obvious question that should have come up here by now, but just hasn't for some reason.

Has anyone seen one the Googlebots access their CSS file in their server logs in the last month or so?

If the answer is no, one would think it might offer some disclaim here. But the more I thought about that... I am not sure GOOGLE needs to, to deal with this issue.

On the other hand, if the answer is "yes", that might be a fairly diffinitive indicator also.

Ken

[fuzzygay]

In reviewing our logs I found that googlebot spidered our .CSS file on 10/18/05, coincidental? No spidering of this file in the last 2 months. But this just reinforces Matts quotes, no surprise, they are reviewing style sheets.

Our CSS file doe not have any spam techniques, however we were hit hard by Jagger update. We are "dancing" around and I think the update is not over yet. Too many fluctuations are the main indication.

greeneagle, I agree on the hidden text part, but you went too far with the font size in titles. All of us agree on what a spam practice is. The only changes I see here is the way google is detecting it. How do you detect a hidden text where you have as a background an image the same color as the text itself? Well, they might be smart enough to go to the image content itself and detect its most predominant color, an hueristic approach tough.

[ADAM Web Design]

Originally Posted by faglork

Yeah, if you put Goog's nose right onto them. Otherwise, not.

As I said in another post: If Google were to penalize all those "minor issues", then many good no-spam sites would fall out, and the real spammers who know their CSS would stay in. This is not in Google's interest (as I believe).

We have seen that Google is - in many cases - not able to detect even the most blatant spam. Do you think Google catches the small fish ("there is a 4 percent chance that this might be spam ... lets kick it out") because it can't get the big ones?

I don't buy that.

Cheers,
faglork

Actually, faglork might be right, Ken.

In this case, I posted your "web development pricing" scenario in Matt Cutts' blog on Saturday, two days before you posted that it was removed. I don't know if that's coincidence or not, but it may have something to do with yours truly.

Mind you, I don't know, but it's a possibility.

[greeneagle]

LOL - I missed that Adam.

So you guys just reckon that the GOOG couldn't work out anything for CSS spam except in case of hidden links losing relevancy since they will never be selected?

And, Matt's Blog uncharacteristicly stern warning timed with Jagger was just some huff and puff until they figure out what to do about the rest of the CSS Spam Issues, hoping to scare abusers off in the meantime?

LOL - This speculation is getting amusing now!

Ken

[MarcGrobman]

Thanks Faglork for your thoughtful reasoned comments here and elsewhere. Faglork posted a very helpful Matt Cutts quote at http://www.webproworld.com/viewtopic.php?t=55705 .

My understanding is that they manually ban people based on flags by the spiders or from complaints.

At the same time, the example you show Greeneagle is such a lame obvious attempt at spam. It would not be hard for Google to see that the text is the same color as the background. Was the same-color-text/background penalty ever hard coded or has the threat of manual banning always been the deterrent? You'd have to show me more examples before I'd attribute it to Jagger.

The real spam pros are using techniques much more complex and impossible to detect. As futile as it seems, I still think manual penalties must be at work.

[mktplace]

my "now removed" site is www.owenwalcher.com

It used to be a PR6, then went to PR7 after Jagger2, and now has fallen out of Google altogether.

Let's see, I have defined:
table.menutree {display: none; }
td.menutree {display: none; }

in my printer.css so the side menus don't print, and in my regular style.css, I do have:
H4 { margin: 0; background: #336699; font-family: Verdana,Arial,sans-serif; font-size: 100%; color: white; }

so that my H4 tags are a white text with a dark blue background - of course, the:
body { margin-top: 0in; margin-left: 0in; margin-right: 0in; background: WHITE; font-size: 80%;}

is also white background, so one might think all my H4 text is white on white (hidden), which it isn't...

...unless someone can see what else I might be doing "wrong" according to the "rules".

[greeneagle]

mktplace,

I have reviewed my casualty also and may be experiencing similar collateral damage, just in font color issues.

That's the first thing I will chang, in test.

Ken

[tshesse]

It used to be a PR6, then went to PR7 after Jagger2, and now has fallen out of Google altogether.

Let's see, I have defined:
table.menutree {display: none; }
td.menutree {display: none; }

javascript it instead

sinze you have fallen out completely I'd ask google about it. just don't expect an answer [/quote]

[wrockca]

Thanks for the great information...

It seems that the biggest concentration of the Jagger update was Google filtering CSS spam, from what I understand from this post.

Is there anything else or other post that we can reference a good source on how to continue our optimization for future days.

[johnca]

greeneagle,

http://www. sygon. com/styles.css
BODY Background color = #003366
hidden links = class “key” with #003366 colored text and hover, matching the background

They have also used:

<td valign="top" width="1" class="gray"></td> = 19 x

<td valign="top"></td> = 9 x

28 x web design & web development is just as spammy as the hidden text. I think this was also taken into account as I saw a PR6 fitness site get banned for alt. keyword stuffing two months ago.

john

[greeneagle]

wrockca:

"It seems that the biggest concentration of the Jagger update was Google filtering CSS spam, from what I understand from this post. "

Right now, that is pure speculation, and even if it "pans-out" to be true, it is only one nugget in a complex alg shift rollout.

See the article in my signature. But remember it only speculates on part of what may have occurred and why.

Ken

[bj]

I read over Matt's words in both his blog comments and in the Threadwatch thread very carefully. He states very clearly that if you are hiding text and it's the exact text in the header image via an image replacement technique, and the text isn't spammy, then you would most likely be okay. I use this technique on a few sites. They've all gone UP in PR and ranking. None of these sites use this technique for spammy reasons. I would suggest that we don't all need to redesign all our sites to get rid of skiplinks and headertext/image replacement. That's an overreaction.

However, if you do hide a lot of text, even for a good reason, you're getting into a gray area. If your site divebombed during Jagger, then yes, you should be looking at that.

My personal feeling about this is that if you do hide text for a good reason (Bob Easton's use of hidden text as oral keyboard help for blind surfers comes to mind) and your site takes a nosedive, you need to leave the site just the way it is and put in for a site review. Google needs to be made aware when their algorithms screw up or they'll never get adjusted to trash only the ones who deserve it. I also recognize that financial considerations sometimes make that tough.

I sometimes wonder about this term "Google Dance" and exactly who it is who's dancing-- Google or us?

[ogito]

Thanks for the thread, but what if I use that type of CSS:

/css/style.php?main.css

for example by www. ebayer. com ? Did you tried to load their CSS? I couldn't...

[Keimos]

Hi all,
see above.

robots.txt
disallow/ mmmm.css example but not right so dont be picky.

Linked CSS can be blocked from the spiders so most of the above is irrelevant as long as you lin your CSS and block the robots/spiders from looking at it.

Keith

Some of the above CSS hides are crap anyway. Learn CSS

[greeneagle]

Keimos,

Is there really any reason to make your CSS file unavailable, other than suspicious issues from one side or the other?

Copying a CSS file isn't really a big deal anyway is it? They all have to be customized anyway, unless you are copying the entire site, and then what? In fact that openess is a great source of learning for many!

I haven't seen a single site that dissalows their CSS files yet.

You don't think that that would raise an alg flag on it's own, if they decided to utilize the CSS file/s on reference and had to use a BOT to get there?

Let's get "sinister" - LOL:

Sounds like a real "cat and mouse" game anyway. Do they even need to use a bot to siphon off a site's CSS file? It has to be truly accessible to the "general browsing public" anyway. GOOGLE already has technology that determines if Sites are feeding the SEs something different than the public. That's been around a long time. Suck it up and slap it down... it's just that easy.

Maybe it would also be good SEO and serve some purpose to disallow the robots.txt file as a last command in the robots.txt file?, or dissalow "google mediapartners" on a 10,000 page site with Adsense, post Jagger3?

Let's do it... Put down your pint of Tennets Lager and I'll put down my glass of whiskey!... Let's go talk to Matt!

Ken: "Matt, I understand from your Blog that Google may be looking at CSS Spam as a serious issue. Do you care to comment?

Matt: "I don’t recommend that people use CSS to hide text..."

I just can't emphasize enough; That, that alone coming from that an authoritive GOOGLE voice should be enough to send every CSS Spammer scurrying to make changes....

The problem with the spammer mentality is though, they are defiant and continue to stand their ground thinking they are outsmarting.... until they eat dirt.

It's pretty rare that the GOOG gives that poignant a hint coincidental with a MAJOR alg adjustment! That being said... every SEO company should advise their clients to continue grevious behavior or conduct it for them in the face of probably the most stern warning from a leading GOOGLE Engineer/Public Spokes Person ever issued?

The whole scenario is really quite amusing and a great spectator sport!

LOL

Ken

[espectations]

Hiding things is just something I do not do - too risky - whether it is text or anything else.

I make very sure that all links have distinctive color differences - KISS (keep it simple) is still the best philosophy.

Why tempt fate if you know it is potentially something that could harm your own sites and those of your competitors?

I think Jagger was in any case a combination of several factors, not just one or the other...

I understand that you are discussing specifically the CSS but is there not a possibility that it could be CSS hidden whatever in combination with say non-relevant links???

[Faglork]

Originally Posted by bj

I read over Matt's words in both his blog comments and in the Threadwatch thread very carefully. He states very clearly that if you are hiding text and it's the exact text in the header image via an image replacement technique, and the text isn't spammy, then you would most likely be okay. I use this technique on a few sites. They've all gone UP in PR and ranking. None of these sites use this technique for spammy reasons. I would suggest that we don't all need to redesign all our sites to get rid of skiplinks and headertext/image replacement. That's an overreaction.

Exactly. It would be absurd to penalize technique - they would have to penalize <font color= > as well ...

Google knows that. As Matt Cutts said in the same thread: "We try hard to avoid throwing babies out with bathwater."

Originally Posted by bj

I sometimes wonder about this term "Google Dance" and exactly who it is who's dancing-- Google or us?

LOL - that's a good one.

faglork

[greeneagle]

espectations,

You are right on. I touch on just a few of them in this article:
http://www.mountaineagleweb.com/Sear...ger-Update.htm

Ken