The complex code above was even loaded with keywords (using asp code). These keywords signaled to the Web server at the spam target site the type of dynamically generated page that should be served in response to the query. These tactics are not approved of if they're done deliberately to manipulate search rankings. Visitors to this site were totally oblivious to the devious intent of the site owner, and search engines were fooled as well.
Publishing Empires
When a publisher builds a vast array of interlinked Websites, it can generate high PageRank and subsequent rankings. This form of spam is difficult for a search engine to penalize, since the links are legitimate. Any single business entity has the right to interlink its own Websites. The company can create further overlap between the sites' content themes so that the links are truly valued by search engines.
This kind of activity is exemplified by one of the Internet's largest publishers. The business has 120+ Web properties, all of which are carefully linked to the others. Perform a search on one of these sites, and you're virtually guaranteed to see one of the company's other Web properties in the search results.
Many among the most successfully ranked sites use this system -- this form of spamming is extremely widespread. Perpetrators basically collect PageRank and link reputation within their network, then use it creatively to dominate the best keyword phrases. Search engines haven't found a way to stop this technique, but they'll have to. This form of spamming is a major threat to the quality of search results.
Wikis
Wikis are Web repositories to which anyone can post content. They can be a great way to present and edit ideas without close censorship, and have proven extremely successful for the creation, management, and maintenance of projects that require input from users around the globe.
However, despite their considerable advantages, the often un-scrutinized nature of wikis makes them ripe for abuse. Like a link farm, a wiki's links are free for all. Ironically, the value of wikis is consistent with popularity-based search engines. Some of these wikis boast a very high pagerank, which can make the wiki an attractive place from which to gain a link to your site. But without close human control, users may simply add their links as a means to take advantage of the wiki's PR. Until another user of the wiki removes the link, the linked site enjoys the benefits of this unscrupulous activity. The search engine spammers have control.
Networked Blogs
Blogs can be a source of precise, up-to-date and technically detailed information, presented by specialists and experts. Blogs are thus very valuable to info-hungry searchers, and are extremely popular.
However, some spammers start a blog, plug it full of garbage content such as comments on what they thought at 5:15, along with a link or two and some keyword rich text. Keyword rich musings don't present real value to deceived searchers. Worse still, blogs often operate in a free-for-all link structure that further validates the linked sites in search engine indexes.
Forums
Like blogs, forums can be a rich source of relevant information.
Unfortunately, some forum participants make comments in forums only in an effort to publish links back to their own sites. This may be acceptable if the user provides help or assistance to another forum member. Indeed, they should gain credit for that information, which they may have worked hard to discover.
However, when the posts become excessive and are comprised solely of glib or irrelevant comments, then value of the link, or indeed, the whole forum, can be put into question. Some forum owners only start forums in the hope that they will raise search engine rankings.
Domain Spam
Probably the most popular spam technique today involves creating and hosting a number of Websites. These sites rarely have any intrinsic value other than providing ranking support for the owner's main Website.
I've had several former clients who had used this technique -- and had been penalized for it. After I got them to get rid of the duplicates completely, their rankings were repaired.
Duplicate Domains
Why can't Google detect two exact duplicate Websites that only differ on domain names? Why would Google give these same sites first and second rank for the very same phrase? This happens all too frequently and is due to Google's preoccupation with linking between topically related sites.
Domain spam is usually the result of a corporation's attempt to have Web sites for each of its company departments or subsidiaries. Those with many subsidiaries get a big boost from these domains. Realizing this, spammers are increasingly encouraging clients to have sites hosted on different IP addresses and even in different geographical locations.
The link pattern detection used by Google has difficulty dealing with this practice, and is currently failing to cope with it. Google's new emphasis on authority sites actually makes this matter worse, as the authority can gain trust it really doesn't deserve.
Links Inside No Script Tags
One top publishing site I recently discovered secretly interlinked its sites using the no script tag. Although I can't name the site, I can show you how the technique worked.
Used legitimately, the no script tag provides spiderable links when a user's browser (or a search engine robot), has its javascript turned off. Anything that appears inside the no script tags is not visible on the Web page itself.
To be used authentically, the no script tag must contain links that replicate those used within JavaScript code in the actual page.
But in this case, the links went to sites that strategically collected PageRank. They were basically hidden, acting as underground network of links to support the publisher's rankings. This code appeared in almost all of the site's many domains -- and perhaps exists in the Websites of others, who may not even know it's there! Some of the pages only used a closing tag, which could also confuse search engines.
Please check source link for code demo..
The complex code above was even loaded with keywords (using asp code). These keywords signaled to the Web server at the spam target site the type of dynamically generated page that should be served in response to the query. These tactics are not approved of if they're done deliberately to manipulate search rankings. Visitors to this site were totally oblivious to the devious intent of the site owner, and search engines were fooled as well.
Non-Robot JavaScript Detectable Redirects
The use of mouseover code like that shown below is quietly spreading across the Web:
Please check source link for code demo..
There have been rumors that Google is taking action against this tactic. In the cases I discovered, the JavaScript code automatically redirected the visitor to another page, but only upon the cursor being moved over the page itself. It was almost impossible for the user to avoid setting this code off.
I found the code on a site ranked number one on Google for its primary keyword phrase. As search engine robots don't use a mouse, they're blind to the spamming activity. In this case, the tactic was combined with a server side redirect to another page, which was relevant only in some cases. The purpose of the redirect may have been part of bigger ploy to support another ranking strategy.
Dynamic Real Time Page Generation
It is possible for a Web server to produce and serve different, optimized pages according to the referrer of any page request.
In theory, there is nothing wrong with serving a page that's customized to the circumstance in which it was requested -- indeed, many ad campaigns serve up different ads based on the type of banner that was clicked. Customized ads are seen as being far more effective and useful for users.
With dynamic page spam, however, the site is loaded with hundreds of these phantom pages (dynamic urls) that act as affiliate links to some other site. Search engines don't want affiliate links. In the case I found, all the links were credited to the site's backlink count.
I don't think this is what search engines had in mind when they began to spider dynamic urls -- they certainly don't want to allow affiliate link spam.
Here's what the links typically look like:
Please check source link for code demo..
DHTML Layering and Hidden Text
Using DHTML layering, spammers can hide layers of keywords beneath graphics. One layer covers the other visually, yet the text hidden on the lowers layer is readable by the search engine robot -- another highly illegal technique.
HTML Hidden Table Cells
The combined powers of CSS and html and the loose dtd allow the unscrupulous site owner to hide the content of table cells loaded with keywords and heading tags.
CSS permits the flexible positioning of Web page elements; it's a flexible coding language that search engines do not fully understand at present. In short, the search engine doesn't really know what's being displayed. This trickery can be specified in a separate CSS sheet (.css file), which a search engine may or may not index. This CSS style sheet file, however, does affect the display of content on the page.
In this example, the CSS affects the display of the body of the Web page, which is set to 97%:
Please check source link for code demo..
Within the regular code, .gif files can be placed in the page at a width of 150%, ensuring that part of the page is not seen. That extra 50% provides plenty of room for keywords stuffed .
Enormous Machine-Generated Websites
Those Webmasters who are not adept to html, dhtml or css tricks may try something simpler. When there's not enough content to go around, they often try to stretch a minimal amount of content across thousands of pages. The pages are built with templates and the sentences within them are basically shuffled from one page to the next. Unique title tags are plugged into each page that's generated.
This technique basically sees the same page repeated hundreds to thousands of times. It can even be done using a computer program that systematically stuffs the text sentences, paragraphs and headings, including keywords, into pages.
This technique is most often used with ecommerce sites that have a limited range of products for sale. Often, the products are simply re-organized, or shuffled to create another page that appears to be unique. It's actually the same selection of products presented countless different ways.
Link Spam
To maximize Pagerank distribution throughout a Website, some spammers will fill a page with links to the point where it is just a links page, and every page links to every other page.
Why do this? Well, by maximizing the number of links, the spammer more equally spreads PageRank throughout his or her site. When links from all those pages point to a single page on a keyword topic, the site can gain higher rankings for that phrase.
Link exchanges are also considered link spam. The links are fabricated -- not a real reflection of personal choice. Most link exchanges are now being filtered out of search results; however, some links in link exchanges are still being recognized.
This system allows the server to give the robot different content than that which is delivered to human visitors. And that means the search engine could be deceived.
Invisible Text
Invisible text is invisible because the font color is the same as the color of the background or background image.
In one example I saw, a site used the font color "snow" to make the text white on a white background. The author also used this font tag in a way that caused it to overlap another tag, thereby confusing the search engine robot further.
The example below uses a black color .gif as the background to hide black text. It also has a dhtml layer directly above it, to further hide the text.
Please check source link for code demo..
A robot can't detect whether text in a dhtml layer is the same as the background used in a layer below it. The layer can even be set off-screen so it is never visible to a person.
Link Farms
Link farms are still prevalent on the Web, even though search engines can detect their presence through link pattern recognition. Since link spamming is being done at a macro level, so search engines must be able to view a large sophisticated network of links and delete those that are machine-generated and not true, human-chosen links.
The hilltop algorithm is one filter that minimizes the advantage gained by hundreds of useless links.
Spamming Penalties
Each search engine has its own distinct prohibitions and related penalties. Each penalty is a response to the degree of threat the search engine that a given spamming technique represents.
Spammers may receive demerits, through which the ranking of their sites on a particular phrase might drop significantly. Alternatively, a zero PageRank penalty may be awarded to a particular page, or whole sites may be banned if the search engine so chooses.
Now that these techniques are widely known, I strongly advise you not to try them. The search engine engineers may be embarrassed that these tricks really do work, and will move swiftly to take action against spammers.
No comments:
Post a Comment