Spamdexing is an English neologism composed of the substantive Spam and suffix dexing taken on the term indexing meaning referencing. Into French, one translates spamdexing by abusive Référencement .

It is a whole of techniques consisting in misleading the Search engines on the quality of a page or a site in order to obtain, for a given keyword, a good classification in the results of the engines (preferably in the very first results, because the users seldom go beyond the first page which, for the main motors, includes/understands by defect only ten addresses). It is sometimes severely punished by the engines, even if there is no precise code of conduct for the référenceurs (it is sometimes difficult to distinguish abusive referencing from SEO, “honest” optimization). The usual techniques of abusive referencing consist for example with truffer a satellite Page of lists of keywords (to attract the users of engines which make a research on these words), or to create tens of sites which point the ones towards the others ( link farms or seedbeds of bonds) to improve their classification in the engines which judge the quality of a page according to the number of bonds pointing towards it.

Operation

In theory, the search engines classify the results according to the quality of the pages and their relevance compared to the request; but the current engines (being thus opposed to the directories, produced by the human ones, which refuses the sites of insufficient quality) try to estimate the quality and the relevance of the pages by automatic processes, whose principles are known, in their broad outlines, by the polluposteurs and the optimizers of sites:

• a page is supposed of good quality if a great number of external bonds point towards it (when an originator of Web page places a bond towards a page, he is thus supposed " to vote " for this page); it is easy to create several sites which point towards the site that one wants to promote (or to exchange bonds with friendly sites, managed by other people. It is the " Netlinking ", literally " Setting in bonds of the réseau" , commonly called " Exchange of bonds ").

• a page is supposed to be relevant, in answer to a given request, if it contains many words present in the request
• the various words of the page obtain a more important weight according to their site (for example, if the expression " sale of voitures" appear in the title, the page is most probably devoted to this subject)
• the engines also take account of the words present in the address of the page (what explains why one finds sometimes URL long, with repetitions of words, like www.exemple.com/voyages-pas-chers/voyage-en-chine/voyage-en-chine.html )

The techniques of referencing evolve/move in time and adapt to the engines. A novel method is born: the " saturation by integrations multiples". The principle is the following: the holder of the site to be promoted proposes his contents with a series of partners who have a domain name with a high pagerank and a raised number of pages, which will facilitate their rise in the results. Example: www.site-du-spamdexeur.com proposes the contents. Then, one finds the same contents on http://mot-cl é.partenaire.com, http://mot-cl é.partenaire2.com, etc In a saturation results from the page of results of the search engines. One can thus obtain 80% of the results of a search posted on first page by the search engines. As the majority of the clicks are done on the first page of results of a request, they secure a maximum of visibility thus and évincent their competitors.

Abusive referencing and the Right

Abusive referencing enters in total contradiction with the Loi on the practices of the Commerce. Moreover it is connected with data-processing Fraude since the goal is to divert an automated data-processing process of its initial goal and this with an aim of personal enrichment. This article covers subject well: http://www.journaldunet.com/juridique/juridique040127.shtml

Ethical referencing

In opposition to the techniques of referencing known as abusive, certain people advance the idea of a referencing " ethical " supposed to rest on a code deontologic. Various sites, or association of référenceurs, are advanced to propose their vision of a code deontologic as regards Marketing of the search engines. Of course these precepts do not have any force of law, vary from one individual appreciation to another, and engage only those which want to be well recognized in such models " éthiques". These same codes of ethics are written by intimidation of the search engines. It is however strange to note that the search engine which occupies 90% of market share adopts various positions with regard to the spamdexing. Sometimes it tolerates it by prohibiting it, sometimes it reprimand heavily (blacklisting of the index) without preventing those which have recourse there… These actions are connected in some kind with an dominant position abuse because the actor in dominant position distorts the play of competition.

Dissimulation of the junk email

Not to give suspicions to the user who would see on his screen a long list of words, the many terms placed in a page “to trap” the engines are often camouflaged by various processes:

• relegation of these lists of words at the foot of the page

• writing in tiny characters
• words placed in a section " noframes " , " noscript " or " display: nun " (generally not posted by the navigator, but read by the robots of the engines)
• of the same characters color than the bottom of the page (what makes the text invisible)
• driving or directories posting of long lists of “last research” or “popular research”
• dynamic pages - for example those of search engines - disguised on static pages, with addresses such as example.com/trouver-requete.php: such an address resembles that of a static file which would be invited to find-requete.php, which would be located on the waiter of the engine, on the waiter, whereas it is acted in fact of a dynamic page (the exit of a script, posting the results of a search) created at the time of the request: the fact “of thus disguising” the URL makes it possible to facilitate its indexing if it is supposed that the dynamic pages can not be indexed by the engines, or obtain a classification lower than that of the static pages. In general, the pages of results of the main motors have addresses such as example.com/search.cgi?requete, where the contents of the request are not disguised in file name; moreover, these engines expressly prohibit the indexing of these pages by means of a file robots.txt
• Retrait of the words via a script (e.g.: Javascript)
• a satellite Page ( doorway ), truffée of keywords, is read by the robots of the search engines; but when human consults it, it is redirected towards another page (and thus it does not see the can page).
• the Cloaking (occultage) consists in having different results according to the software used to post the page: an alleviating page for a navigator Web, an optimized page, filled out of keywords, reserved for the robots of the engines
• the companies of SEO, on their banner page, give examples of sites which they optimized, each one of these addresses being placed behind a word describing the subject of the site in question; what makes it possible the pages of the optimizers to contain words which have nothing to do with their activity (and thus to appear among the research results relating to these words). They can also put a bond towards their own site in each page which they modify

See too

• Splog

• satellite Page
• Pagejacking
• Occultage (also called cloaking)
• Métaélément
• Referencing
• Optimization for the search engines

External bond

• Technical of abusive referencing - Article on Prohibited Web
• SpamReport for FireFox easily Defer a spam in the index of Google
• free Repertoire of search engines and directories giving access each time possible to the form of indexing - free tools.