A lot has been said about ArticleBot on the Web. The blogging world, especially, has made a huge deal about the program and the name seems to be spoken with the same fear as a boogeyman.

However, that’s not at all surprising considering some of the rumors that have been going around regarding the software. One site recently referred to it as a “Tool from hell �? and said that it “Takes your content and tries to automatically rewrite it in order to prevent detection of the fact the content was stolen from you.�?

However, when searching for information on the program, I wasn’t able to find many hard facts on what the software actually does. Even the ArticleBot Web Site is strangely devoid of information on its public pages.

Concerned and worried, I decided to contact ArticleBot’s creator, Don Harrold, to hear what he had to say about the product. The answer was very surprising.

What ArticleBot Does

ArticleBot, fundamentally, is a text manipulator designed to create permutations of an article. It’s function is not to steal content, but create it, lots of it.

The idea is pretty simple. If you had a sentence that read “I love my brown dog�? ArticleBot could use synonyms for the words “love�? “brown�? and “dog�? to create new ones. If you had five synonyms for each of those three words, you’d have 125 permutations of the sentence immediately. “I love my brown dog�? could be come “I like my gray puppy�? or “I cherish my black mutt.�?

If you applied this same logic over the length of a 300 word article, you could come up with hundreds of thousands of potential permutations, each of them unique. Also, ArticleBot is capable of editing HTML code in much the same way, changing alignments, lengths and creating not just thousands of original articles, but pages as well.

These articles (or pages) can then either be uploaded to the Web, copied and pasted into a blog or otherwise reused in any fashion.

What ArticleBot Doesn’t Do

Contrary to popular belief, ArticleBot does not scrape from RSS feeds nor does it have the capability to do so. Yes, people can copy and paste into AB and they can use third party software to automate the process, but AB is not a feed scraper in and of itself. Instead, the only XML capabilities it has are to create a feed for the content, not read another feed. According to Harrold, an RSS scraping feature has been requested many times and he has flatly refused to add one.

There is indeed a scraping tool built into AB, however, it scrapes solely from search engines based upon terms provided by the user. Though there are some copyright questions here since the collection of information is copyrighted by the respective search engine and that, in turn, is made up of the copyrighted content of thousands of Webmasters, it mainly seems to consist of site descriptions and short snippets. Quite frankly, the scraping tool produces mostly unreadable and meaningless gibberish.

On the other hand, the articles themselves are, by in large, very readable. Though they aren’t likely to win any Pulitzer prizes and, many times, read as if they were written by someone who was new to the English language, they are intelligible and do resemble what a human might write.

Still, while there are some copyright concerns in dealing with certain AB features, the tool itself is not designed as a platform for plagiarism or copyright infringement.

The Idea Behind ArticleBot

ArticleBot is designed to create a large volume of search engine-friendly unique content in rapid order. Some have called this “Black Hat�? Search Engine Optimization (SEO) and others have flat out called it “cheating�?. However, Mr. Harrold feels that it’s a legitimate technique for dealing with search engines that, he feels, are crippling free speech and creating an oligopoly of information.

But even though the ethics of the software’s search engine goals will be debated for a long time to come, it is actually a very poor tool for use in plagiarism. Though many, understandably, fear that AB will be used to create derivatives of their work that will be undetectable to search engines, the software is very complex and requires a great deal of time to prepare an article of “spinning�? as AB calls the permutation process.

If you’re trying to produce ten thousand derivatives of a piece, AB saves you a great deal of time. However, if you’re trying to simply modify a piece so that it looks more like your own work and is more difficult to trace back, AB is a very slow way to do it. You’d have a much easier time editing it by hand using a word processor and you’d most likely get better results.

The Potential for Trouble

The danger of AB seems to lie less in the software itself than the people who use it. It seems logical that people in a hurry to create hundreds of permutations of an article would also, very likely, be in an equal hurry to obtain the content to start with. Though I can not and have not confirmed this, it is very likely that at least some of AB’s users are currently lifting content illegally for use in the program.

A potential solution to this problem would be to encourage AB’s users to pull their content from the very large library of public domain material or from sites with Creative Commons Licenses that permit such reuse.

Still, those who do steal content for use in this service are going to create countless variants of the original. Though it’s somewhat likely these modified clones would be considered unlawful derivative works under copyright law, tracing them back to the original source would be an incredible challenge as the two could, very likely, look nothing alike.

As such, if anyone were to read both works, it is unlikely that they would ever relate the two pieces to one another and searching for it in Google or Copyscape would be an exercise in futility. There is little doubt that this is a very frightening proposition for many Webmasters who might have to compete with thousands of computer-modified copies of their work.

However, committing plagiarism or copyright infringement is still a decision left to the end user. Like a lot of software packages that can be used to aid the act, including Microsoft Word, AB does not directly encourage the act, but does have features that, though designed for other uses, could be turned around.

Article Bot: Preventing Plagiarism?

One suggestion that Harrold made was to use ArticleBot not as a plagiarism aid, but as a tool to prevent it. The idea is to create an original work that you’re proud of and then use AB to create derivatives of it. If any of the derivatives get plagiarized, the original is still safely in your hands.

While this defeats much of the purpose of posting an original work to the Web, especially since it’s not entirely your work any more, some might take solace in this kind of protection. Still, as I said before, using AB to create one derivative of something doesn’t make a great deal of sense. A writer wanting to create a single derivative of a poem or article would probably be better served by doing it himself.

Nonetheless, it is an interesting concept that certainly bears some thought.

Conclusions

Certainly, there is much about ArticleBot that will worry and perhaps anger many Webmasters, especially those in the SEO field and those who have to compete with largely computer generated sites. Still, it’s not the massive content thief that many thought it was. That, as it turns out, is more a matter of misinformation and confusion.

Though it is a stretch to call it harmless, like most things that aren’t understood, AB has taken on a reputation that is far more dangerous than its reality.

Still this relief comes with a very serious caveat: That the software we have been fearing is still out there, being used by truly sinister people and enjoying all of the attention that has been paid to ArticleBot.

Link: ArticleBot Tutorials – See firsthand what AB actually does.

Protect Your Work. Subscribe to Plagiarism Today via Email or RSS.