What are "fake", "shady", and/or "predatory" journals?
A "fake" or "shady" journal is a low-quality journal that does little or no quality control.
They are often called "predatory" because they prey on people who are under a lot of pressure to publish, charging high author fees for promises of quick publication. (Since they do little, if any, peer review, the time from submission of a paper to publication is often very quick in these journals).
They often engage in deceptive practices to make themselves appear legitimate, such as:
- Pretending to be affiliated with a reputable professional society
- Claiming an "impact factor" when they do not have one, or when they have an "impact factor" from some entity other than Thomson Reuters ISI
- Listing important academics on their editorial board, when these people never agreed to serve in this capacity.
Publishing in one of these journals can be very damaging to your academic career. At best, it shows that you don't know what journals are considered reputable in your field; at worst, it makes you look like you are trying to "get" publications without doing the work required to publish in a reputable, high-quality journal.
An academic librarian named Jeffrey Beall kept a list of open-access journals and publishers that he considered to be "predatory." (The links are to archived versions of the pages). His lists were quite well-known - you may hear people refer to "Beall's List" when talking about predatory journals.
Between the growth of open-access and the Internet making it possible for literally anyone to start a "journal," these "journals" have been popping up at an alarming rate. The NY Times even ran a story about it recently.
Jeffrey Beall, who works for the University of Colorado, Denver, maintained a list of predatory journals. He also publishes his criteria for determining which journals/publishers are predatory. (both links are to archived version of the pages) The criteria are numerous (and I find many of them amusing). The main thrust of the criteria are such things as:
- Lack of transparency (in both business model and editorial process)
- Disregard for intellectual property
- Promise to publish anything in exchange for money
- Dishonesty (in business model and editorial process)
- Poor (or no) peer review
... and a very good question indeed. The present answers do a good job, but it's worth emphasizing that there is a whole continuum between legitimate journals and what amount to conning schemes.
"Predatory" is used to describe a journal that hinges its business model on getting article processing charges from authors, whilst providing a very low quality journal. "Fake" is less defined, but it carries the connotation of a journal that pretends to carry out rigorous peer review but in fact doesn't (which is what makes it damaging on a CV: it looks like you too want to skip formal peer review but pretend you still did it).
"Shady" is a much more informal term, and denotes a journal anywhere in that continuum. I intentionally used this term in my answer to keep it broadly applicable, in terms of any journal of possibly not-so-good standing which is, after publication, thought to have a negative impact on one's CV.