I wrote a post a while ago on what made for a 'good' (as in reliable) journal, in terms of the nature & quality of the research reported in it. Now it seems that there's a new low on the scale - in 2003 a pharmacological company paid a publishing house to produce a 'fake' medical journal: much advertising, & reprints of papers that had appeared in other journals in the publisher's stable. Orac provides a very thorough commentary on this - valuable reading if you're interested in how publication works & what makes for a good place to publish. As he points out, the quality of the reprinted articles isn't in question - but the fact that they were in a 'pseudojournal' stuffed with the drug company's advertising, is just downright wrong. As one of Orac's commenters notes:
A fake academic journal creates the appearance of a conflict of interest. It is designed to look like legitimate peer reviewed research published by a nominally independent publisher (Elsevier) who in fact is being paid to publish research in this journal which supports the products of the particular medical products company that is paying Elsevier to do it. Thus there is a conflict between Elsevier's ostensible role as a leading publisher of peer-reviewed scientific research and their actual role of being paid to produce advertisements for a particular company. And it was ethically wrong for Merck to suborn Elsevier's conflict of interest.
It's one thing to create a throwaway journal that is clearly identified as such and kept at arm's length from any academic journals published by the same company. It's another thing to create a throwaway journal that is specifically designed to look like another of the many academic journals published by the same company. The arm's length relationship between marketing and scientific publishing is key: without it, we scientists cannot trust the scientific publishing.