I am reading with astonishment an article in IEEE Spectrum on the origins of DOS. The author, a self-proclaimed expert on software intellectual property analysis, describes his attempt at a forensic comparison of early versions of MS-DOS and CP/M, to prove or disprove once and for all the allegation that MS-DOS was a result of theft.
But I find the article poorly researched, and also a thinly veiled attempt to plug the author’s company and analysis tools. Childish comparisons of identifier names and code fragments… really? The issue was never verbatim copying but the extent to which QDOS (which is the operating system Microsoft purchased and renamed) was derived from CP/M. It is clear that it was heavily influenced by CP/M, just as CP/M was heavily influenced by its predecessors, including operating systems written for the PDP-11. Does this constitute infringement? I certainly do not believe so. Indeed, something very similar (albeit more formal) occurred a little later, when the first IBM-compatible “clones” hit the market, and companies like American Megatrends, Award and Phoenix created binary-compatible versions of the IBM PC BIOS using “clean room” reverse engineering.
Some online commenters went so far as to ascribe ulterior motives to the author and question his sincerity. I think that is uncalled for. However, I do believe that this article should not have been published in its present form. At the very least, the author should have been advised by competent editors to tone down the plugs; to do a little bit more research on the topic; and to shift the emphasis from meaningless code comparisons to an analysis of the functional similarities between the two operating systems, the possible origin of these similarities, and the question of whether or not they might constitute infringement (and the extent to which the law may have changed, if at all, in this regard between 1982 and 2012).