[ndiddy]

This has never been a serious question. People have looked at DOS for years and never found any stolen CP/M code. Tim Paterson has written about how DOS was designed here: https://dosmandrivel.blogspot.com/2007/09/design-of-dos.html https://dosmandrivel.blogspot.com/2007/08/is-dos-rip-off-of-.... His employer was selling an 8086 computer but had no operating system for it, so he proposed making QDOS, a placeholder operating system that they could sell until CP/M-86 was available. He made QDOS compatible with the publicly documented CP/M API, but did not look at how the API was implemented. This made it possible for software companies to easily make QDOS ports of their existing CP/M software by automatically translating the source code from 8080 assembly to 8086 assembly. Paterson himself took advantage of this. The 8086 assembler and source code translator were originally CP/M programs written in 8080 assembly, then he translated them to 8086 in order to make QDOS self-hosting.

It's somewhat ironic that Kildall was angry about DOS copying the CP/M API because Digital Research went on to release DR-DOS, an 8086 operating system that was API compatible with MS-DOS.

[shrubble]

Killdall viewed it as a serious question, at the least.

More accurately, “the assembly language programmer Tim Paterson has always claimed that he did not look at the implementation”.

That Kildall later decided that turnabout is fair play, is not that surprising…

However my point is that now, everyone can assess the source code for themselves.

[ndiddy]

> Killdall viewed it as a serious question, at the least.

The best source I've seen for Kildall's viewpoint is his chapter in the book They Made America. It quotes sections on this dispute from his unpublished memoir (his children later allowed the Computer History Museum to publicly release a portion of the book, but not the part that They Made America quotes from). Kildall was angry that DOS was API compatible with CP/M, and seems to have believed that this constituted a copyright violation:

> “Paterson was not writing a computer application according to DRI specs, i.e., cooking from a recipe. He was creating a derived work based on the cookbook copyrighted by someone else. When Paterson wrote QDOS with Kildall’s manuals “at his side” (in the words of Gary Rivlin in The Plot to Get Bill Gates), he was using materials marked on every page: “All Information Contained Herein is Proprietary to Digital Research.”

> “So that software developers would know how to write programs for its still-secret project, IBM had to let selected programmers have a list of API function calls. Kildall was angry to find how much of CP/M’s proprietary list appeared there. He had no idea IBM had a deal with Gates. He was just upset that IBM itself seemed to have copied his interfaces. In his unpublished memoir he says he furiously got through to IBM. They immediately dispatched a manager and an attorney to Pacific Grove. “I showed the IBM attorney definitive evidence that PC-DOS was a clone of CP/M and immediately threatened a lawsuit for copyright infringement. The IBM attorney compared the API interface, and I can say clearly that he fairly blanched at the comparison and stated that he was not aware of the similarity. I told him that he should take note and become aware at the earliest opportunity, or else he should face a major lawsuit.”

> “By the time he wrote his memoir, Kildall saw the decision not to sue as a fateful error. He grew increasingly irate about the similarity of PC-DOS and CP/M. He writes: “The first twenty-six function calls of the API in Gates’s PC-DOS are identical to and taken directly from the CP/M proprietary documents [CP/M manuals].” Then he poses a challenge for his old friend. “If you think Bill Gates invented those function calls, ask him why print string (function 9) ends with a dollar sign. He will not know.”

> “What Paterson essentially did was rewrite the bottom part of the software—improving the way files were stored and adapting the program to a 16-bit machine—while copying most of the top part of Kildall’s operating system interfacing mechanisms. Even if QDOS and CP/M were 80 percent different, as Paterson has insisted, he took almost unaltered Kildall’s interrupt mechanism—the key innovation.”

> “To demonstrate how far Paterson mimicked CP/M’s interface, the first 36 Int-21 functions, Kildall’s memoir devotes an appendix to comparing the sequence and language of CP/M and those of QDOS and MS-DOS. A few words were changed. Kildall’s “Read Sequential” function became “Sequential Read”; “write sequential” became “sequential write”; “Read Random” was called “Random Read.” And so on.

At the time Kildall wrote his memoir, and when Harold Evans wrote They Made America, the question of whether APIs are copyrightable had not been answered. However, Google v. Oracle has settled that writing an implementation of someone else's API is fair use.