Don wrote:I cannot imagine Ken Thompson, the Bell creator and world renown computer scientist and Unix creator being sucked up into a purposeful lie for example.
I'm not excluding the possibility anymore that that might have happened, maybe with the word 'purposeful' removed and 'lie' replaced with 'confusion'.
I didn't see that coming during the process, while I was thinking I was specifically looking for that pitfall at the time. As it happens, there is a potential hole in the investigation, which I now regret I haven't addressed properly a year ago. Actually I recall I have addressed it briefly but for some reason I didn't follow it to the end while I should have, and I don't know why.
The point that is that the letter of rule 2 as formulated in 1977 is quite clear. But the spirit and how it evolves over time, with exemptions made verbally in players' meetings and what has been accepted historically is not so. The contents of older application forms and prior rulings relating to rule2 are the
only tangible manifestations of that spirit, and these were not extensively discussed within the panel. Ken would be as unaware of them as are newer members and outside experts.
Assuming that the 1977-spirit of that rule exactly equals its wording, VR violated rule 2. Not by code copying, but by writing code that was derived from Fruit's design.
But the spirit evolved/refined and the wording was never updated.
For example: Deep Thought's design was based on Belle's (1) but not copied at implementation level (2). This happened with knowledge and consent of Thompson (3), and it was 'known' at the time (4) and Belle didn't compete anyway, so it was ok (5). But was this relation with Belle actually -written- on the DT's entry form? I don't know and I want to see it. Because one can argue that the same case applies for VR: Rybka was based on Fruit's design (1), but not copied at implementation level (2). This was public knowledge by VR's statements on his forum (4), with consent of FL (3) (indirect permission: by open-sourcing Fruit. The GPL license that Fruit came with only protects distribution, but playing in a WCCC is not distribution. This is precisely why long time ago Crafty's license was changed: because the GPL didn't prohibit this. But Fruit has GPL, not Crafty's license). And Rybka didn't compete at the same time with Fruit (5).
That, combined with one of the secretariat member's explanation that he only considers literal reading of words but that in his opinion gut feeling is enough to recognize a clone, while the ICGA journal clearly writes that rule 2 only has a meaning in historic interpretation, pushes a 'reset'-button with me. Maybe the panel
was unknowingly directed too much into one line of thinking, a literal one. I was part of it and I can't answer that question.