garym: The key question is how is the originality of free software to be judged? My guess is that while the great mass of orginality software ideas (how on earth do you measure originality?) lies in proprietary software, in proportion to the time spent on it, free software is massively more original than proprietary software. One of the features of free software that hasn't attracted enough attention, I think, is just how efficient the best projects are in terms of getting maxiumum results from minimum effort. This property naturally fosters originality.

Politics Free Zone revisited
cmm and raph both replied to my last diary entry, both being rather less positive about the community standards here on Advogato than I was. A couple of points:

1. Raph thinks that the current anti-war movement is a general exception to the usual informal rule against talking politics here. I find this odd: why this conflict? Is it because he cares a lot, or because a lot of people care a lot about this conflict, or something else? The former is in effect a universal exception; in the latter case, shouldn't this mean that pro-war postings are as justified as anit-war postings?
2. My own feeling is that one should be considerate of the general advogatan sensibilities when writing diary entries (which I'm afraid I am not always, looking back at my earlier entries), but beyond that there are no rules about what is appropriate. I don't think it is a problem if diary entries are not particularly free software focussed - my own experience is that writing "off-topic" entries seems to be a kind of therapy that helps keep up my motivation to work on free software projects.

While I'm thinking about the almost-certain-to-come conflict, this article at the London Review of Books is about the best article I have read on the divisions created by the conflict. Raph gives this article at interesting-people.org high praise: I have some reactions to it, but they will have to wait until I have more time.

Proof of Correctness Wars
This ACM article from last summer is required reading for the now rather dormant discussion on web-based proof assistants. I think it might already have been mentioned here on advogato, but it makes good points and I think folks interested in the issues might benefit by looking over it again. Serious point: I think if the not-too-clear ideas going around about web-based proof assistants come to something, then we will be revisiting this debate again. Not so serious point: Dijkstra's halo doesn't look so firm in this retrospective.

Postscript fxn pointed out only ACM Portal users can access the above article: I'd be grateful for any pointers to non-crippled URLs of the text.