Python's new way of representing floating point numbers is only a partial solution. The bigger problem is accumulation of errors. To see this: add 10 copies of the number "1.1". Now subtract 11. The result will be tiny, but non-zero (-1.8e-15, on my machine). This is a much more serious problem than the 10.99999999999999998 representation problem.

This is almost, but not quite, what I want. I want sloppy focus for mouse-wheel scrolling only.

Is anyone aware of such a thing? It would make my windows work just a wee bit less bothersome...

I gave an impromptu talk at an EuroFOO conference 5 years ago about exactly this problem: http://homepages.cwi.nl/~jack/presentations/OpenSource-EuroFoo.pdf.

My feeling is that the basic problem is that, in open source, at most 5% of the people involved are non-programmers (read: non-geeks). And for most projects the number is probably exactly 0% of the people involved. for shareware projects it's close to 50% (half of the developer:-). For commercial projects it's somewhere in the range of 20% (small vendors) to 99% (Microsoft, big software houses).

The input of the non-geeks, while usually dismissed by us geeks as fluff, can be really, really important. Because their interested in such technical trivialities as documentation, ease of use, learning curves, market acceptance (and, yes, financial bottom line too). Those trivialities are important even to hardcore geeks when the software in question is just a tool you need to get the job done (as opposed to the labour of love you've been spending years of your life on).

This article is pretty interesting from a philosophical point of view. One of the first things I realised is that if there is indeed an effect from our measurements it must mean that no sentient beings have been doing such measurements earlier in the lifetime of the universe.