[oclug] Linux kernel development loses BitKeeper

[miden]

At band camp, one time, Brad Barnett said:
> You have missed one important thing.  These directions tend to take into
> account the tolerances of those sampled for reaction, which are generally
> not human beings.
>

My point was that not enough account is taken of the probable rate of 
failure to comply with instructions (protocol). This is a serious failing.

> There have been many cases of people being "sensitive" to pesticides. 
> Even following directions, these people are horrifically effected by their
> use.  It is the same with resistance to radiation.  Every individual has a
> unique tolerance to radiation, and the same can be said for tolerance to
> pesticides.

Very true. Cost-benefit ratios are simply a prediction of the number of 
people who will probably be affected in a harmful way. If the number is 
considered low enough then the benefits to society are judged to 
outweigh the risks to the individual - and those who are damaged are 
acceptable casualties (nice, huh?). Business growth seems to be viewed 
as a benefit.

Is not having to get down on your knees and dig out weeds really a 
benefit to society? Or is it more a benefit to an industry? I think the 
answer is obvious.

-snip-
> There are many cases of products that take decades of use, before people 
> notice the subtle and devastating effects they have.

True. But this raises a very real quandary - do you ban everything that 
can't be proven to not have long-term effects (which means everything) 
and if you do how do you prove that they do or do not have long term 
effects (on human beings).

I for one don't want to see development freeze-framed to the present. I 
would, however, like to see much, much more caution in the approval process.

-m