> > Mine is quite a bit more pretty, and only slightly less pretty. I think
> > though, even if you through away the rest, you should keep the
> > no-repeated-names rule (important for position-independent parseing, since
> > otherwise, there is no unique index key), and the format rules for values
> > ("0x"ed hex, MSB first, yes/no bools).
Please try and quote a bit less when you reply. It's tedious wading
through acres of text.
(I would have sent this by private mail, but Richard never seems to get,
or at least take notice of, any mail I send that way).
> Okay, now you are talking about making a standard that will allow the
> entries to be parsed with <don't flame me> obsolete Unix tools
> like `ed` `cut`, etc. This would certainly be useful, but do you want
> to design the specification around this?
Those tools aren't obsolete. Maybe you don't use `ed' any more, but 'cut'
(and things like `awk') are definitely still in widespread use. And yes,
I think it would be beneficial to be able to parse /proc with them.
> My guess is that this discussion will go no where because everyone
> has their own idea (rightfully so) of what is best. Hopefully, at
I think you may be right. Certainly as long as people keep raving about a
binary /proc.
p.