On Tue, 20 Apr 2010, Bart Smaalders wrote:
> It's clear that the current "design" of the interface
> embodied in power.conf is poor when reading the man page.
(Personally, I think this is a generous evaluation).
>
> What is needed is a design for a new interface, and
> the SMF properties needed to implement that interface.
> Proposals to hide data elsewhere are simply wrong.
>
> The complex current semantics of power.conf are not desirable
> or maintainable, and it should be abandoned as an interface.
>
> Arguing about which properties are needed vs. which
> government standards is pointless without a design
> for a new interface.
>
> Good OS interfaces are about abstraction of hardware/platform
> differences. Trying to design your way to a correct
> interface from the current pile is pointless; decide
> how it should work, keeping in mind both technical limitations
> and business priorities, and design a new interface from
> scratch.
To which I fully agree. It is similar to the comment I have
actually have been making for a while: decide how we want it to work,
and make that happen, and don't worry at all about what is there
already. After we know how we want something to work, we can see
if there is anything there that we can reuse. And if not, so be it.
When we have the new thing defined, it is a trivial effort (probably
even self-review) to deprecate the old as "superceded by the new".
---- Randy
>
> - Bart
>
>
>
>
> --
> Bart Smaalders Solaris Kernel Performance
> [email protected] http://blogs.sun.com/barts
> "You will contribute more with mercurial than with thunderbird."
_______________________________________________
pm-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pm-discuss
