On Mar 15, 2006, at 1:07 AM, [EMAIL PROTECTED] wrote:
We already have secure properties defined, and there are a set of
*-secprop interfaces to deal with them. We could follow the same
convention for other types of objects.
But secprop have different operations you're proposing are created/
delete/show, vs the other properties which support only set/show.
A user already has to understand that the dladm subcommands act on
specific types of objects. Once this is understood, subcommands of
the
form setprop-link seem a "natural" fit.
I don't find it quite so natural. In particular, all of the existing
subcommands (and most proposed subcommands) have a single verb
acting on a
single object. Here, we have a single verb acting on an object (prop)
that is associated with another object (link).
The way I see it fit the current dladm(1M) model is that the object
is the link, and the verb describes the operation of setting or
showing a property of that object. I think that's the model used by
CLIP as well, that's why I'm proposing the issue be taken to UIRB too
(which doesn't prevent the community from discussing this issue on
this forum of course). BTW, there should be more visibility into UIRB
and CLIP from OpenSolaris. I'm a kernel guy and don't claim to own
our UI guidelines :-)
--
Nicolas Droux, Solaris Networking
[EMAIL PROTECTED], http://blogs.sun.com/droux
_______________________________________________
networking-discuss mailing list
[email protected]
