Re: Final remote(4) / remote(9) API proposal

[Robert Elz <kre%munnari.OZ.AU@localhost>]

    Date:        Fri, 10 Oct 2008 06:05:50 -0600
    From:        Herb Peyerl <hpeyerl%beer.org@localhost>
    Message-ID:  <AF8DF9C1-F6F2-4E05-893A-906B1AD26CB5%beer.org@localhost>

  | > I'm sure we can brainstorm something that we can all agree on ;-)
  | nirc?

Naming - the question everyone can have an opinion on without
anyone knowing anything at all...   So why should I remain silent?

I'd suggest that the API (and hence the name) should be independent of
any suggestion of transport technology (so should not involve IR) - there's
no reason a bluetooth remote, or a 802.15 remote, or anything else
(including 802.3, or RS232) should present a different interface to the
application, it is the function the application cares about, not the
signalling technique by which the user supplies that information to
the system.

Similarly, there's no reason to imply remote - a device like this might
be built into the system being controlled, what matters is its capability,
not its location.   So I'd also avoid anything that suggests "remote"
(in the name used for the API).

None of this affects particular device/driver names as used by the kernel
drivers, or to identify particular devices accessible through the API of
course, so a driver for IR Remote Devices can be named to indicate that,
and a bluetooth driver could be btrc or something, that's a different
question (kind of like the relationship between wscons and video cards,
or wsmouse and different kinds of pointer devices.)

Now, all the above is actually intended to be serious ... in a previous
version of this message (that never made it to the list, (or anywhere)
I followed with some pretty absurd (non-serious) suggestions of what
the name might be.  This time I decided to omit that, I'm pathetic
at naming things even when I'm just attempting to make jokes, so I
won't bother...

kre