Re: [Usability] Screensaver and idle time

[William Jon McCann <mccann jhu edu>]

Hi Joachim and Matthew,

Matthew Paul Thomas wrote:
> On Mar 8, 2006, at 8:01 AM, Joachim Noreiko wrote:
>
> > ...
> > I have been giving this some more thought.
> >
> > For the idle time label, how about one of:
> >
> > "Treat the computer as idle after:" * minutes
> > "Treat the computer as idle after this length of
> > time."
> > "Treat the computer as idle after this length of time
> > without user input."
> > ...

I'm not sure I like the term "treat" perhaps "regard" (in the sense of 
deem to be) is better.


> That's still a level of indirection. "Treat the computer as idle"? Does 
> this mean the computer starts criticizing its own lethargy? :-)
>
> Perhaps this should all be defined with reference to the screensaver, 
> since it's the most visible thing. So the screensaver preferences 
> themselves can have:
>     [/] Turn on the screensaver when the computer is inactive for: ...
>
> Then either the Screensaver preferences or some future Security 
> preferences can have:
>     [/] Require password when returning from screensaver
>
> And your IM client can have:
>     Mark as [Away   :^] automatically:
>     [/] When the scrensaver turns on
>     [/] When inactive for: ...

Interesting.  I suppose it depends on how you look at it or perhaps what 
kind of computer user you are.

Many of us are lucky enough to have at least one working hand.  Perhaps 
consequently we totally love our switches, knobs, levers, etc.  For 
example, I have a stick shifting manual transmission in my car.  We'll 
always have some type of control in hand.  However, what usually happens 
is that over time discrete low level controls are replaced by more 
broadly scoped high level (for the time) controls.  Correspondingly, the 
details and visibility of the lower level controls recede into the 
background.  We'll likely end up driving with computers in our hands 
before too long.

I'm not a user interface expert so I'm not sure what the state of the 
art is with regard to using abstractions in an interface.  I don't see 
how we can avoid it though if we are to move forward.

When I left my house this morning I had to do a number of really boring 
things like closing the blinds, turning off appliances, and locking 
doors.  I previously programmed my thermostat to my regular schedule so 
that I wouldn't have to turn it down manually.  There are two primary 
reasons for all this stuff: power saving/management, safety/security. 
And of course there is the internal contradiction - it is hard to do it 
all at the same time - which is why people leave lights on when they go 
away.

However, might it be nice if I could lock my house instead of locking my 
door?  Or might my house be able to detect when I'm not there so that I 
don't waste energy running the furnace on a timetable?

Some people already have things like this.  Though you probably need 
infrared motion detectors to do it so your lights won't go off when you 
are sitting still reading a book.  And maybe you can use some kind of 
accounting how many bodies enter and exit a room.  And maybe you don't 
want the lights to turn on when you walk into a room while someone is 
watching a movie.

So, what the heck does this have to do with screensavers?

The one thing that I must tell my hypothetical smart home is how long 
should it wait to regard itself as unoccupied.  It doesn't make sense to 
define this only with reference to the window blinds - even though they 
are more visible than the state of occupation.

I expect there would be some kind of configuration like:

Regard the home as unoccupied after [30 minutes] of inactivity
[x] Close doors, windows, and blinds when unoccupied
[x] Lock the doors and window when unoccupied


Actually is is a little bit more complicated than this because of 
multiple sessions and other details.


I hope this metaphor doesn't confuse things even more.  To summarize, I 
don't think people should be controlling how their screensaver works - 
but rather how their desktop/computer works.  I don't think we'll have 
succeeded until the screensaver cannot be identified as a discrete 
component.

Jon