Re-Examining the PC-as-Appliance Model
Jan 25, 2008, 23:30 (19 Talkback[s])
(Other stories by Brian Proffitt)
Re-Imagining Linux Platforms to Meet the Needs of Cloud Service Providers
By Brian Proffitt
There have been several instances where I have written something
along the lines that all users really need from their computers is
a toaster. Just plug it in, turn it on, and it works. No
configuration, no installing.
Applying that kind of model to a PC with any operating system
has thus far been an impossible goal to achieve. Even if a device
actually had all of the software a user needed, those needs could
change, and very likely some of that software would break. Or need
Macs come close to a toaster model, with Linux right there,
given the security and automated update features available.
Windows? Right, that's our laugh for this week's column. But
"close" is still a long ways away. Even the pre-loaded Linux
machines being sold out on the market now fall short of a
"plug-in-and-forget" operating scenario.
This week, however, I begin to wonder if such a goal is worthy
I raise this question, not because I think it would be
impossible for a Linux PC to achieve if developers really put their
mind to it, but because I wonder if being a toaster is a really
good idea. Maybe there always needs to be an ability to poke around
inside a PC's guts to fix or improve it.
What brought this up is the rather mundane problem we have had
in our family since Wednesday, when our dryer up and died. Now, I
sit here in the home office, waiting for a repairman to come out
and see what the deal is. They're late, naturally, and I'm sure
this will either be something simple I could have fixed if I knew
how (thus setting up resentment at paying for the service call) or
something that may warrant replacing the entire device.
It's a bit frustrating, not knowing what's wrong, but knowing I
will have to pay someone for a skill set I don't have to fix
whatever it is.
With this (albeit minor) crisis in mind, it has occurred to me
that perhaps there is merit in leaving the PC open to
configuration. I am not saying that every user will have the
ability to fix any problem, but at least they have a chance to.
That chance is higher than my chances to fix a dryer, even for a
For one thing, the wealth of howtos on the Internet is far
greater for computer problems than for appliances. While this may
be a less-than-subtle indicator of the quality of software
engineering versus appliance engineering, it still affords the
potential for someone to try to fix their PC. Another pro in the PC
fix-it column: you don't need special tools or a special skill set
in electrical systems to fix them (typically).
I should emphasize that I don't think there's a lot of
simplification to be done in software. Usability-wise, every
operating system I have seen has issues. Especially from the point
of view of "Joe User." But there needs to be, I think, some way to
open the hood and let users fix things if they can.
Make things easier, but don't make them wait for a