Re: Clueless users are bad for debian (was Let's CENSOR it!)
On Fri, Mar 26, 1999 at 01:10:31AM +0000, Kevin Dalley wrote:
> "Pablo \(Tombstone\) Averbuj" <email@example.com> writes:
> > Executive Summary
> > =================
> > 1. Stupid Users are Bad.
> > 2. Stupid Users are Bad for Debian.
> > therefore:
> > 3. Stupid Users should be ignored.
> Reading "The Design of Everyday Things", by Norman changed my mind on
> this issue. I dislike struggling for 5 minutes to determine how to
> set the time on a phone. I dislike struggling for an hour to install
> a Debian package in which I *might* be interested.
i must be missing something here.
what is it about "apt-get install <PACKAGE...>" that takes an hour to type?
> A user should not need to be an expert on the entire Debian system in
> order to install it.
no, but there is a bare minimum of knowledge about computers which is
necessary for operating them - regardless of what software or operating
system is installed. basic computer literacy.
i've done a lot of help-desk/support type work over the last 20 years
(not much in the last 5 years...these days i say "I don't do help desk"
at job interviews), and the problem is NOT that computers are too hard.
the problem is that a large percentage of people who have to use
computers for their job are too stupid, or too ignorant, or too lazy or
all three. many of them actively resent the computer and do not want
to know how to operate it - they are GLAD when it screws up because it
confirms their view that computers are difficult, unreliable, and a
waste of time.
I used to believe that it was a generational thing and that the
problem would vanish as more people who had grown with computers
entered the workforce. if anything, the problem has got worse: about
the same percentage of them are completely useless with any form of
computer...the difference is that they have the mistaken belief that
they know what they are doing and that a pretty GUI is an adequate
substitute for knowledge and/or intelligence.
> As far as possible, a novice should be able to use a system.
"as far as possible". there's the crunch. the fact is that a novice is
not able to use a system very far. in some areas, novices should not be
*allowed* to operate the system (e.g. system administration, network
design and administration).
this should be obvious, but it isn't: how the hell can someone DESIGN a
network when they don't have the faintest idea of what a network is or
how it works, or what a protocol is, or how routing works etc etc etc
(the list goes on...)
why isn't it obvious? as far as i can tell, it's because most of the
industry has been suckered into believing some absurd things, and Joe
Public (not knowing any better) accepts what is published in the trade
press as an established fact.
the first absurd thing is mentioned above: that a program CAN be an
adequate substitute for a knowledgeable and intelligent mind. This may
become true one day in the far future with high speed AI systems running
on neural network computers, but that probably wont be in my lifetime
(and i'm expecting at least another 50 years...or a lot more, depending
on how good medical tech gets in the next few decades). computers
are dumb. people are smart (or should be). dumb jobs need only dumb
computers. difficult jobs need real brains. use the right tool for the
the second absurd thing is "easy to use is the same thing as easy to
learn". this is simply not true. it is not impossible for a program to
be both easy to use AND easy to learn, but most "user friendly" programs
focus on being easy to learn at the expense of being easy to use.
the third absurdity is "graphical programs are inherently easier to
use". this is easily disproven...there are many graphical programs that
are clumsy and difficult to use and many non-graphical programs that are
easy. for example, compare the Award BIOS with the AMI Graphical BIOS.
The Award BIOS is much easier to use and much easier to learn. it's also
more effective at getting the job done.
the fourth major absurdity is that ease-of-use is an absolute, that it
can be objectively defined. this is nonsense - what is easy for one
person may be difficult for another, and vice-versa. e.g. selecting
some text with a mouse may be easy for Fred Bloggs, but tedious and
difficult for Joe Blow who has all the vi keystrokes "hardwired into his
there are other absurdities, but these are the main ones. if you think
about these and trace through the ramifications, then they go a long way
towards explaining the sorry state of the computer industry.
> Yet anyone who is an expert in an area should be able to control that
> area of expertise. These are difficult requirements to combine. The
> expert should never be left out, but the novice shouldn't be ignored
some things are not suitable for novices. they shouldn't be ignored,
they should be directed to the documentation and other sources of
some of them will learn enough to advance beyond 'novice' and actually
become capable of operating complex systems.
some will not. there's not a lot we can do about that and we shouldn't
be too concerned, either: not everyone has the skills or aptitude for
being a systems admin or programmer or computer operator or user, just
as not everyone is capable of creating decent music or doing well at
some sport or writing a novel or making a business deal, etc.
what is the point of this message? that debian shouldn't be aiming
at the lowest common denominator, we target a more knowledgeable
("sophisticated" if you like) audience than some other dists.
complex systems can only be simplified so far...they can't be simplified
to the point that ANY idiot can operate them, but they can often
be simplified to the point where any reasonably intelligent and
knowledgeable person can.
we strike a good (but not perfect) balance between ease of use and
this is not to say that we can't or shouldn't do better. we can and
we should...but it's a matter of priorities and goals.
overall, i think we're heading in the right direction, there's no need
to change....just aim for more of the same, only better.