[OT] Blowing my mind about Linux

Peter Alcibiades palcibiades-first at yahoo.co.uk
Sun May 13 04:46:02 EDT 2012


There are really two interesting issues, one being the effect of choice on
non-computer people who are using the systems, and the other being usability
and standards.

Richmond is probably right that choice is to a degree unwelcome for most
people.  We may like it, we may know what we want, but in a way for the
ordinary user its like being confronted with a request to by a bicycle by
assembling a collection of parts.  People simply do not want to learn enough
about computers or operating systems to do this, and they find the existence
of 350 different Linux distributions, if they ever find out about it, as
utterly bewildering and a bit threatening.

Its probably what accounts in part for Ubuntu's success - if you manage to
think of one distribution as being in some sense Linux, its easier to
handle.

Richard is right about the advantages of choice however.  The same thing
that underlies the choice gives rise to very rapid evolution.  Its not just
about people being able to choose, its also about people being able to
customize, and in the end that is the unique advantage of Linux.  Wherever
there is a niche that it can be made to fit, someone will do it.  This is
the really remarkable and unique thing from a developer's perspective.  You
really can customize the whole works.  A kiosk application, for instance,
can be set up in an hour or so using stripped down Debian and pessulus.  You
can have appliances going with one or two appls only, similarly easily.

The other question is about usability and standards, and here I differ from
Richard and think we are going through a bad patch, and for basically bad
though understandable reasons.  My own experience is that Gnome2 worked just
about perfectly for my users.  I set it up with one bottom task bar, a
virtual desktop manager, clock,  shutdown button, and whatever the few
frequently used applications are.  Mail, web, office, a few others.  I set
up the file manager in browser mode, and put a terminal icon in the task bar
for the very rare occasions when people call and we need it.  You can do
exactly this with xfce now.  Once done, people get used to how it works and
it vanishes for them.  Its just their computer.

The thing I always have to spend time on is how to take advantage of the
virtual desktops, and that's really an indicator.  When you find people
leaving their work in place, moving to another one to get another task done,
and avoiding all that mess of overlapping windows, you know they are going
to be OK.  

You can imagine that with this experience the idea that the latest
incarnations of either Gnome or KDE will increase user satisfaction seems
absurd, and so I have spent a bit of time finding how to set up xfce to
duplicate what they now have, and it will be fine.  At the same time if you
are doing that, a light DM is logical, and the one I am now using is
lightdm.  See, there was nothing that needed any improvement as far as the
users were concerned.  We had got it just right.  They are OK for the
applications to evolve, though the latest evolutions of Kontact have just
about destroyed it, but there was nothing to be improved in their desktop
interface.

I think what has happened to both the gnome and the kde teams is they are
fixing things that worked.   Kontact is the worst example of this, but its
the whole idea that there was any need to change the user experience of the
OS.   Its understandable, people always want to improve, and the thrill
comes from radical innovation, but sometimes you have got it just right and
need to stop or to evolve very carefully and incrementally.  We need to
listen to the users far more.  In the end, if its not improving usability
for them, its not improving it, no matter what the academic studies say.

Peter

--
View this message in context: http://runtime-revolution.278305.n4.nabble.com/OT-Blowing-my-mind-about-Linux-tp4627046p4630172.html
Sent from the Revolution - User mailing list archive at Nabble.com.




More information about the use-livecode mailing list