1i5t5.duncan at cox.net
Mon Nov 2 09:06:34 GMT 2009
Anne Wilson posted on Mon, 02 Nov 2009 08:06:02 +0000 as excerpted:
> On Sunday 01 November 2009 22:29:15 James Tyrer wrote:
>> You totally missed my point which is that I need to documentation to
>> configure my system.
> No configuration is required to make this work, so I assume that it is
> some change you have already made to your system that is stopping it
You guys are using the same words... but mean different things by them.
=:^( (FWIW, from my observation, fortunately or unfortunately, this is
the base problem in at least 2/3 of differing viewpoint discussions, too
bad there's no magical way to suddenly understand the base definitions
the other side is using, or that proportion of the arguments would pretty
much solve themselves!)
The system does indeed need to be properly configured for this to work.
On distributions of any size, there's at least one person tasked with
learning all this stuff in whatever depth is necessary (including reading
the sources to groke it, if necessary) to properly configure the system
to have it all "just work", which is why it appears to Anne that "No
configuration is required." But where that gets broken for whatever
reason, or where the system doesn't fit the norm so the standard
distribution config doesn't work, or for those building more from scratch
(not from a distribution, or from a distribution designed to be built
from scratch) and thus relying directly on the upstream packages and
whatever documentation they can find, this doesn't work the way it does
for what is hopefully the majority of normal distribution users.
The problem in this case is that there's no basic user level
documentation on the configuration available, and the configuration
itself is incredibly obtuse and "brittle", resulting in an all but
unconfigurable (from the viewpoint of a normal user) system that either
works or doesn't, with no user-level documented way to get from the
"broken" to the "working config" state.
As I said, KDE's exposure of the underlying hal config is actually
surprisingly good and as simple as I can imagine it being, but
unfortunately, that doesn't change the underlying fact that the base
functionality is not designed with user configuration in mind at all, so
the KDE abstraction of the config is necessarily rather more advanced
than the ordinary user should or can reasonably be expected to
understand. This one's certainly not KDE's fault at all, except in their
choice to go with the "automagic" technology in the first place, with no
fallback to more traditional methods as used on kde3, for instance, but
it does remain a fact that if the user finds himself with this element of
the system non-functional, it's all but impossible to get it back to a
functional state, without simply reinstalling. And that's the situation
MS has been rightly criticized for being in for years, and a step back
for Linux, which had to this point been know for at least being
configurable to functional, even if it didn't simply "automagically" work
for most people out of the box, as one must assume is the goal with the
changes that leave us at the current situation.
Duncan - List replies preferred. No HTML msgs.
"Every nonfree program has a lord, a master --
and if you use the program, he is your master." Richard Stallman
This message is from the kde mailing list.
Account management: https://mail.kde.org/mailman/listinfo/kde.
More info: http://www.kde.org/faq.html.
More information about the kde