Robert Donovan wrote:
I do wish we could all agree an a standard GUI, features, tools, attributes,
and a standard way to extend it to accommodate individual taste.
I don't simply because it won't be *my* standard. That's why I use Linux
to begin with. If there is something I don't like with some aspect of my
computer use, I usually can find something more to my liking.
Selecting one GUI standard also presumes there is one use standard or
one hardware platform standard, or any other number of "Standards".
I don't want Linux to become the next Windows or the next OSX or
whatever. I want it to stay nimble and flexible and to easily adapt to
the needs of the individual rather than try to create a mass market of
lazy drones.
This does
not mean that the nonstandard stuff has to be discarded, just an aggreement
on a common base to start from and an efficient way to integrate new stuff.
And which criteria decides which stuff is to be non-standard? The One
True Way crowd already have several choice OSes I don't care for. I
don't really need another poor solution. "Standardizing" the Linux GUI
will simply spawn Linux's Linux in the same way MS OSes encouraged the
use of Linux itself.
In any case, any judgment of what makes a good user interface is
subjective at best. "There's no accounting for taste". I like these
colors, he doesn't. She likes cascading menus, I don't. They like CLI,
we don't. On and on. If you can somehow guarantee that, after having
achieved an industry-agreed GUI standard for Linux, that _no_one_ will
provide at least one substantial alternative GUI to satisfy those who
don't like The Standard, then you'll have your way. Otherwise, forget
it. As long as Linux is not controlled by one entity, there will be no
mono-culture of Linux standards. I mean, really!, we can't even agree on
file systems!
I realize that this means nobody will be entirely pleased with the result, but
too
many degrees of freedom can have the effect of diluting effort and causing a
lack of focus. To the degree that it hurts efforts to get Linux on the desktop,
I think a bit of constraint might not be a bad idea.
RD
That would sound like perfect sense if it came from Microsoft or Apple.
Of course, if such thinking takes over the Linux world, Linux will cease
to be Linuxy, and those who prefer the somewhat chaotic, but quick and
nimble way of Linux-style development will move on to whatever looks
more like the old Linux (or Unixy or BSDixy or...) than like the New
Shiny Polished Linux.
One of the things I like about the managed chaos of the FOSS software
experience is that it recognizes that life is too short and too mutable
for one to stay in one place long enough to produce perfection. Once
you've accomplished that, you can be pretty sure you've also achieved
obsolescence.
The tree of human development produces new wants and needs way faster
than the stasis that is "Perfect" technology can satisfy. As soon as the
perfect solution arrives, the need that begged it has long become
anachronistic.
The driving motive for Linuxy development is often likened to scratching
an itch. Itches eventually go away. If they're chronic, it probably
means the problem's also pathological. All warts and blemishes are not
pathological and so don't need to be "fixed" by developing a perfect
cure. "Better is the enemy of Good Enough". Linux is thriving because it
only needs to be good enough. Then it moves on.
When I finally settle down and wish only for stability, it will be the
day I realize how really old I've become. For now, I prefer a technology
environment closer to Niven's and Pournelle's Watchmakers, where a
solution to a problem is only useful as long as the problem it solves is
important or a better solution comes along, and it's abandoned when
that's no longer true - perfected or not.
--
Best Regards,
~DJA.
--
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list
