On Wednesday 01 December 2004 12:22 pm, Phil Mattison wrote:
> I agree that "from scratch" is not user frieldly now, but it could be made
> just as turnkey as any of the major existing distros. The only difference
> would be that you end up with a minimal system that boots up in command
> line mode, and you install your favorite GUI distro as a second step.
I guess I still don't see the rationale behind this. Most distros out there
allow a non-X install, X w/Gnome, X w/KDE, or some other variant. Ultimately
the non-technical user will have someone else setup the box and the GUI will
be Linux in their mind until they use it and start to learn about open
source.
> The difference is
> that most people have no commitment in principal to Linux, and would rather
> pay for convenience than learn something complex just to save a few bucks.
> If you could get the kernel for free and get a will-written GUI front end
> relatively cheap, I think that could be a viable contender for the mass
> market. Right now KDE/Gnome are the weak elements in Linux for attracting
> the mass market.
Well OS X has the UNIX backend with a well-written GUI frontend .. And for
certain people, it is viable, though, oddly enough, the people it has
attracted were already UNIX heads and overall, it hasn't made inroads into
other markets that Apple didn't already have strength in.
I don't quite understand why you think KDE/Gnome is the weak elements. KDE has
been quite strong over the past year or two in polishing its interface,
getting certain core apps (openoffice.org, gecko, etc..) running with a
native KDE look, adding additional GUI interfaces for common tasks,
reorganizing many of the defaults to provide a simplified interface, etc.. I
have been using it on a daily basis for over a year and a half and have had
very few issues.Granted, I am one of those guys who uses the command line a
LOT so perhaps I don't push things like konqueror quite hard enough. Though I
do use many of the core apps a lot and they work quite well.
What type of desktop environment do you envision that would provide a "viable
contender for the mass market" that KDE/Gnome doesn't provide? Right now, if
someone was to start on one of these interfaces, it would probably take
atleast 3 years to get something equivilent to the current interfaces, then a
few years on top of that to polish (ie take a look at OS X's development).
Seems like we are better off working with the already quite solid foundation
that the existing desktop environments provide? I'm truly interested in
hearing your perspective on this.
> I wouldn't argue with that. But the fact remains that many of the "Windows
> surrogates" for Linux are not very well written. It is precicely that
> philosophy of always building on the existing foundation that makes many of
> those applications into kludges. Very often it also results in poor
> performace and high latency in the UI. I still say that without a practical
> commercial motivation that effectively addresses the needs of the mass
> market, Linux will never be a viable competitor to Windows. The experiences
> of guys like you and me are not relevant because we actually *like* the
> technical aspect.
From what you have written, I get the clear feeling that these are your
positions:
1. Linux kernel & shell are a-ok, no problems.
2. X based desktop environments (kde/gnome/etc) too slow, buggy, crappy.
3. Free alternatives to commercial products are not good.
4. Commercial motivation = good products
Lets start from the bottom and go up..
#4 -- commercial motivation = good products. Ok lets look at the market leader
-- Microsoft. They had the most commercial success and they have gone on
record as claiming to "give the customer what they want.." -- but lets see,
we have a system without clear metaphors, complex APIs, massive security
issues and all around kludge.. And this is for a system that was initially
developed just over a decade ago (WinNT).
Even look at Apple .. they gave the end user ease of use. Didn't care about
many of the core fundamental principles found in UNIX (multitasking,
security, protected memory, etc..). After trying to hack these features on
their old system, they realized it was not viable and did the right thing and
developed a new system based on UNIX.
Seems to me that commercial motivation leads to focusing on features that will
push product and meet short-term quotas. Doesnt' seem like a good idea.
#3 -- You have a few issues here. The biggest is the simple fact that a free
open source general computing desktop hasn't been around very long. So it has
been a game of catch up.. Without large user bases providing feedback and man
power, its slow going. However, as more use these tools on a daily basis
(myself included) more input will occur and these programs will get
significantly better. I have seen a dramatic increase in quality over the
past year and expect this trend to continue.
#2 -- This one seems to be getting better. Develop is strong with KDE & Gnome
with both expanding their development teams to include great artists,
usability analysis, documentation writers, more developers, etc..etc.. With
more contributors, more users and corporate backing (ie Novell) providing
funding and additional push toward "end user friendly", its looking good
(well atleast to me) .. and about that X thing, well we have X.org now which
seems to be doing a great job of getting X back into active development and
meeting the current needs of users.
#1 -- You like it, I like it.. Seems like a lot of companies are embracing it
and liking it as well. The open source development model has proven itself to
be successful. I think it will not be too much longer when we start to see
even more adoption of GUI open source and it becoming the first viable
alternative to so many who have been locked into Windows. The question ends
up being, are they willing to learn something new? When you get right down to
it, many of these people use many Microsoft products and I don't foresee
Microsoft porting their apps, so there will be a learning curve. No other GUI
on-top-of Linux, commercial or free, is going to solve that.
---------------------------------------------------
PLUG-discuss mailing list -
PLUG-discuss@lists.plug.phoenix.az.us
To subscribe, unsubscribe, or to change you mail settings:
http://lists.PLUG.phoenix.az.us/mailman/listinfo/plug-discuss
(text/plain)