Joe Pluta wrote:
David Gibbs wrote:
From my perspective ... Linux still has a long ways to go to make
inroads onto the desktop.
There's a number of factors ...
1. Lack of first class driver support ... video, network, etc. which
ties directly into...
2. Lack of vendor support.
Such as? What standard office equipment isn't supported by Linux?
Video cards, network cards, wireless cards, etc.
1. Dell Latitude D630 with the Dell branded wireless card (Broadcomm
chipset) is *NOT* supported in Linux. You need to use the NDISWRAPPER
and the windows wireless driver in order to get it to work. I've tried
to get it to work, but it's not nearly as smooth and integrated as with
devices with first class drivers (Such as the Intel A/B/G wireless card
I picked up for just this reason).
2. Shuttle micro-atx system with builtin ATI video card ... ATI does
*NOT* provide drivers that support the X.org server provided in Fedora
9. The stock Radeon video drivers work reasonably with the video card,
but not nearly as well as the WinXP drivers provided by ATI do.
These are just two examples where lack of first class driver support and
vendor support are lacking in Linux.
3. Wide variety OS version and distros (which effects #2).
Picking a good version is important, agreed. But that's a one-time
cost, and if it can save you boatloads of cash in the long run via
licensing and hardware costs, it's a hurdle which any good IT manager
simply has to address.
It's not a one time cost ... it's an ongoing cost, because as releases
change, so do the internals ... and, in the case of drivers, if the
kernel changes sufficiently, it will cease to function with an updated
kernel.
4. (Relative) Lack of applications.
Similar to above, which business applications are lacking? Open Office,
Firefox and Thunderbird handle just about anything a large percentage of
users do.
All Microsoft apps, Client Access, WDSC/RDi, Photoshop, etc.
Yes, there are open source apps that currently have compatibility with
many of the apps I've mentioned ... but there's no assurance that that
compatibility will be maintained. I'm not saying that's GOOD, I just
think it's something that has to be considered.
5. Usability factors (having to edit config files, having to restart X
to apply changes, etc).
I don't see much of that with modern releases. I have yet to configure
my EEE PC manually. I have had to do an apt-get or two to get
applications, but that's something IT would do in a bigger shop. Where
Linux would REALLY make sense is in a thin client environment: really
cheap desktops that download your apps from a central server as needed.
I've had a few situations where manual adjustments to the config files
and/or removal of lock files are necessary with X application's.
See, I've already begun the switch. While I have a Windows laptop, I
also have a Linux laptop and frankly the Linux laptop is easier to use
for some things - and the bootup time is unbeatable compared to
Windows. The primary problem with the EEE is the miniscule screen, but
whaddya want for $350?
On my own Laptop, I run 4 or 5 applications primarily ... Thunderbird,
Firefox, Pidgin ... for which there is no issue wrt Linux support.
However, I also run Quicken and Photoshop Express ... for which there is
*NO* Linux support (and likely never will be). Running those apps in
Wine isn't really an option. And there isn't any open source
applications that are viable alternatives (at least that I've seen).
IMO, Linux was created by developers FOR developers ... and not
necessarily for end users.
When it comes down to it ... the question I ask myself for ANY
application or OS is this: Would I be comfortable setting it up for my
mom (whom Joe has met, FWIW). A Linux desktop is absolutely not
something my mom could cope with.
When Linux gets to the point where someone like my mom can cope with it
... then I would say it's absolutely arrived at the desktop. When I see
Linux getting closer to that point is when I'll start my conversion.
Just as a side note ... aside from the redevelopment effort, I really
don't see a justifiable reason for companies such as Intuit *NOT* to
port their application over to Java (with SWT, such as it is, for UI).
Then they can release Windows & Mac versions of their products that are
100% compatible with each other. PLUS they could make their products
available for other OS's ... Linux & BSD for example ... perhaps without
direct support.
'course, this is all JMHO.
david
midrange.com
