No, people will almost always be the largest cost in doing business - however the cost for IT and support personnel will always be there, leaving the question: How can one reduce IT costs by millions per year (In a mid-to-large sized company?) The quick, correct, AND long term answer is: choose open source solutions whenever possible. The expertise you need is largely the same, the salaries are largely the same (Note: I am not referring to paper MCSEs here, I'm referring to real network engineers/sysadmins with half a clue), so that cost carries over with EITHER solution. When you start counting downtime, er, "scheduled maintenance windows" then the cost of operating under Windows is driven up even higher in comparison to Linux.

This might be true from a purely support perspective, but in reality useability and user-friendliness matters to productivity of others involved. Even if linux were easier to update, if it's more difficult to use for normal people you will end up paying more.

Believe me, I'm no supporter of proprietary software. Where I work they do use only MS software, but largely because it is what comes on the machines. I can't imagine that HP would charge less to lease these computers with linux on them instead, because that would entail higher support costs for them. Why? Because linux issues are generally more difficult to resolve for regular users and IT staff. Sure, I admit that the level that "have a clue" in windows and in linux are probably equally salaried. But when it comes to the general populace/users the percentage who even have an awareness of linux in most places is pretty low.

At my interview, for example, I just mentioned that I use linux, and I browse with firefox, etc. instead of IE. They didn't even understand "linux" but when I said I don't use IE, they were surprised. They said something along the lines of "we don't know anyone who uses anything other than Microsoft software."

I'm sure this was primarily ignorance on their part, but this ignorance is pretty common.

Now, Linux is almost to the point of user-friendliness that in many cases it matches windows (some places even exceeds it). But the general linux user still has to know a lot more to get things working than the general windows user.

My family, all of whom I've convinced to at least try linux out by putting it on their computers, still comments on how much typing they see me doing in order to get things done. I try to explain what that typing is for, and explain the commands but in general this is just not immediately accessible to them the way that learning windows applications is.

And, frankly, I can't imagine trying to get quite a few things working without using console at all. Maybe there are some linux distros that allow this already, but they are probably more limited in other ways.

Linux still has a a way to go, IMO, but that doesn't mean that I don't want it to get there. That also doesn't mean I don't think it's close enough in some cases for businesses to use it. I just think that for many it is not ready yet.

xanas3712>

I sit novices down in front of Linux (KDE desktop), show them OpenOffice and Firefox and they're generally good to go. It is every bit as friendly (more so!) and easy to use as Windows.

I come from a mainframe background. I started IT in 1968, and we did things then that were standard. What I mean is that we organized applications to have known prefixes. The commercial name may be xyz-gazoo, but error messages and modules had to adhere to a given three character prefix.

For example, one application, called APLSV, had all modules of the form dqcapl001, dqcaplcalc, ...
messages for APLSV had to have a format...

"dqc?nnnn: Your system is up and running"

the ? in the above had a single letter, such as "I", "W", or "E". Thus dqcinnnn was informative message, dqcennnn was an error message, and dqcwnnnn was a warning message.

We could look up each message in the application and determine the module(s) that generated it, or the actions to take.
I don't see this with Linux. Instead, I see a free for all for messages, and that makes support difficult.

Can we not code applications names. It would prevent so many problems, should a developer unknowingly put his module into the search path ahead of a system one.

DMSCMD103I I agree!
R; T=00:00:00.305

Seriously, I think that we will very-quickly see the emergence of Linux into the desktop world, far more-so than it is there now, but it will occur not because it is Linux and not as a direct result (as in "preference") that it is open-source. Simultaneously, Linux will continue to smash down barriers in the server-world.

There's a big advantage in open-source, collaborative development of what has, in the past, been de rigeur a "this is closed, this is mine," perspective. One of the simple reasons for this is that it enables greater progress than any single company, working alone and funding itself alone, has been able to achieve. As the hardware platforms continue to diversify, software can no longer continue to be tied to it. Since closed, proprietary business models are "inherently tied to it" (because they cannot afford to hit all those targets at once), cooperation becomes an economic necessity .. as well as "a big win."

But software isn't "free," not by any means, so the question remains... how do the companies pay their people? How do the people who are doing the work make it their full-time job, which is also what the stakeholders ("customers") of the product do require? That's a problem that really hasn't been solved yet.

I think that we're at a point in time when MS-Windows is nearing the end of its twenty-year run, and Linux is about a third of the way along, and this is therefore probably the first "major turning point" in the PC industry as we (until now) know it. Open Source has definitely been a surprising evolution (in an industry known for "surprising evolutions"), but it has its visible problems too. Instead of crowing, "The Wicked Witch Is Dead! The Wicked Witch Is Dead!", we need to figure out what to do next.

Dont you just love Novell ad on that page....

G'day all,

I'm a school principal and have been with computers since CPM80 days. I have been using Linux (FC4) for about a year as my main home desktop (the rest of the crew use Ubuntu). Recently we needed to upgrade our student server at school; M$ licences were too expensive for us so we have gone with Red Hat. We have the Linux Server taking with the M$ Admin server through Samba. On my new work laptop I'm using Star Office; my purpose is to change the culture of the school from M$ and in the process save lost of money. Changing people's understanding that XP is the only operating system is the biggest problem I think. I fear however that the new "Longhorn" or what ever it is called now, is going to start the catch-up game again.

Regards
John

One real slick way to convert people who are hesitant to part with their friendly neighborhood Microsoft server OS is to bring them halfway to let them get their feet wet. Lots of open source applications are cross-platform these days. Give a Windows desktop user Firefox, Thunderbird, the GIMP, OpenOffice, and GAIM to play with. On a server, throw on Apache for Win32 and OpenLDAP. Later on, when/if they're impressed, mention that they could use these same exact apps under a Linux OS, and they would work virtually identically.

Yes, all our admin XP's are running Firefox and T/bird.Open office will go onto our classroom units when we are due for an upgrade. Should save us thousands.

Cheers
John

With real cases like the one you share with us John it's clear why they need the "Linux killers". They are running out of good reasons and arguments.

While it is true that we convinced linux users feel that linux is better, I sometimes think that we have a few stumbling blocks to overcome. One of these is the integration of what is called license software (mp3, wmv, etc). Whenever I install a linux system, and the user does not find these applications... then within the week, linux is out and XP is in.

We need the ability to have mp3, wmv, and all "free to end user", but licensed to Linux Companies, as being available via a simple install.

It took me quite a while to find the plugins for mp3, etc. Totem has never worked on my system and finally I was able to use XINE as the viewer for WMV's. Prior to XINE. I was persistent, but new end users want no less than what they will get with basic XP home edition.

One of the things that M$ learned early on, and that Linux distros seem yet to learn, is that end-users do not want to have to "think." And I don't mean that in a disparaging way... People want to choose what they have to think about, and for most other folks (unlike "you and me"), the computer ain't "it."

End-users want to do. To them, it's not a JPEG... it's a picture. Not an MP3... a song. And, "either this newfangled computer can show me pictures and play songs, just like XP can do, or it can't."

End-users do not understand XP, in "our" sense of the word. To them, the computer is the machine, and XP is simply part-of how the "machine" works. Either "the machine" works, or it does not work.

It only takes a day or three for people to learn a new system, and only a week or three for the details of what they did before to fade away... perhaps gratefully. But if something happens that jars that transition; if they run off the road and wind up in a ditch; then it doesn't take but a split-second for "the way things used to be" to be instantly remembered as Nirvana. And you ... are now the goat who screwed everything up when it was working perfectly before...

So the burden really is upon distro writers to create a smoothly integrated system that is "drop-dead easy." And as the old country song says, "it ain't no thinkin' thang..."

And this is any better on Windows? I should really collect all the screenshots I've taken at work of a blank error message with just a stop sign or some other icon with no text and post them on my web page.... (not that you don't have a very good point... but Windows isn't consistant with it's errors either...)

GE, Siemens, Phillips, and Westinghouse all use Linux for their healthcare applications. These applications run ultrasounds, CT scanners, MRI scanners, Cath equipment, life-support systems, lab equipment, IV pumps.

If Linux is used by these major healthcare equipment companies, which is able to sustain human life, hope I am never on a ventilator running Windows OS, when a stop error or buffer overflow error occurs!

As far as ease goes, i sat my wife(only knows xp) down a few years ago, gave her the cd's and told her to install. she was able to do a full install and even created a network connection on her own. linux(mandriva) configures everything on my laptop perfectly while windows doesnt even config my sound

I have tried to help others switch. some gladly jump while others say " I dont mind taking my computer in to be cleaned of spyware once a month" some people are happy with whats put infront of them because thats all they know. My small business will be making a big push this year in regards to making linux more "known" By 3rd quarter i would like to have the funds to have atleast one commercial on t.v

I remeber a while back i heard microsofts new release was supposed to have permissions and a new technology.....tabbed browsing! to bad mozilla beat them to it. If linux is so bad why is microsoft almost mimicking linux? I dont remember where i read this and if i am wrong i do apologize for being misinformed.