2008: Linux's year on the Desktop will hurt M$, yay!

[RealBlackStuff]

http://blogs.zdnet.com/storage/?p=260&tag=nl.e540

[dkpw]

Here's hoping....

[jdhurst]

This one (this, whatever year, is the year of Linux) has been going on ad nausem. I get the impression from reading that there is no longer much hope for Linux winning any greater share. I maintain people should use whatever they want, and I have no axe to grind, but important business software is no closer (which is not close at all) to having a Linux version or a MAC version that the business software was a decade ago. The OS does not matter - it is what you run that matters. ... JDH

[lightweight]

JDH,

I agree with your general sentiment (Linux taking over the world has been predicted for years), but many figures have Linux growing from 2% of the Desktop market in 2006 to 7% 2007. This is significant growth.

The OS does not matter - it is what you run that matters.

Exactly, and as applications are abstracted from the desktop to server, and web applications and standards grow, the client's OS becomes less important. More important are standards. This bodes well for open source, as well as the small to enterprise company.

[jamesqf]

...as applications are abstracted from the desktop to server, and web applications and standards grow, the client's OS becomes less important. More important are standards. This bodes well for open source, as well as the small to enterprise company.

But unfortunately it does not bode well for me. I use Linux because it lets me work the way I want, instead of being trapped in someone's "intuitive" GUI applications (which I find impossible to use effectively). The problem with Linux getting more desktop share is that to do so, it's essentially had to become a clone of Windoze, with Gnome & KDE trying to look as much like it as they can without copyright violations.

[independent]

This has been going on for years.
Linux won't rule on the desktop in the next 10 years IMO. But I think there is a good chance one day $M will be based-on/running *nix.
GNU/Linux needs to consolidate a bit. It's my opinion it needs to slow down and strengthen (ie squash bugs) it's market share. However, it's the bazaar we are talking about and not the cathedral so what happens is what happens.
I've just about had it with my distro (gentoo) and will probably go to the competition (ubuntu). If it wasn't for audio I'd be going to FreeBSD. But that's another story.

[Troels]

I maintain people should use whatever they want, and I have no axe to grind, but important business software is no closer (which is not close at all) to having a Linux version or a MAC version that the business software was a decade ago. The OS does not matter - it is what you run that matters. ... JDH

Neither do I, but since Wine has seen so much development it seems to be a good way of getting most business software to work, although i have found no way of getting Orcad to run to this day - while everything else works great, including SOME games.
AMD has released and will release more register details of their ASICs as ATi never did.

[awong]

I maintain people should use whatever they want, and I have no axe to grind, but important business software is no closer (which is not close at all) to having a Linux version or a MAC version that the business software was a decade ago. The OS does not matter - it is what you run that matters. ... JDH

Neither do I, but since Wine has seen so much development it seems to be a good way of getting most business software to work, although i have found no way of getting Orcad to run to this day - while everything else works great, including SOME games.
AMD has released and will release more register details of their ASICs as ATi never did.

I have to use virtual box to run orcad, since wine wont work

[jdhurst]

<snip>
I have to use virtual box to run orcad, since wine wont work

I use VMware to run operating systems in a sandbox and to test things, but I hate running an emulator or virtual machine just to run basic business tools. ... JDH

[rmcder]

Man! This conversation takes me back (twenty?) years to similar conversations in the OS/2 community!

Realistically... Linux has enjoyed several advantages that OS/2 never had, and despite that, it STILL hasn't made significant inroads against Windows. I think Windows has simply reached critical mass and anything else, regardless of its merits, is relegated to being a bit player.

Like OS/2 users before them, Linux users are more likely to be knowledgeable about the inner workings of the os and the hardware than are Win users generally. As with OS/2 , therefore, Linux strikes me as being a niche market for high-functioning compuphiles.

The one real hope for those of us who would like to see a real alternative to Windows, imo, is if all the Linux supporters joined with the Mac supporters and pooled their efforts. Together, there would be enough support to take a real run at Windows, but that is obviously unlikely to happen.

Anyway, I'm having fun playing with ubuntu, just as I had fun playing with OS/2 lo those many years ago!

[jamesqf]

Realistically... Linux has enjoyed several advantages that OS/2 never had, and despite that, it STILL hasn't made significant inroads against Windows....

As with OS/2 , therefore, Linux strikes me as being a niche market for high-functioning compuphiles.

I think you're looking at this the wrong way, though. Linux is, AND SHOULD BE, a system for the high-functioning. Not just compuphiles, but for anyone who needs to get certain kinds of work done without interference from a user interface designed to support the lowest common denominator. That may be running servers in the data center, developing complex scientific applications (which I can test on my T60, and then run on a BlueGene supercomputer with just a recompile), or any of the many things with too small a market to interest M$. It's still useful & important work, and having to try to do it in a Windoze environment would make it much harder to do.

The point is that Linux (and other *nix variants) are professional tools. Change them so that they can compete with Windoze for the mass market, and you reduce their usefulness as tools.

[jdhurst]

<snip>
The point is that Linux (and other *nix variants) are professional tools. Change them so that they can compete with Windoze for the mass market, and you reduce their usefulness as tools.

The point made earlier is that the Linux community has failed to adapt Linux to critical business tools out there today. Make no mistake about my viewpoint - use whatever you wish. But commonplace, widely used stuff I that run all the time has no Linux variant and is not likely to for years to come. ... JDH

[lightweight]

The point is that Linux (and other *nix variants) are professional tools. Change them so that they can compete with Windoze for the mass market, and you reduce their usefulness as tools.

It's important to note building some desktop for the masses is not the same as changing Linux. No one is forcing anyone to use KDE and Gnome and stay within their GUI configuration. Respectfully, I think you're overreacting about folks focussing their distrubitions into viable and easily supported desktops.

[rmcder]

Well, what's the goal here? For us oldtimers, the goal was always to have enough support for OS/2 that we could find software on the shelves when we went into the store, could count on new titles for Win being duplicated for OS/2, and that drivers for new hardware would include OS/2 versions. Of course, that never happened.

Linux HAS software, much of it free, but the same problem that afflicted OS/2 continues today for Linux. You go into a store, and you find Win titles and a smattering of Mac titles... And that's it. It doesn't matter how good Linux is, just as it didn't matter how good OS/2 was.

Imo, it also doesn't matter if you get a fancy frontend that "just works" to draw in the masses. What's their incentive to use it? You don't sell the masses based on techie "advantages". It didn't do OS/2 a bit of good to be a multithreaded, object-oriented operating system, that could support multiple processors YEARS before Win did. Or, to draw another parallel, that most banks used it in preference to Win.

Thirty(?) years ago, it didn't matter that the TI-99 series was the only 16-bit processing computer, or that it had sprites - objects that operated autonomously once initiated and which included multiple screen priorities (could overlay one with a lesser priority), or that it had built-in speech synthesis capability, or that its programming allowed for each new subroutine/program to have its own set of registers, three of which were dedicated to passing parameters. The consumer doesn't CARE about that stuff; only techies do!

For someone like me, Linux is the new OS/2, I can go online and I don't really have to concern myself with viruses and other noxious problems, and there're some neat things I can play with. Yeah, I have to struggle with it from time to time, just as with OS/2 before it (and the TI-99 before THAT), but I'm ok with that.

[jamesqf]

The point made earlier is that the Linux community has failed to adapt Linux to critical business tools out there today.

But that is at least in part my point. The scientific/engineering/programming world mostly doesn't have any need or desire to use those "critical" business tools (and often objects when an oblivious management forces such on them - I get people wanting me to look at or send them Powerpoint presentations, fer gawdsakes . And on the other side, the Windoze world does not offer anywhere near the range of technical tools that's available in the Linux world, though the problem's not as bad because it's comparatively easy to port a *nix program to Windoze.

[doog]

I feel that Linux on the desktop has done well in the last couple of years.
And I feel in the last year Vista has helped to add to the number of people moving to Linux.

Will Linux ever catch up to M$....No but if M$ keeps putting out OS like this last one. With the prices they charge. Linux will make some more gains.

I thought this was a good read.
http://www.news.com/Five-reasons-not-to ... =nefd.lede

I do think one thing that does hurt Linux is, there are so many distro's to choose from. But then again I guess Linux is about freedom of choice.

I would be curious how much the percentage has gone up in the last couple years of people who now dual boot. Windows/Nix.

Lowes and Home depot use Linux as there OS to run things. "google it once"

[independent]

Agreed regarding the number of distros. It's a dog's breakfast. However, I think the thing missing is a lack of standardisation among distros rather than the abundance which is hurting GNU/Linux. If there were standards for packaging then maybe there would be a shot. However, I don't think $M has anything to worry about from GNU/Linux, but there has been a shift toward Apple, which is just as bad IMO. Bad in the sense that there is the same locking-in technologies slash thinking with Jobs' OS as with Gates'.

[ThinkPad]

Interesting news !

[ThinkPad]

Interesting news !

[wswartzendruber]

You guys want to see Linux take over the world? Here's where to start:

1. A SINGLE GUI toolkit. We have Qt, Gtk+, and even wxWidgets running on top of Gtk+. We need to pick one, and put all of our efforts behind it.

2. A SINGLE desktop environment. Pick either GNOME, KDE, XFCE, whatever.

3. NetworkManager, HAL, and DBUS need to keep growing.

4. Hardware vendor support (that's YOU, nVidia and VIA).

5. Consistent user interface standards.

Freedom of choice won't necessarily make you strong.

I've noticed over the years that a lot of our community efforts are spent on the great many "KDE vs. GNOME vs. XFCE4 vs. Your Mom" threads. If Linux is ever really going to take over in the desktop market, we need CONSISTENCY.

Check it out, dude A can sit in front of dude B's Windows computer and know exactly what the hell to do. Why? Because Windows is CONSISTENT.

Try going to some software site to download a Linux application. The only download that won't have special directions is the source code, and you have to build that yourself. Otherwise, you'll see something like, "Tested on Fedora Core blah, Ubuntu meh, check with your distro provider for more information."

Which is something else. We've got a billion different distros running around each with a different goal. Ubuntu aims to be easy, Gentoo aims to be elite, Arch is this, Red Hat's that, blah blah blah. And the thing is, each distro has a different way of doing things. Last time I check, Fedora put Apache's configuration file in /etc/httpd/httpd.conf, while Gentoo puts it at /etc/apache/httpd.conf. So this makes it [censored] near impossible to provide generic support (HOWTO's, etc...) for Linux users because each one has a different breed of funk going on.

I remember a few years back there was this thing called "United Linux" between SCO, SuSE, and I think Caldera. Whatever happened to that? Nothing? That MIGHT be because each distro is too busy running off on it's own f***ing program thinking their way of doing things is the best.

Listen people, the United States wouldn't be where it is today if the Founding Fathers would've said, "Yeah, we're all going to go do our own thing." Hell no, they stuck together, made compromises, and along with that, a kick-[censored] nation.

Richard Stallman, you have followers, UNITE THEM.

[Paul Unger]

I couldn't agree more (apart from the political bit at the end)! Linux's greatest "strength"--the variety of 'flavours'--is its greatest weakness. There is a bewildering array of configuration issues for the new user. You've got to learn Linux to use Linux, which is a dead-end catch-22, because you've got to use Linux to learn Linux. Unless you've got someone to teach you, you're left to wade through a bazillion pages of helpful folks telling you how to implement something on build X of distro Y. But if you're working with build P of distro Q, you're hooped. In the end, Ubuntu is for human beings who know Ubuntu, because it doesn't "just work" . . . you've got to know how to make it work.

I made the same point as wswartzendruber to a friend recently: if Linux want's to make it to the world's desktop, these brilliant minds (and they really are brilliant) should collaborate on making--at the very least--displays (current and past) and wireless (again, new and old) 'work'. Go ahead, use a 'safe mode' monitor for setup, but then autodetect the resident graphics chip and implement the drivers with a 'package manager'. But don't make me learn command line to get my display to work. We're not all as brilliant as the builders! Right now getting Linux to work is like walking into a garage with all the necessary parts for a Chevy Malibu, but no manual on how to put it together and only knowing friends with a Ford Focus, a Dodge Charger, and a Mazda MPV to compare with. Getting the engine and running gear fitted to the chassis might be possible for someone with automechanic experience, but wiring the tail-lights (let along the stereo) may prove impossible.

Having said that, there is room for different distros: maybe someone needs a '[censored] small' OS, and another wants a 'free' (in the Linux spirit of the word) distro. But most of us want something we can use and can help our friends to use. Despite the obvious benefits of Linux, I would never recommend it to my parents; they regularly seek my help in sorting out Windows issues (the latest being a drive that was stuck in PIO mode), but I could never do that with Linux. I'm not smart enough . . .

Streamline Linux and you've got a winner!

PS I have a dual-boot environment (XP/Ubuntu) on all my machines. I'm trying to learn, but it's a steep curve.

[wswartzendruber]

I couldn't agree more (apart from the political bit at the end)! Linux's greatest "strength"--the variety of 'flavours'--is its greatest weakness. There is a bewildering array of configuration issues for the new user. You've got to learn Linux to use Linux, which is a dead-end catch-22, because you've got to use Linux to learn Linux. Unless you've got someone to teach you, you're left to wade through a bazillion pages of helpful folks telling you how to implement something on build X of distro Y. But if you're working with build P of distro Q, you're hooped. In the end, Ubuntu is for human beings who know Ubuntu, because it doesn't "just work" . . . you've got to know how to make it work.

I made the same point as wswartzendruber to a friend recently: if Linux want's to make it to the world's desktop, these brilliant minds (and they really are brilliant) should collaborate on making--at the very least--displays (current and past) and wireless (again, new and old) 'work'. Go ahead, use a 'safe mode' monitor for setup, but then autodetect the resident graphics chip and implement the drivers with a 'package manager'. But don't make me learn command line to get my display to work. We're not all as brilliant as the builders! Right now getting Linux to work is like walking into a garage with all the necessary parts for a Chevy Malibu, but no manual on how to put it together and only knowing friends with a Ford Focus, a Dodge Charger, and a Mazda MPV to compare with. Getting the engine and running gear fitted to the chassis might be possible for someone with automechanic experience, but wiring the tail-lights (let along the stereo) may prove impossible.

Having said that, there is room for different distros: maybe someone needs a '[censored] small' OS, and another wants a 'free' (in the Linux spirit of the word) distro. But most of us want something we can use and can help our friends to use. Despite the obvious benefits of Linux, I would never recommend it to my parents; they regularly seek my help in sorting out Windows issues (the latest being a drive that was stuck in PIO mode), but I could never do that with Linux. I'm not smart enough . . .

Streamline Linux and you've got a winner!

PS I have a dual-boot environment (XP/Ubuntu) on all my machines. I'm trying to learn, but it's a steep curve.

I haven't used any recent versions of Ubuntu, but you're welcome to PM me with any problems, and if I have time, I'll see what I can do.

In a couple weeks though, my unit's packing up and leaving Iraq so I'll be offline for a bit.

[Paul Unger]

Thank you. I honestly appreciate the gesture. I think I have things working at the moment: I've enabled compiz on my T42 (ATI Radeon 9600), my two atheros cards (one [IBM a/b/g I] "just worked"; the other [IBM a/b/g/n] works with an ndiswrapper [?!]), and two Logitech mice (using btnx with VX Revolution and V450). But please don't ask how I did it--I don't think I could reproduce the steps. And that's my biggest problem: I find a piece of command line magic, use it, and promptly forget it b/c I have no idea what I've done (/i -t ^m sudo apt-config &*%$#! is &*%$#! to me . . . ) Maybe that's what they mean when they say, "It just works": "type these commands and quit whining. It works, don't it?!" With that, I'll quit whining and get back to work.

May your unit find it's way home peacefully and safely.

[rmcder]

May your unit find it's way home peacefully and safely.

Amen!

[ditkuss]

This one (this, whatever year, is the year of Linux) has been going on ad nausem. I get the impression from reading that there is no longer much hope for Linux winning any greater share. I maintain people should use whatever they want, and I have no axe to grind, but important business software is no closer (which is not close at all) to having a Linux version or a MAC version that the business software was a decade ago. The OS does not matter - it is what you run that matters. ... JDH

What important, common business software are you talking about? Just curious because you always mention this mysterious business software.

[tylerwylie]

This one (this, whatever year, is the year of Linux) has been going on ad nausem. I get the impression from reading that there is no longer much hope for Linux winning any greater share. I maintain people should use whatever they want, and I have no axe to grind, but important business software is no closer (which is not close at all) to having a Linux version or a MAC version that the business software was a decade ago. The OS does not matter - it is what you run that matters. ... JDH

What important, common business software are you talking about? Just curious because you always mention this mysterious business software.

In some cases it's I.E.

I just run Windows in VMWare Workstation for all my Windows needs. Now if only I would stop bouncing between distributions.

[jdhurst]

<snip>
What important, common business software are you talking about? Just curious because you always mention this mysterious business software.

Quickbooks does not work in Linux, Simply Accounting does not work in Linux, AccPac (so far as I know) does not work in Linux, UltraEdit does not work in Linux, Juniper Netscreen does not work in Linux. Then, for people like me who don't want substitutes, Microsoft Office does not work in Linux. You catch my drift - some stuff has no counterpart in Linux; some stuff I don't like the Linux alternative offered. ... JDH

[tylerwylie]

<snip>
What important, common business software are you talking about? Just curious because you always mention this mysterious business software.

Quickbooks does not work in Linux, Simply Accounting does not work in Linux, AccPac (so far as I know) does not work in Linux, UltraEdit does not work in Linux, Juniper Netscreen does not work in Linux. Then, for people like who don't want substitutes, Microsoft Office does not work in Linux. You catch my drift - some sutff has no counerpart in Linux, some stuuf I don't like the alternative offered. ... Jdh

OpenOffice does work in Linux i hear though.

[nilfilter]

Howdy thread,
Some have already made valid points as to why Linux won't make it to the masses anytime soon, and I mostly agree with them.

As stated before, it's not important whether Linux is or is not superior to competitors. It's applications commonly used in business, the OS doesn't really count. As an example: I am in the localisation business. I receive files from various companies that in turn received these files from their clients. We're all part of a production chain that heavily depends on certain applications and file formats, e.g. Trados. Document reviews are often done with Acrobat's comment tools.
Eventhough there are powerful translation tools available for Linux (e.g. Heartsome), I see about zilch chances for Linux to get into the business simply because the whole production chain depends on tools that are available for Windows only (and Mac, to a lesser degree though). I am positive this principle applies to other trades as well.

Cheers

[lightweight]

Quickbooks does not work in Linux, Simply Accounting does not work in Linux, AccPac (so far as I know) does not work in Linux, UltraEdit does not work in Linux, Juniper Netscreen does not work in Linux. Then, for people like me who don't want substitutes, Microsoft Office does not work in Linux. You catch my drift - some stuff has no counterpart in Linux; some stuff I don't like the Linux alternative offered. ... JDH

But dude...

http://www.linux.com/articles/41026
http://www.desktoplinux.com/articles/AT2282537026.html
http://www.sugarcrm.com/crm/ (Our company uses this integrated with Asterisk. Click on contacts to call? Coming integration with Zimbra/Mail? Pimp.)
And a text editor? Dude, vim, emacs, or if you like gui, Bluefish. Linux options are king.
You don't find secure, configurable, portable, and efficent VPN solutions under Linux? Really?

OS's do not matter, applications do. I get and agree with that. And if one is attached to certain programs for whatever reason, I get that too. But I do not understand how that is an argument against an open os, especially one where even beginning programers can write, adapt or do whatever they want with most software.

But regardless, CRM, databases, telecommunications, and certainly great text editors and secure tunneling is perhaps where open source and Linux is the most cost effective, efficient, and stable for the small to medium buisness, and it's best implemented past the desktop and into the server infastructure, lowering the sum of deployment and maintenence. Dell sells and supports Asterisk/Trixbox VoIP servers with SugarCRM on CentOS now and users access it all via web UI (on whatever compliant browser on whatever OS they want) for Pete's sake.