Linus on SCO, and the Desktop Being 10 Years Away

An anonymous reader writes "In this interview from last week's Linux.conf.au in Australia, Linus Torvalds talks about how the SCO lawsuit 'riled' him and led him to spend a week writing an application to archive his email, and how he think Linux will take 5 to 10 years to become mainstream on the desktop."

Comments

I agree

[PatrickThomson]

Linux on the desktop is a long long way off from being as easy to use for beginners as windows is. I think we need to just grit our teeth, clench our buttocks, swallow our pride and set out to emulate windows's simplicity.

Re:I agree

[Leroy_Brown242]

Emulating someone else is not the path to being a hero. As I have talked about before, we need tog et some fresh blood intothe design of a new GUI. OSX, Windows, windowmaker, KDE, all suck. Outofo them all, I like OSX and windowmaker the best, but they suck.

Linux/unix will be popular on the desktop, when the GUI is not designed by a geek.

Re:I agree

[Anonymous Coward]

Dont emulate it, better it.

Im tired of all this .blah hidden config files which has no standard formats except tabs. how about using something more structured. like XML

How about a DB file system to the core. A better windowing system, and a better coding architecture its just too archane these days.

This is the way to get support , drivers, apps etc.

Drop the ego sheild.

Re:I agree

[vpscolo]

At the end of the days users want something that works with their existing apps and documents. They don't care whats going on underneath as long as it works

Rus

Re:I agree

[Anonymous Coward]

There are many definitions of "The Desktop"

For many, it doesn't necessarily mean anything to do with beginners, or home users, or kiddie-eyecandy.

Personally, i see it as being a strength on the desktop in a business sense, where an organisation like IBM or Telstra or NTT has 50,000 workers all needing a desktop computer to easily email, browse, collaborate with users, plan their day, type documents, organise stuff etc.

For those users, the whole setup and install thing is irrelevant, and that's the hardest part at the moment. When it comes to actually using say, a good KDE install set up by a company for its own users, Linux is ready for the desktop in the middle of last year.

Re:I agree

[tomcrick]

I think with the lower end of Linus's statement (5 years), the use (and awareness) of Linux will become much more noticeable. I've noticed recently that the SCO lawsuit has made some waves in UK papers, where previously you'd be hard pushed to find a mention of Linux whenever a computer-related article is published (Microsoft, Microsoft, Microsoft!). Possibly something to do with the fact that the big name of IBM is involved, but surely this is a good thing - getting the Linux name actually recognised!

It's still amazing to see the puzzled look on people's faces when they ask what 'Red Hat Linux' is and when did Microsoft release it.....

Re:I agree

[CaptnMArk]

Windows simplicity? LOL

Re:I agree

[Durin_Deathless]

when the GUI is not designed by a geek
Yes!! I'm not alone!
True usability is defined(for me) as a machine that my Grandma can use. Not my geeky friends, but my parents and grandparents that aren't into computers.

Re:I agree

[caluml]

Im tired of all this .blah hidden config files which has no standard formats except tabs. how about using something more structured. like XML Why? No reason on earth. foo = bar. Why make it more complicated?

Re:I agree

[starseeker]

Well, yes and no. Two things to remember when considering how to get Windows users over to Linux:

1) most of them don't care what OS they're running
as long as it works

IP issues don't matter, freedom doesn't matter. What matters is things working, being straightforward, and being able to do what other users are doing. Computing is a social activity - people don't use them in isolation anymore. (Insert ironic geek social misfit comment here.) So falling down in any of these camps is enough to prevent people from switching.

2) Inertia is the most power force in the desktop
computer world.

Ordinary users Don't Like Change. If they take the time to relearn something, it has to be because it's so much better than what they have they can't live without it. That's a very rare condition. OSX is better than Windows, but not enough better that everyone is willing to abandon Windows. A few do, but inertia in computerland is a group effect, and as long as the group inertia is strong in one direction everyone goes that way. This is why Microsoft has a natural monopoly, much more so that telephones or power lines. Technology was able to find new ways to provide telephone service, and things like solar and wind power can generate power independant of power lines. But if people need to expend a lot of effort to learn a tool, THEY WILL NOT THROW AWAY THAT EFFORT. The software market, particularly the OS market, must face this. Change can occur, but very, very slowly. Which leads us to our first two guiding principles:

Taking over the World - Rule #1

Patience is not a virtue - it is a necessity

Taking over the World - Rule #2

There will never be a "Year of the Desktop"

Media and fans like explosive, dramatic changes. But that is not how things happen on a large scale. This is more like a river cutting through rock. So don't build up Linux as "about to take over the world/desktop/White House/whatever" because it won't be so dramatic. Particularly in light of

Taking over the World - Rule #3

"Desktop Ready" is not a well defined target,
and as such "making it" is like chasing a
mirage.

Each person has their own definition of ready for the desktop. Linux met mine years ago, and it's doubtful Windows could meet mine now. But I don't worry about what most users worry about - consistent look and feel aren't an issue for me. So who defines "ready"? For me, ready was a while back. But I'm clearly a geek. For my Dad, it might be close. For my Mom, I doubt it's close. It's a fuzzy thing.

With SCO making as much trouble as humanly possible for Linux and open source, and Microsoft lurking in the background, I know it's hard to remember this last rule. But do try, because it's the only reason we got as far as we have, and it's the only reason we'll go anywhere in the future.

Taking over the World - Rule #4 (The important one)

Have fun!

Re:I agree

[Anonymous Coward]

Don't kid yourself, a computer simple enough for granny to use will be useless to other people. Face it, the older generation got left out of the whole tech thing, and it's pointless catering to them or the other simpering idiots that find KDE or anything comparable 'difficult'.

The REAL problem, which is too late to fix, is the dumbass desktop scheme. How many people find Folders, Files and other abstract concepts apply well to computing? All you end up with is a user left scratching his/her head, saying 'now where the fuck did I put that file'. Add to that the logical yet difficult unix system tree, and you've got a mess on your hands.

I will admit, however, that having a /home directory each user has all to themselves is a big leap over Windows. And before the windows trolls flame me..wait I'm at zero, fuck it.

Re:I agree

[DoraLives]

emulate windows's simplicity.

True usability is defined(for me) as a machine that my Grandma can use

Which is what's endlessly hanging everybody up in the field of GUI design. They all want to be DIFFERENT from windows, but they fail to realize that windows isn't just decided upon by fiat, but instead is the result of endless focus groups and user surveys to determine exactly what grandma actually works most comfortably with! Microsoft has huge resources and can afford endless focus groups and user surveys to arrive at a smooth, intuitive GUI. We're not asking to emulate Microsoft, but instead emulate (or invent independently) a smooth user interface as defined by the users themselves. Which is apparently such a simple concept that nobody seems able to grok it.

Re:I agree

[westlake]

When you have 50,000 users at all skill levels and none of them geeks, designing a Linux desktop that "just works" for everyone is a difficult problem and by no means solved.

Re:I agree

[Simonetta]

You can be sure that when Linux actually overtakes Windows as the most popular desktop interface, the current Linux community will deride it as junk, not-Linux, 'Linux for Dummies', ect...

Part of the Linux/Unix fascination is a transformative mindset that confuses cryptic complexity with power and ease-of-use with restricted utility. It's part of the twisted mindset that exclaims without irony that using mispelled words like 'creat' for 'create' and 'ls' for 'list' and 'rm' for remove encourages flexiblility *it doesn't-it does just the opposite by forcing the memorization of a whole new set of words*. Anyone who in the twenty-first century who would say that typing something like:

>$ ls * rx -adlk- (*lf39(0309)) splatxy -3&**

makes programming more powerful just doesn't get the point of all the computer interface research that has been done since they were born.

This isn't a troll: it's a trueism. Linux/Unix must give up its 1970's mentality if it is ever going to be taken seriously by people who take computers seriously.

Thank you, (here I go again, another mod 0)

Re:I agree

[starseeker]

Uh - you don't "have" to do anything. Linux is like a takeover by democratic vote - no one forces you to use it unless you want to. Republicians don't get stamped out when Democrats win an election - they still exist. The challenge is to make Linux good enough that the logical decision is to use it instead of Windows, not somehow force users to remove Windows. Indeed, if Linux is strong enough it might compel Microsoft to respect common standards, and everyone would be better off. Taking over is defined as gaining enough momentum and user base so that people have to pay attention to making their stuff work with a product that everyone has complete access to, not leaving a smoking crater where every Windows box was.

The Linux War is a war to break a monopoly, and restore choice as a default situation. Restoring choice does NOT mean destroying Windows. But thanks to the position they've taken, it DOES seem to mean humbling Microsoft. Not end users mind you, but Microsoft. IBM used to be like Microsoft, but after they fell off the high peak they learned how to play nicer with the rest of the world. Maybe Microsoft can too.

You've been listening to Stallman too much. By all means, use Windows if you like it. No one is prying it out of your hands - if you ever let go it will be because YOU want to, not because anyone made you. Prying it out of your hands would be Microsoft's idea of victory, not ours. Open source doesn't say you have to use "our" OS in order to have what you need/want. It does say you should be able to make a choice - i.e., be presented with more than one option. Forcing OS choice is like converting people to a religion at gunpoint - insincere and meaningless. Anybody with guts will let their product fight on a level playing field. Open source wants the level playing field back, and that's what the war is about.

Re:I agree

[zulux]

that windows isn't just decided upon by fiat, but instead is the result of endless focus groups and user surveys to determine exactly what grandma actually works most comfortably with!

Focus groups suck for determining design.

Rememeber cars in the 80's that had a computerise voice to tell that "You're door is ajar"?

Focus groups *LOVED* the idea of the car taking to them - it really sucked in reality.

Just like Windows XP has that sucky search dog thing - it's cute the first time, buy annoying as hess lafter that.

Re:I agree

[LearnToSpell]

We're not asking to emulate Microsoft, but instead emulate (or invent independently) a smooth user interface as defined by the users themselves. Which is apparently such a simple concept that nobody seems able to grok it.

It may not be as simple as all that. Yesterday I went to my aunt's house to help her, since her AOL wasn't working right, and the three printed pages of instructions that tech support gave her may as well have been written in Phoenician.

She showed me how the startup was normal, but the "pictures" on the desktop seemed bigger, and when she fired up AOL, the background (default clouds) went "psychedelic." I took ten seconds to go into display settings, up her resolution, and bump the colours up from 256.

The point is, a smooth user interface is not some point to be reached where we can sit back and say "yes, this is it." These concepts of resolution and palettes are so nebulous that it can be very difficult for "normal" people (i.e. not /. readers) to describe properly what they're seeing, and how it's something different from what they encountered last night, when they last checked their email.

I think I understand what you're getting at, and I agree that there needs to be an evolution towards better user interfaces, but I'm not sure that a one-size-fits-all smooth, intuitive GUI can even exist. Some people like KDE, and turn everything on. Some people go straight to Blackbox. I use KMail; others swear by Mutt.

Until we get to a point where programs can ask "Do you want more options or fewer? Do you want clicky stuff, or do you consider a pointer to be the method for switching between xterms?" we'll continue to muddle along trying to balance the needs of the people who want to "do email" and those who want 3D overlapping alpha-transparent Everything, with sound.

There really are no simple answers. There are certainly better options than others, but determining what constitutes the perfect GUI is a pipe dream, because all users are different, and there are too (damn) many of them (grumble grumble).

Re:I agree

[damiam]

Windows doesn't "just work" for everyone either. That hasn't kept people from using it.

Re:I agree

[Bloater]

But XF86Config can be configured by all sorts of GUI utilities. I'll tell you what, you try to teach my dad about XML - then tell me that he will use Linux if he has to understand *that*.

On second thought, try teaching my dad how to use Notepad...

Re:I agree

[Anonymous Coward]

The difference between Unix's habit of sticking everything in /usr and Windows' habit of sticking everything in C:\Program Files is that the latter OS puts apps in their own directories.

That is, you have Foo installed in C:\Program Files\Foo, and Bar installed in C:\Program Files\Bar. Foo will be C:\Program Files\Foo\Foo.exe, with its documentation in C:\Program Files\Foo\Foo.hlp.

On Linux, foo would be installed in /usr, and bar would be installed in /usr. Foo would be /usr/bin/foo, and bar would be /usr/bin/bar. Foo's documentation would be stored as /usr/man/man1/foo.1 and /usr/info/foo.info; bar's documentation would be in /usr/man/man1/bar.1 and /usr/info/bar.info.

That's assuming they weren't installed into the identical directory structures based in /usr/local, /usr/share, and /usr/local/share. Or, as you point out, /opt. Or /usr/X11R6, indeed.

You may not consider it a problem that a Linux application has its files scattered over the hard disk, but people like your parent certainly like being able to uninstall an application just by deleting the directory it was installed in, instead of relying on an esoteric tool called a "package manager".

Of course, Windows _has_ a rather good package manager called MSI, but that's another story.

Re:I agree

[Bloater]

My dad hasn't been able to use a computer effectively since we replaced our old Amstrad 1512 complete with command line and text-mode, key-combination operated word processor with a more modern PC with windows and mouse.

He could remember the keys to press, but for the icons and GUI he must refer to the picture instructions I printed out. That means changing his glasses every 10 seconds. Imagine taking longer to do your work on a top-of-the-range PC than on an ancient rust-bucket :/

But hey, he thinks clippy is fun! :)

If Linux desktops mimic Windows, then not only will *I* find it harder to work effectively, so will my dad - For me, the argument is over. Especially that for the last 5 years or so, tech-support to my dad has been provided over the phone.

Re:I agree

[bangular]

I too think a desktop isn't necessarly defined by kiddie-eyecandy and all that stuff. I think the defination of desktop is a little skewed too. We have a few office suites, we have many many email programs, fast reliable browsers, Instant Messaging,Calendars, etc. etc. What's the problem? You'd think if you told most people that they could have a desktop that didn't crash every 5 minutes and set them up with 5 desktop icons of the stuff they use the most they'd be more than happy to use it. Nope! The problem is people are so stubborn. I think if you did the same for someone on Mac OSX very few would switch even though the GUI is far superior to Windows. I don't think it's about usability and all that stuff, I think it's people are just too stubborn to use anything else.

I think it's the fact people _think_ they know a lot about computers because they know a lot about windows. When they get onto anything else, they feel like an infant again and they hate it. They will spew out things like "it's too hard to use. It can't do this. It can't do that." etc. etc. You could take a windows user and put them on THE perfect operating system, and they probably wouldn't use it because it's different.

I've personally been using linux on the desktop almost exclusivly for about 4 years now. I can't stand to go back onto windows computers. They are just so slow and crash way too much and why deal with all that when I can use something superior.

I mean look at all the things Windows doesn't have. It doesn't even come with a ram disk driver!! And the one MS lets you download it utter crap. Let's see... no 64-bit chip support, no ability to run windows without a GUI (why should the whole OS crash if the GUI crashes), very weak ssl support (netbios stuff not encrypted), office can't export to PDF's, there's almost no basic scripting ability, updating your OS is a chore and much more complicated than emerge -U World, lack of a decent plain non formatted text editor. You get the picture. It's about perspective.

Re:I agree

[oconnorcjo]

but instead is the result of endless focus groups and user surveys to determine exactly what grandma actually works most comfortably with!

Actually I don't think that is how MS works. MS gui is an evolution of what they had for win3.1/95 and most gui changes have been mostly cosmetic or coppied from Mac. I personally think that the MS gui is a piece of shit but when people are asked how a gui should work, they say it should work like that of MS but ONLY because that is the way they learned and expect it to work. In essence because most people are too "short sighted" to imagine something better.

Re:I agree

[jeremyp]

Don't kid yourself, a computer simple enough for granny to use will be useless to other people

Patronising pile of shit. My mother is a grandmother and she has no problems using a computer. Any reasonably intelligent person of any age can learn to use a computer given a bit of time. I will not be migrating my mother to a Linux dektop however, because a) she is used to Windows, b) a lot of the software she uses is Windows only and while there may be equivalent Linux packages out there, she sees her computer as a tool not as a plaything. She is not going to be impressed with having to learn a whole new set of conventions and tools just because her son thinks Bill Gates is the spawn of satan.

Re:I agree

[Anonymous Coward]

The point is that I as a programmer and the normal people who use computers are human beings. I don't care if a DB backed file system is necessary, it's more convienient. XML has problems with datatypes and complexity, I don't care if its editable with Emacs.

POSIX was written by reverse engineering the BSD Unix in the 70s, its complicated and rooted to C. Look at how J2EE is taking off, it might not be as simple as I'd like, but at least you can plug a servlet into a server expecting it to work.

The day desktop apps come as little plug-in desklets of some sort (OCAML modules, ColorForth vocabs, whatever) cannot come too soon.

Some people love fiddling with weird bits of crap, the rest of us have LIVES!

Re:I agree

[jeremyp]

Nobody except a developer goes near the environment settings dialog. That's left over from WinNT. I expect they just forgot about it.

I really don't understand what they were thinking about with the new services box. As far as I can see "advanced" is only called that because the left hand edge is completely blank meaning you have to squeeze the useful info into less space.

I think the file sharing idea is to make it harder to do stupid things. In the default "simple" mode you have to move the files you want to share to a special folder. This contrasts with the old way where as soon as anybody discovered sharing they immediately shared the whole "C" drive read/write to anybody on the whole Internet.

Re:I agree

[ca1v1n]

That's funny, ours is trying desperately to get everyone running 2000 to upgrade to XP, because 2000 is such a security disaster.

Then again, our whole network is on publicly routable IPs, and most of it always will be for a variety of reasons.

Re:I agree

[chunkwhite86]

At the end of the days users want something that works with their existing apps and documents. They don't care whats going on underneath as long as it works

I agree, however it has to work for it's target audience. Most Linux distros are trying too hard to be all things to all people. They end up becoming what the Toyota Camry or Honda Accord is to the car market - a bland, boring, transportation appliance, which may be "good" for a great many things, it is not "excellent" at anything.

Another automotive analogy might be to compare your average slashdot Linux geek's computer to a Ferrari. Joe Geek has an overclocked Athlon and the latest -pre kernel compiled with optimization flags out the wazoo. The Ferrari is similarly tweaked to it's maximum potential, is designed to be screaming fast and handle like it's on rails - but unlike the Honda or Toyota, the Ferrari requires more maintenance, a skilled Ferrari specialist to work on it, and is more demanding of the driver. It also provides a much more rewarding experience than than Honda - this is what makes the Ferrari so desireable - but it's also what gives it such a limited market share (besides the price tag).

For Linux to be sucessful on the desktop, there needs to be a clear line between what is a corporate desktop distro, and what is a home desktop distro. This is exactly what RedHat is attempting with their new "Advanced Workstation" product, versus the Fedora Core.

Re:I agree

[cHALiTO]

As someone said before, I don't think usability can be thought in terms of "what grandma can use". I don't think such a thing is possible or even desirable (except for the companies wich only want to sell sell sell). You can make a friendly UI, easy to learn, but at some point you *have* to learn about some basic computer related concepts to understand what you're doing and what you can and cannot do with your PC. Granted you shouldn't need to read a 500-page manual, but at least some basic concepts (what parts make up a computer, what is an OS, an application, etc), it's not a matter of users caring about it, it's a matter of NEED, they need to know this, as they need to know that their car has an engine, four wheels, and it runs on gasoline and not alcohol (except if you live in brazil). Whatever tool you use, you ALWAYS have to learn at least the minimum about how it works in order to use it properly. Call me a zealot, but I don't agree with the MS-view of computers in which even a monkey can use a computer. Computers are NOT simple. They are NOT toasters. They are complex tools that require some knowledge to be used. If you want to use them, you should expect to sit down and read a manual or two, or take a course or have someone explain to you how it works. Otherwise, go buy a dvd/vcr/typewriter/hi-fi/etc.

This doesn't mean that overly complex UIs are ok either. Most people won't want that, no, but what i mean with all this is that maybe at some point it's not ease of use in the sense of how the UI looks or how many clicks you have to do to do something, but the coherence of it all. I think it's far more important to follow the same concepts everywhere in your system. For example, if you have to do something to configure your video card, you should look for something SIMILAR to configure your sound card.

oops.. gotta get back to work :)

Re:I agree

[Afrosheen]

Your mother isn't YOUR grandmother. I think that's an important distinction to make here as it determines her age and therefore her generation. I agree with AC here, if you dumb something down too much, it becomes useless to the average computer user. Just take a look at WebTV. My grandfather (no not my dad, my actual grandfather) does better with a pc than he ever did with WebTV. He isn't stupid, but he's technophobic to a certain extent.

I just wonder if making the computer more physical isn't better for older people. Everything is so virtual..I think if we gave them more buttons to push and meters, switches and gauges they'd like it better. You shouldn't have to click start > shut down > shut down to turn off your computer anyway, that's what the power switch is for. One press should shut it down on ANY pc.

The technological revolution took the physical and turned it all into the abstract. The world of computing is 75% software (and to the end user, much more). It's all theory and code and electrons, none of it is things you can touch and manipulate by hand. My grandfather's generation is one very much aligned with the physical; men who can fix cars and build homes or extensions to them. They're lost on a pc, just as most of us would be lost with a leaky pipe and a pipe wrench.

Re:I agree

[azzy]

Your 'parent' said that they couldn't handle folders.. they probably therefore don't know what 'gift', 'incoming', 'video', or 'porn' are in your example. So you shouldn't be so quick to agree with them.

Re:I agree

[gregmac]

Im tired of all this .blah hidden config files which has no standard formats except tabs. how about using something more structured. like XML

How about a DB file system to the core. A better windowing system, and a better coding architecture its just too archane these days.

So, instead of config files that are easy to modify with just a text editor, or some fairly simple shell scripts, we should switch to a database format that requires all sorts of client libraries to load and modify?

In, for example, a network environment, it can get difficult to automate tasks such as changing program settings. An example is the Outlook Express mail store root. I hate to write a program in delphi to do it, because microsoft decided to store it in a deep location using random unique identifiers as key names. (Though mozilla does the same thing with their directory structure, which annoys me...)

Anyway, config files that are easily changable with simple scripts are a huge benefit. Adding complexity is bad. XML is sort of a step in between. It could have its benefits, as long as there was a 'config file' DTD that meant there could be a standard editor to modify ANY config files. There could also be command-line tools to do the same.

I don't see any point in requiring some sort of database server, though, just for config files. Remember, not everyone uses linux with a GUI. A lot of times it's used in embedded devices with limited memory. Should we start having the text-config-file version of apache for those systems, and the XML or Database-aware version for systems that are bigger and have the database server? That would be a step in the wrong irection, for sure.

Re:I agree

[anthonyrcalgary]

Maybe most people just don't care. Learning something new when they can get along just fine with what they have is too much to ask. Seriously, the fan on my roommate's CPU died, and it's an Athlon so I told him he shouldn't use it until the fan is replaced. He just shrugged and did something else. That's how little some people care. They're not going to spend a weekend expanding their horizons because they (correctly) don't see any benefit. It's something they don't care about, so even a large improvement won't give them anything they want.

When it doesn't require any new skills (eg, when some OEM uses a distribution that's just as restrictive), then people will start using it, and they won't be much better off than they are now. There are deals like that now, but they haven't really hit the mass market.

There will always be people that don't care, and they will never have as much freedom or as much power as the rest of us. And they won't mind, because they don't care.

The real uphill battle Linux faces is no longer MS, it's user indifference. You can't use merit to sell someone on something they don't care about.

Re:I agree

[Rary]

I keep thinking we're trying to solve the wrong problem. Instead of making computers so simple that a complete moron can figure out how to use it instantly, teach the moron how to use it.

People don't get to drive a car until they learn how to drive a car. Using a computer is significantly more complicated than driving a car, yet we expect people to just sit down in front of it and start being productive. And when they're not, we blame the computer, or at least its designers. This is fundamentally wrong.

Modern computer operating systems, whether Windows, Linux, OSX, whatever, are all quite easy to use if you learn how. I think this problem will largely be solved over time as we have generations growing up in a computerized world. But as for your parents and grandparents, if they want to use a computer, teach them, don't expect the computer to be "click here and magic just occurs exactly the way you want it to", cuz it ain't gonna happen.

Re:How much is your time worth now-a-days

[kalidasa]

It was worth it to him. Me, I use Zoe; but then I also use an operating system someone else wrote. I'm not going to gainsay what Linus does with his time - I don't have an entire industry built around what I decided to do as a hobby.

Linus and the P2P Fileswapper victims of the RIAA

[cervo]

I've never seen a lawsuit up this close and personal before

This is what the "lucky" 300 must also be thinking. I don't think they will be spending their time writing an e-mail indexing program.

Linus is the only person I've ever heard of taking a lawsuit as an opportunity to write some new code. The world needs more Linuses!!!

the biggest barrier of all

[Anonymous Coward]

Lack of games. The odd FPS game crops up, but dual booting isnt an option for mot point and click users.

Re:the biggest barrier of all

[edwdig]

Games aren't a huge barrior. Particuarlly not FPS games, since those seem to be the type of game most likely to get ported to Linux. Most people play games on a console, not a PC. The most commonly used PC games are things like Solitare.

How many business require the use of games? If anything they'd be happier with an OS without many games. How many of you have parents that play Quake? My mom never played anything more than simple card games on the computer.

Really, the majority of the people who would care about the issue are the people who have nothing better to do than see how they can get an extra 1 frame per second out of Quake 3.

Games go where the market is. Not the other way around.

Re:the biggest barrier of all

[Laser Lou]

Lack of games. The odd FPS game crops up, but dual booting isnt an option for mot point and click users

This is a chicken-and-egg problem. Loki showed the world that games run fine in Linux. Once point-and-click users start using Linux, developers will port their games. The question is, what advantage does Linux offer them?

Re:the biggest barrier of all

[Tom]

Depends on your target audience.

Would Linux offer enough games for me now, if I were still in school or university? Definitely not, I was eating games for lunch at those times, and could hardly go a week without a new on.

Does Linux offer enough games for me today, where I work fulltime and have a bunch of other things to do as well (including my own game, see below) ?
Absolutely yes. In fact, I have quite a few Linux games on my shelf that I haven't played half as much as I'd like to. (Dominions 2, Terminus, Mindrover)

Re:the biggest barrier of all

[Anonymous Coward]

Most people play games on a console, not a PC.

There's an enormous amount of games which are only available on the PC. The PC gaming market is huge. Many "PC gamers" would like to play their games using Linux. I know because I'm one of them. There is a reason for sites like linuxgames.com and happypenguin.com.

Re:the biggest barrier of all

[stangbat]

What do you mean lack of games? Yahoo Games and PopCap seem to work fine on my Linux box. You mean there's something else?

All joking aside, to most people solitare and sites like these are computer (PC) gaming. At least that is the case for my wife and my mother. I'm sure they are not all that much different than many others.

Re:the biggest barrier of all

[PetoskeyGuy]

And the reason for Lack of games is lack of standard game development system. OpenGL is a good library, but to get a generic or high end 3D video card working smoothly on Linux is still something of a black art with drivers and X-Windows configurations from hell.

Redhat, Debian, Gentoo? Which distribution to support? What pacakage manager? The market is too small, and the support costs too high. I've worked on video games where we've had to evaluate these things.

Making a Bootable Linux Gaming CD was an option I've read before, but that just puts more of the setup and configuration steps into auto-detection where people can't get the full use of their hardware.

Linux is wonderful for porting apps with source, but porting binaries can be a pain compared with making a single windows EXE.

Re:Linux Desktop

[toddler99]

I've been linux as my primary desktop for 2 years. Its been working great for me. I write my school papers with abiword, my presentations with open office impress and do all my coding with vim. gnome hardly ever crashes on me and when it does i can typically do one of two things: either login remotely and restart X or cntrl+alt+backspace. Then i can file a bug report and in most cases the problem is solved. Linux just requires patience and an understanding of what and how you plan to use a tool. I think what linus means is it won't be ready for a generic user for a little while longer...

Re:Different interpretations?

[10Ghz]

There seems to be a lot of different interpretations of Linus' views of the future of Linux floating around. There was a recent post on /. entitled "Linus says 2004 is the year of the Linux Desktop" or something like that. That seems to be a bit of a conflict with this article.

Not at all. Basically, he thinks that in 2004 Linux will really take on in the desktop-market. But that wouldn't mean that Linux would be mainstream in the desktop-market. Let's assume that number of Linux-users doubles in 2004, and that's due to increase in desktop-use. That would give Linx a market-share of around 5%. If that happened, 2004 would be the "year of the desktop" for Linux, but being mainstream would still be several years in the future.

Re:It's all about the desktop journey

[cerberusss]

I think the key to the desktop is preloaded machines by big-vendor being available at retail stores

I think the key to the desktop is preloaded machines that can flawlessly interoperate with the existing Windows monopoly. If it would include the ability to run MS Office for instance (free CrossoverOffice included, or a better Wine), that would be good. That way, it would run most things that Windows can, and then some more.

Another interoperability issue would be internet-connection. The various ISPs should support Linux as well.

I'm sorry..

[Anonymous Coward]

But I can't just imagine there will be "another" Microsoft. Microsoft will never be completely destroyed, and linux will never completely take over; I think they will form a balance, and this is nothing but good for everyone; it creates competition, and drives innovation more than just having one behemoth in the playing field.

Witness the true age of information technology...

Re:Linux becoming commercial?

[10Ghz]

Hopefully it won't be all bad, but I'm worried that Linux will just turn into another Microsoft (obviously with open source, but still)...

Linux cannot become another Microsoft. Microsoft is about monopoly prices, lock-in, proprietary technologies etc. etc. None of those are possible with Linux. If Linux gained 100% market-share, there would still be several distros competing (and several free versions of Linux), the core-systems would be open and free, so moving between different vendors would be easy. And you could fork your own version from existing distros (for example Red Hat ==> Mandrake, Gentoo ==> Zynot)

You mentioned Red Hat trying to make a profit. How would that affect Linux? Easy: Red Hat would have even more money to spend improving Linux.

lol

[DrSkwid]

for some people 'archive' doesn't mean 'zip up into a binary format nothing else understands'

Linux on the desktop

[Anonymous Coward]

Didn't Linus say that he pretty much thought it would be a battle between Windows and Linux in 2005 and MacOS wouldn't be around ?

Looks like he was pretty far off. Then again Jobs did pull out a lot of white rabbits these last couple of years. I'm certainly not going back to Linux on the desktop any time soon. Not saying it isn't usable - au contraire - I'd rather use it than Windows because of my needs, but OS X just does everything so much better as long as you have enough juice for it.

Five to Ten

[DumbSwede]

I have used UNIX the majority of my computing career and LINUX for over five. But at work we are still struggling to get to rh9.0 with many systems at 7.2 and 8.0 though the are now considered depricated. We recently had to reconfigure a machine back to 7.1 to regenerate data for a client who is still using 7.1. Not only this, but our code is notoriously unstable if not running on the OS revision and patch level it was compiled on. I'm sure some will flame about the skill of our sysadmins and make script maintainers, but I think that would be unfair. We produce a lot of floating point intensive code that depends critically on the underlining OS calls, and while the code may run, it becomes quite a chore to justify to the customer (government) why the results may differ from earlier versions. This tendency for code to be brittle with compiler and OS upgrades is not something we observe under IRIX and SunOS, the two other platforms we support, and have supported for longer than LINUX.

I am not saying that SunOS or IRIX are superior, just that the upgrades come at a more manageable pace, and tend not to break our code base when upgrading compilers. I think the reason Linus thinks five to ten years before really conquering the desktop is based on two things. By then LINUX should have slowed down in its development and will be a beast you can run two to three years before upgrading. Secondly, Windows will probably sink under the weight of it is haphazard code base, which is guided not by what is best for users and cleanest in design, but what makes sense commercially to support and lock-in their other products in as covert way as possible to keep from running afoul of the antitrust laws.

Looking forward to the day though!

Re:It's all about the desktop journey

[po8]

For me, the journey has been more like 20 years. I was running a desktop window system on a UNIX-like OS at home before there was such a thing as X (Smalltalk on LynxOS on a Tektronix Pegasus box).

I have to say that I think the folks who are all over the deficiencies of the Linux Desktop, and how we have to emulate the Windows/Mac/BeOS/Xbox/Sinclair/whatever desktop experience to have a usable desktop are mistaken. I think they underestimate the ability of users to adapt, and overestimate the degree to which familiar = better. For many years I had a PC or Mac sitting on my desktop next to a UNIX/X box. Now I have a Windows box and a Linux box at home. I have always found that I almost exclusively use the UNIX/X box. The monopoly (at best duopoly) is real, and most folks haven't had my experience. I think it's clear that they're going to, and I think it's going to be enlightening for them when they do.

I'm working hard to make the Linux desktop experience better for everyone. But it's pretty darn good now. So good that I finally threw away twm a couple of years ago. :-)

Let's enjoy the ride.

Re:Linux becoming commercial?

[10Ghz]

But what if these distributors like Red Hat "develop" Linux to the extent that it outdoes every single other Linux distro on the market (because of the availability of capital, which seems to be the argument you're using - more money = better development?)? Then, it would seem, customers are effectively locked in to one particular system even though it is theoretically open and free for anyone to change (most users won't bother).

If one distro wins by simply being superior to everyone else, then I fail to see how that could be considered bad. And there would be nothing stopping you (or anyone else for that matter) from creating your own version of their distro (or creating one from scratch) and competing with them with your own version. It has happened several times in the past (like when Mandrake was created from Red Hat).

And having money DOES help developement. For example, Red Hat (or some other company) could hire full-time kernel-hackers that would have more time (and better equipment) at their disposal, instead of if they hacked only in their free time.

Re:It's all about the desktop journey

[PReDiToR]

I prefer to go the other way, whenever someone asks me to get them a pirated copy of Office, I encourage them to get OO.org instead, or at least try it for a week, and if they need more features, come back and see me.
Only then do I tell them that pirating is illegal and I refuse to partici *yawn* sorry...? er.. oh yeah participate in that sort of thing.

Nonsense

[RoLi]

Technically, KDE/Linux has been good enough for the desktop for 1 or 2 years already.

What is missing is applications (especially games) and to a lesser extent drivers.

The 3d-modelling niche is a very good example on how fast Linux can take over a market when the apps are there.

In the next years, expect other niches to go to Linux, the next being non-US government desktops. When Munich migrates and ports their apps, it gets easier, cheaper and faster for other cities with similar application-needs to follow.

The only problem is that such migrations take a lot of time, that's why it is taking a decade (and it already started).

Re:Nonsense

[RoLi]

Certainly not, but at least Windows' interface isn't designed by techies, and instead GUI specialists.

KDE's interface is much better. Multiple desktops, MMB-pasting, single/doubleclick consistency and tabbed browsing are just a few of many examples of it's superiority.

Yes, as usual, I put forward real examples while the Wintrolls make claims like "being designed by GUI specialists" without even an hint of proof.

Can you put up an example of what is so terrible about KDE/Linux? Of course you can't because there simply are no major shortcomings compared to Windows.

I had no driver issues, but when they occur it is an issue for Linux on the desktop. Poeple expect it to work right off the bat. They're gonna say "fuck this, I give up" if it doesn't - mos people don't get a thrill out of programming their own drivers like Slashbots would suggest.

It's exactly what I said. Essentially Linux is missing 3rd party support. 3rd party means hardware vendors including easily installable and well tested drivers on the driver disc. 3rd party support means software vendors offering Linux versions. 3rd party support means computer vendors preinstalling it.

However, 3rd party support doesn't have anything to do with Linux itself.

I usually don't get personal, but are you really too dumb to realize the difference between problems caused by programming mistakes/missing features (like MS Blaster) and problems caused by ignorant 3rd parties (like that USB-camera that doesn't work on Linux)?

Nope, merely pointing out that Linux isn't ready for the desktop for the average user. In case you missed it, Linus Fucking Torvalds agrees.

Linus Torvalds has realized that it takes years so that Linux can pick up the 3rd party support to become usable for the masses.

As I said, that's no technical problem at all. It's a matter of getting Linux entrenched and established - and that takes some time.

Re:Nonsense

[Jondor]

RedHat's CEO said it plainly "... when you go into a bestbuy and buy a digital camera that can sync with linux out of the box, then you know its ready".

Well, in that case it's ready.. I have used two digital camera's up till now, both needing driver installation under windows. Under linux/kde I plug them in, fire up konqueror and below.. I can just copy the images from the camera.. Wow!

And as it goes, hardware makers don't have to write drivers. They just have to release specs.

Re:How selfish of him

[k98sven]

Yes, he has only given us the best, most stable, FREE OS in the world. My God!

Well, that's all good and well, but I think declaring him to be your God* is a bit much.

(*Actually, to parody a saying about Larry Ellison, the difference between God and Linus Torvalds is that Linus doesn't think he's god..)

Re:Games!

[Dave2 Wickham]

It looks like you're trolling, but...

1- not required
2- Works fine for UT2003, ArmyOps, Savage, RTCW, etc...

Re:Linux becoming commercial?

[Anonymous Coward]

Nope, not at all. I see where you're coming from, but if that was the case, it'd already have happened -- Linux is a multi-billion dollar industry right now.

Any enhancements Red Hat makes (and distributes) to the kernel, glibc, gcc, GNOME etc. all have to be under the GPL. Then any other disto vendors can pick them up. Seeing as RH are the main force behind gcc and glibc at the moment, there's the proof -- it's not like RH are on GCC 3.3 while all other distros have to suffer with 2.95.3.

Competitive edges will come with service and support. It's virtually impossible, and pointless from a competitive standpoint, to try and "hog" certain aspects of Linux development. If a company wants right control over a free x86 UNIXalike, they'll go with FreeBSD[1].

-- A smoking AC

[1]Whenever people talk about Microsoft adopting Linux at some point and releasing "Microsoft Linux", I laugh. They'll go with FreeBSD, absolutely no doubt about it. FreeBSD can run most native Linux and open source apps, and MS can exert more control. There'll be no Microsoft Linux. Just "Microsoft UNIX" (or whatever) based on FreeBSD.

Re:Linus' point

[PhrostyMcByte]

Longhorn won't be out till 2005 if I'm correct and many users are very insatisfied with Windows XP, from Sobig/Blaster outbreaks dragging down productivity levels to random annoyances like messenger popups and a full suite of internet blockers/virus stoppers/software firewalls needed to surf the web.

All laymen users I know will say this- they know about the security, and they will say it's messed up, but it hasn't effected their productivity at all.

Windows XP SP2 due out later this year will fix the popups/virus/firewall problems. With Windows Update v5, users can keep their machine up to date without effort.

The people who do know that there are alternatives out there recognize that they will gain speed and pass Windows someday, but aren't willing to change until durastic changes take place in ease of use. When mentioning Linux most think it's CLI, and that's a downgrade.

Re:I agree (Actually, I don't)

[mwdib]

You write: Linux on the desktop is a long long way off from being as easy to use for beginners as windows is.

I've done more than my share of teaching total newbies how to use Windows. There's nothing intrinsically logical or sensible about the Windows desktop (95, 2K, XP), Windows' naming schemes, etc. It's extraordinarily difficult for an adult newbie to pick up. -- We tend to think of Windows as "easier-to-use" simply, I think, because of familiarity. Ditto with the Mac interface -- it's easy to use once you've learned how to use it. Come to Mac from a pure Windows or pure newbie background and there's still a learning curve.

Frankly, I don't think there will ever be a desktop that is "simple to use" from a newbie standpoint (at least until the computers can engage in an intelligent dialogue with the user and actually figure out what the user wants to do).

Consequently, I don't think any great re-imaging of the Linux (or any other) desktop is particularly required. Rather, I think the greater value will be in continuing to support a diversity of desktops with some focusing on new-user needs as much as others focus on the needs of sophisticated users.

After wading through four levels of menus on a default KDE install, I wish I had the skills to do some interface design myself. Grin.

Re:Linux becoming commercial?

[10Ghz]

Like what? GPL makes it really hard to abuse Linux (or more presicely, users of Linux). Like I said, I just can't see Linux turning in to another Microsoft. Yes, Linux could achieve complete market-dominance some time in the future. But the very nature of Linux makes it EXTREMELY difficult to abuse that dominance.

Re:Linux becoming commercial?

[Xabraxas]

I think the point that you are missing is that no matter how good a distro gets it's still open source. Anyone can fork it. Anyone can use those features in thier distro if they were so inclined. A monopoly is not possible because the 'trade secrets' are not secrets at all.

Observations

[gone.fishing]

Microsoft has dominated the desktop for over a decade. Unless something drastic and unexpected happens, it will take a minimum of five years from now for it to lose dominance. Having said that, I do think that 2004 is a watershed year for Linux and for Microsoft. Years from now, we will look back and identify 2004 as the year where the tides bagan to change.

Why do I feel this way? Very few companies in very few industries ever achieve the dominance that Microsoft has in the computing industry. Competition always keeps the underdogs going for the golden ring, and profits like Microsoft enjoys have other companies salivating. History shows us that very few companies can hold onto such an amazing lead over the competition.

Linux and other "free" operating systems hold a unique advantage over Microsoft's offerings. They are free. Microsoft can not afford to compete on price alone. Every day that goes by, the gap between Microsoft's offerings and Linux's offerings narrows the gap in quality. With Novell and IBM in the fray, that gap is sure to close even further. At some point, Linux's offerings will become the most logical choice for everyone. Microsoft's grip will sliip and they will slide. It won't be fast, they will lose by percentage points.

At least this is what I hope. I have no crysal ball. They have quite a war chest and they have a lot of lawyers. Maybe one of these hair-brained lawsuits from the likes of SCO will work. I don't know, and I sure hope not.

Linus is probably right but I hope that it is 5 years and not 10.

Have to disagree

[bogie]

"I'd love to have an easy to use system that I could handle without much difficulty while still having the power of Unix at hand should I want it.

This is not Linux."

But it IS Linux. I know this will come as a shock to Apple fans, but OS X isn't the be all end all of Unix desktops. I like many Linux users don't want a pc equiv of OS X. OS X does many things right, but it also does a lot wrong. OS X for x86 would be a real threat to Microsoft and would no doubt get more users using a semi-Unix but it's not what I'm looking for.

The only thing missing from Mandrake, Red Hat etc is real support from software and hardware makers. Documented hardware IS truly plug and play. Getting software installed/uninstalled IS moron proof provided that its packaged correctly. Like you said installation is easy as pie.

Imagine a distro running the 2.6 kernel with full oem hardware support, KDE 3.2, and the support of all the big software ISV's. At this point you have an OS that is easily as good as OS X and XP. So your right that we are indeed waiting, but not for OS X to come to the PC. We are in fact just waiting for Hardware and Software OEM's to fully support Linux. Maybe that won't ever happen, but if it does then you can rest assured that there will be no reason to pine for OS X on the PC.

The way I see it you have 3 options. 1) Buy an expensive Mac, thus putting yourself under the thumb of Apple and in a situation which is NOT an improvement over running XP. 2) Wait for OS X to come to the PC. 3) Wait for hardware and software makers to get off their asses and finally support Linux. It has been a long road, but I'm sticking with number 3. Number 1 is not and never will be an appealing option to me and most others.

Re:And the number is ....

[Smthng]

So what if he literally manages to use "literally" twice on the same line ;) .

I will disagree.

[khasim]

It would be simple.

Put all the apps that they would use for work in a folder on their desktop.

Also, have all those apps open when they first log in.

When they log out, save all the information about those apps so they will appear EXACTLY THE SAME when the user logs in again.

Then, have the items that the user is ALLOWED to change in a different folder. Like backgrounds and themes and sounds and junk like that.

Everything else is locked down.

The user info is saved to a server so any machine that the user logs into will have the exact same desktop as the last machine.

This is VERY hard with Windows (unless you're running a Citrix desktop). But it should be very easy with Linux (all apps served from the servers).

I important part is getting them connected to the apps they need, seemlessly and reliably. Every time, every machine.

All the end user should NEED to know about the computer is how to turn it on and where the blinken lights are that show that it IS turned on.

Everything else should be covered by training on the applications that the company uses.

Re:People get that Microsoft is garbage now

[tomcrick]

People are starting to blame Microsoft for their failures.

Definitely - people seem to be taking notice that Windows is not the complete package they are told it is when they are sold the computer. A good example is the amount of Critical Updates people have to download when they have a brand new computer just to make sure it isn't vulnerable. I had to do this for my parent's new computer with XP Home - it was no fun trying to download 16Mb of updates on a 56kbps connection.

I think it will be a massive step forward to get the average home user (whatever that is!) to actually know there is an alternative to Windows and it's not just for 'hackers and programmers'.

Re:I don't

[romanr]

I don't. The GUI design is at best inconsistant - they may be trying to play catch up now - but a lot of what is happening is based on behaviour that was thrown together years ago and can't / won't be fixed. I don't actually believe they test usability with their focus groups - they probably concentrate on what eye-candy looks best.

A perfect example of how non-user friendly Windows is the way your keyboard focus gets stolen. I touch type - I don't spend a lot of time looking at the screen - i end up get very, very irritated because some window/dialog has decided to open and steal the keyboard focus - at best, my keystrokes end up in a black hole, at worst - they're invoking some action that I don't want to do.

The Amiga got this right 15 years ago - the programmer guidelines stated that you don't steal focus - Microsoft would do well to re-think a lot of their GUI guidelines (or at least follow their existing ones - they tend not to do that for their own apps anyways).

Chasing Taillights Is

[Christian Engstrom]

Why are we all so focused on cloning something we all agree is awful?

I actually belive that that is an excellent question, and I'll be happy to provide the answer:

Because 90% of all computer users are used to Windows

(The rest of the following rant is essentially a repost, so I apologize if you have already read it.)

You can feel that it shouldn't be like that, and you can make hundreds of snide and clever remarks to the effect that Windows users are too stupid to recognize their own best interests, but you can't change the facts: at least 90% of the people who are using a computer today are using Windows.

It is not every day that a court of law makes an official market survey and releases it freely on the net, in line with the finest traditions of the Open Source movement. Yet it seems that the very people who really believe the most in the benefits of free and open information, are remarkably reluctant to use it when it's available. Think what you will in private, but please please listen to judge Jackson: if Linux is going to have any impact at all in the desktop market, it is Windows users that will have to be converted.

There are a number of good reasons to make the switch to Open Source --- open file formats, control over future license costs, etc., etc. --- but if it means that you have to spend six months cursing all the little things that are different, so that you can't focus on what you're supposed to be doing because you have to relearn all your automatic reflexes, how many people will decide that it's worth the effort?

A lawyer might perhaps consider switching from MS Word to StarOffice simply to make sure that all the files that he creates today can be opened and read on another computer ten years from now, when the case has finally reached the Supreme Court or whatever. But how may chargeable hours is he prepared to let it cost him in the first six months?

It somehow seems that a lot of the people who develop Open Source applications take a special pride in inventing amusing little pitfalls for the Windows user who might be prepared to switch camps. In StarOffice, the keyboard combination to insert a non-breaking space is "Ctrl-Space", rather than Word's "Ctrl-Shift-Space". Please, somebody, why? Of course this is something that one can relearn if one has to, but what's the point of it? The first time a would-be convert, who has been using non-breaking spaces in Word, tries to insert one in a text in StarOffice, it won't work. Whether he decides that non-breaking spaces are not available and that the product does not fulfill his needs, or interrupts what he was originally trying to achieve and starts exploring the help system to find out what it is that he has to do, he will not feel more favorably disposed towards Open Source programs for having tried one. And so unnecessarily.

I could recite any number of examples: if you type "Ctrl-A Ctrl-Return" to mark all posts in a newsgroup as read, Mozilla will instead choose to open a couple of hundred windows (one for each post in the newsgroup), which will cause the system to freeze, so that it has to be rebooted. Excellent marketing ploy.

To change some settings in Mozilla you should of course look under "Edit" in the menu system, and not under "Tools" like in all other programs in the Windows world. Brilliant. How could you possibly fail when you make it so convenient for the user?

And please, don't come and say "RTFM" now. Why the **** should someone who has been using a computer for years have to consult the FM (provided there actually is one, of course, but that's a separate issue in its own right) to perform a so completely trivial standard task as the ones mentioned here?

And please don't come and say "but you can change that if you spend a couple of days learning how to reconfigure the program from the bottom up" either. Pe

Pre-Installation would make a big difference.

[jmors]

While I would have to agree that linux is not quite as simple to use as a typical windows machine, I think the biggest thing preventing linux from being accepted is the lack of pre-installation on store bought machines.

Think about what would happen if you gave the average desktop user, not technical user now but simply someone who wants to use a word processor, send and receive email and browse the web, a computer with a blank hard drive and a windows OS cd. If windows did not come pre-installed would it be the desktop of choice for average users?

I have friends and relatives who would never have used Linux if not for someone to help them through the install process but after setting up the basics they find it every bit as easy to use as a Windows OS. I honestly believe that if Linux pre-instralls were as available as Windows, we would see a much higher rate of adoption on the desktop.

Penguin Logo Program

[Anonymous Coward]

Users need a central, glossy penguin logo website, where approved hardware (pc's motherboards, video cards, sound cards, etc) are listed and unapproved hardware is blacklisted (until drivers are ready). Kind of a Consumer Reports version of Linux supported hardware, where to be listed the hardware has to be fully supported with drivers and proven to install without any bullshit.

By this I mean a very high standard of compatibility. Naturally, people are going to install whatever and that's fine, but to qualify for the hardware logo website, standards of ease have to be met fully.

Device approval needs to be in a heirarchial format, starting with the motherboard. For example, Radeon xxx isn't approved by itself, but Radeon xxx is approved for install into an Asus xxx motherboard, with Kingston xxx memory, with a Creative xxx soundcard, etc.

Yes, such a site would approve a very narrow set of compatible hardware, but that's ok, the idea would be to give a simple place for newbie users who don't want to hassle it to go to choose products that everyone knows will work without a fight.

Most of us (of course) would ignore it and have fun hacking away at insane hardware combinations, because we like that, but if we're talking about the general user population, they couldn't care less which motherboard or soundcard or whatever they have, they just want it to work without the hassle.

Then, get hardware review sites like tomshardware and anandtech to find combinations that work really well together, and to promote the site. Try to get consumer reports to feature it in an article. Then it's up to the hardware makers to support maybe not all of the stuff they make, but at least some of it.

ACL vs. Unix perms and groups

[beakburke]

Frankly all the possible combinations that ACLs provide only serve to add unneeded complexity to the matter. The judicious use of groups and unix permissions, which, IMO are much simpler to grok that the ACLs, results in a system that is easier for the administator to understand. And thus the system is more likely to be correctly configured, with proper security. Yeah ACLs are "more powerful", and if you want them Linux will support them (in ext3 and jfs or xfs iirc). It's better to have "simple" permissions done correctly, than to have your "fine grained ACLs" configured wrong. It's really an application of KISS and the 80/20 rule. Frankly, I'm not entirely convinced that ACLs really provide anything that groups can't. ACL's make the easy moderately difficult, and the difficult moderately difficult. Where as standard unix permissions and groups make the easy easy, and the difficult difficult. Yeah, you can add all the fine grained stuff you want, but the need for that is the exception rather than the rule. So why compliate the simple stuff, just to make the diffcult stuff only moderately easier?

L.T.: Desktop in 5 or 10 years...

[Saeed al-Sahaf]

One of the things that bothers me is this statement from Linus:

I mean it's going to take, literally five to 10 years before "normal users" start seeing Linux desktop, but in the technical space it's doing pretty well, especially in companies that can support it already. Five or ten years? I L.T. feels this is the time frame, I'm worried.

Re:Its the definition of "Good"

[Anonymous Coward]

> I say it's infinitelky superior to do the Linux desktop the good way, the right way, and emphatically *not* the Windows way right from the start! Sure, the amount people have to learn will be more at first. but then it will be over.

You've missed his point - the more "more to learn" there is, the earlier the user will indeed say "I'm done with this :-(" and go back to Windows without looking back. Your definition of "the right way" is someone else's definition of "change for its own sake".

Enough people do this and yes, it *will* be over, but not in the way you've expected.

World Domination

[0x0d0a]

Can someone explain to me why we need to get linux running on every desktop in the world exactly??

To say it step-by-step:

1) Hackers like Linux more than Windows. It's a nice, powerful OS.

2) Microsoft sells Windows. It wants people to buy copies of Windows. One major weapon in its arsenal is compatibility -- .doc files, for instance, are not easily readable on non-MS products. MS has significant incentive to deliberately attempt to introduce incompatibility.

3) Hackers are not islands. They must interact with other people. Sometimes this means getting DSL service. Sometimes this means having to use a computer specified by an employer. Sometimes this means being able to read .doc files. Since Microsoft is The Institution and tries to isolate itself from other efforts, hackers frequently have to put up with Microsoft's products, even if they do not want to use them.

4) Hackers, frusterated with Microsoft, happily work on Linux and other Microsoft-alternative efforts.

Linux having a 30% market share or more would have major benefits (well, and probably drawbacks as well):

* Hardware compatibility. Someone has to write drivers and test and support hardware. It's expensive, so usually this sort of thing is subsidized by lots of people. If many non-hackers are using Linux, then hackers will get hardware support subsidized by non-hackers. This is a Good Thing for them.

* Games. There needs to be a lot of folks willing to buy a game before a company will port, test, and support it on Linux. It's expensive, so usually this sort of thing is subsidized by lots of people. If many non-hackers are using Linux, then hackers will get games subsidized by non-hackers. This is a Good Thing for them.

* Enabling People. Hackers are human too, and they feel good when they let people do something more. It's rather like the digital artist that introduces a conventional photographer to Photoshop. When the photographer's eyes light up and he realizes what he can do, and his ability to produce value increases, the artist feels good, and has helped society. Linux has a number of capabilities that Windows does not, and introducing folks to them would help society.

* DRM. Lots of hackers are not thrilled with the concept of DRM. Establishing a less monopolistic platform rapidly makes it much more difficult for anyone to get everyone using DRM.

* Environment. I'd love to never have to use a Windows box again. However, I run into them. The more people using Linux, the more folks paying people to work on and develop things for Linux, and the less one has to support Windows machines.

* Elimination of proprietary protocols and formats. Only one person directly wins if a proprietary protocol or format is in place -- the vendor of the software using it. Consumers lose, and competitors lose. Linux, having a large collection of entrenched open source and open specification software packages, has a good amount of inertia to not having closed formats.

Now, I grant that there will probably be drawbacks to a dominant Linux. Whatever the dominant easy-to-use distro is, it will likely have security failings, may force people to use a GUI to configure things, and may have a vendor doing all kinds of licensing deals for exclusives (like Microsoft's AOL icon on the desktop). Trojans and viruses will likely be more common for Linux. Politics will become more involved with Linux, just as it did with the Internet (imagine the same thing happening to the FSF that did with ICANN -- being taken over by less-than-nice corporate interests. Ick.) There will be many packages ported, and some of the existing Linux software that appeals to hackers -- small, CLI programs that can easily be combined -- will lose relevance as folks use ported, large, potentially buggy software packages like MS Office. There will probably be more strict backwards compatibility constraints, and cruft will more easily bu

Stumbling blocks for Linux on the desktop

[logicassasin]

Here's what I see being the real show-stoppers for desktop Linux adaptation:

1. Reliance on the CLI: Yes, in a perfect world, everyone would be comfortable with using the CLI to accomplish tasks from installing a driver to reading email to whatever. REALITY, however, is different. The vast majority of Win32 and MacOS users NEVER touch the CLI. No one wants to be bothered with it. The Linux elite's insistance that everything be centered around CLI apps and whatnot is going to prevent Linux uptake. Yes, we should all learn it before diving into Linux, but think about it this way; Apple, with it's BSD powered OSX, does NOT require it's users to know a damned thing about the command line in order to use their OS. It simply works well without it. Of course, power users can get at it and run as many shell scripts as they wish to, but those that don't know about command line stuff are not forced to learn it.

2. Installing new hardware in your PC should not be harder than plugging it in and installing a driver. In all of the years I've been using Linux, I've rarely ever been able to simply install a new card and not have to install something other than a driver. There have been too many times where I have to fish out my install CD's or search the net for some obscure dependancy package, or worse, have the dep already installed, but the driver's installation script not detect it properly. I've pulled out my hair trying to get my little USB webcam (Cool-I-Cam Stylus 1000) to work with GPhoto/Gphoto2 only to give up after weeks of trying (it took less than 5 minutes to get it up and running under Windows 2000). My IOGear USB2 card STILL doesn't work with Linux (the driver is included with Win2000 SP4 and is also available as a tiny download from the IOGear site). Stuff like this annoys the hell out of me. Honestly, I shouldn't have to deal with it and neither should anyone trying to use Linux for the first time. Until hardware installation is fixed, desktop linux will never happen.

3. Apps. I cannot stress how important having GOOD applications is to the average user. Star/Open Office is good, I'll admit that and it's an excellent start in the direction that things should be heading. However, there's simply not enough applications of this caliber. There are no pro-quality audio applications, no Macromedia authoring apps, games are hard to come by IF they're ported to Linux, and nothing that's truely like EZ CD Creator or Nero for CD burning. Until commercial applications start coming over to Linux, we're not going to see many people moving to Linux.

Think of it this way; The Amiga is/was one of the greatest machines ever built and it had the BEST OS of it's day. It's lack of applications (and lack of marketing push) killed it's desktop uptake. In 1990, I knew more people that had inferior PC's than had Amigas and the sole reason was that the apps they needed were not available for the Amiga. Same for the Atari ST, Same for the BeBox. Apps drive adoption, not just the GUI.

4. Elitism. Linux elitism is rampant. If I ask a question in an IRC channel on how to do something in windows, I get a dozen good responses. If I ask a question in #linux on Efnet or a similar channel, I get a bunch of "did you read the man pages?" "RTFM", "Linux is obviously to difficult for you, go back to Windows" or similar responses. Oddly, I don't encounter the level of elitism when looking for help with any other flavor of unix or MacOS (The guys in #SGI/Efnet were particularly helpful when I had a problem reinstalling Irix on my Indy). The attitude that a lot of Linux users display towards newbies will turn off just about anyone to Linux. Kill the attitude, learn some manners, and lend a hand.

Now, before I get flamed, I must let you know that I AM well versed in Linux. I'm currently working as a Unix admin, overseeing a mission-critical, money making production server farm for a Fortune 1000 company. I make my living using Linux, but cannot see having my wife use it for her business (She's a mortgage broker)