Slashdot Mirror
What Does Open Source Need for Mainstream Desktop?
HesAnIndieRocker asks: "So what will it take to make open source technology a mainstream alternative on the PC desktop? It feels like we've been on the cusp for many years now and the applications available for most common tasks are certainly competitive, but we still hover around a 5% market share by most accounts. I've recently written an article in my weblog about some possibilities, but I'd love to hear what others think."
patience, young padawan.
Foobar2000
Seriously, it has to be the most awesome audio player I had ever used, and Linux doesn't even offer an alternative (other than running it on WINE) that comes close to the pure awesomeness of it. I like Rhythmbox and whatnot, but seriously, Foobar2000 could do it all. I mean, it even cured a cold I had once!
'Yes, firefox is indeed greater than women. Can women block pops up for you? No. Can Firefox show you naked women? Yes.'
SymphonyOS just hit beta one, and it looks like a really nice project. It might be a mainstream hit, and that's what Linux needs. www.symphonyos.com
I thought Ubuntu/Breezy was going to include Beagle, Tomboy, and all the other hip new desktop apps that havn't made it into any distributions yet. I guess it will probably be next release or something. I saw a demo of integration of instant messaging into a bunch of different desktop apps. So any app that has a list of names becomes a "contact list" that you can click on to send a message. Integration of instant messaging might not sound like much, but instant messaging is probably the most non-integrated technology on the PC. I've probably never seen a URL for instant messaging someone. I presume you could do it, but it's just not as mainstream as, say, mailto:// urls.
How we know is more important than what we know.
There should be absolutely one installation method, that should encompass ALL distributions.
/usr/bin, every doc into /usr/doc, every lib into /usr/lib*, or a distribution that keeps every single application's components in it's own seperate directories. All using the same install format.
/usr/lib/xlib1.0.so and /usr/lib/xlib1.1.so are actually totally compatible with each other, so you can erase /usr/lib/xlib1.0.so when installing 1.1.so .. but, /usr/lib/xlib2.0.so has a totally different interface, so if you have programs that depend on /usr/lib/xlib1.1.so and you install /usr/lib/xlib2.0.so, the installer will know to keep the 1.1 version around as well. (this would also eliminate the idiocy of having things like "glib-5" and "glib2-2", when glib2 replaces glib .. don't take any of these examples as examples of absolute truth, i'm just using the names as examples, rather than as case studies)
How each distribution actually DOES it can vary as it wants.
Each application should be packaged, with a file that has a lot of information about whatever is in the archive.
What each file is, wether it's source code, a library, an extension for something else, the main executeable, or some stupid utility to go with it.
Then it's up to the installer, based on WHAT the file IS, to determine where it goes.
Then you can have distributions that use the traditional *throw every executeable in the entire world into
Of course, each would also have version information, and also "compatible with" and "incompatible with" information, particularly for libraries, where
And I really love the idea of "nothing should ever be executed without the installer having previously known about it".. that would be a great thing to add to a distribution, IMO. Hell, the installer could keep track of checksums of the executeables, and make sure they haven't been modified (such as by a virus or worm or rootkit or malicious hacker) before running.
A unified installation METHOD (doesn't have to be the same program on all distros) would solve a huge amount of Linux distribution problems, and perhaps even provide an answer to more general computing problems.
"Champagne for my real friends - and real pain for my sham friends!" http://ericblade.postalboard.com/
I think the main features are there. It just needs an x-factor to grab the public's attention.
When does a rectangle become a line?
After ranting for many years about never upgrading past Windows 2000 (mainly due to having to call MS should you need to reinstall, etc) I finally had to put my money where my mouth was when I purchased a new laptop a few months ago (eMachines m6805, AMD64). To make a long story short - I had a buddy help me install SuSE 9.3 on it and I've been on the penguin at home ever since.
What I have seen from a long time windows user perspective is this: functionality and abilities (and stability) is far greater on (SuSE 9.3) Linux then I ever experienced even under Windows 2000. The problem is the "Win95"-esque problems... such as getting my wireless networking card to work. Now fair enough... I had/have these issues with Windows upon an occasion as well, and I can work my way thru them faster then on Linux simply because I've got ump-teen years experience under Win. My gripes come when I have to follow quite esoteric HowTo's to get my gear to work (or to get this thingy to install, or, or, or).
Most times, I'll get whatever widget I need to working thanks to 2-3 of these HowTo's (mainly because 9.3 is a popular distro). But if I were unlucky enough to be one of the first people with problem 'X' I know I'd be screwed. Just the number of widgets and command line prompts and whatnot I had to tinker with to get my onboard WLAN card working was stunning. Then, after it was all said and done, I still couldn't get to websites 'cause the router didn't have valid DNS IP's configured (Linux seems to be a bit "bitchier" when it comes to certain things). Thankfully I picked up on this before blaming the card!
Now... I'm happy with SuSE/Linux and I cannot ever seeing myself turning back, but I'm a nerd that enjoys the occasional hw/sw challenge (something I've not had on a windows box in probably 3+ years). But for Joe Sixpack? We (as in the all of us, or the royal... take your pick) need to bring Linux's usability up past Win95, because in my opinion, that is exactly where (SuSE 9.3) Linux is currently at.
"1984" was ment to be a warning, not a guidebook. You hear that Kim Jong-il!? BushCo?!
Debs, slackpacks, rpms, etc are a bitch to mess with. I don't want to alien -d xxx.rpm; dpkg -i xxx.deb every time i come across a file of that type. The linux distros should make packages that run on any pc running any distro. also, apt-get, yum, and emerge should come together to make a single repo, so all the linux systems are up to date with everything. this would make our systems much more compatible with eachother, albeit it would take some time.
It's too cheap to be any good...
Seriously. I was recently given the opportunity to take the OS X plunge and I've had the OSX epiphany that goes along with it. It's changed the way I see user interfaces, and I finally understand why the mac "addict" types have been so rabid. It's _that_ good.
FOSS UI's need to integrate or at least peacefully co-exist, and do so with a standards-based foundation.
That's the key (or ticket depending on your view).
Don't think that a small group of dedicated individuals can't change the world. It's the only thing that ever has.
...oh, and don't even get me started on getting my touchpad to "work" (hand editing a config file per SuSE's support pages, then the bastard get overwritten [and loosing the hand-config] by YaST the next time I tinker with other settings - AAARRRGGG!)
"1984" was ment to be a warning, not a guidebook. You hear that Kim Jong-il!? BushCo?!
So, how is Foobar any better than VLC? I check out the website and it looked like it could do _almost_ as many formats as VLC, _if_ you download/install a bunch of extensions.
Do what MS and Apple do? Spend most of our resoures on marketing and PR?
Thats it. Thats all it needs. I know over a dozen people who would switch if it just had mainstream games.
I still have more fans than freaks. WTF is wrong with you people?
ndiswrapper and ndiswrapper-utils are great for wireless nics. ndiswrapper -i inffile ndiswrapper -l modprobe ndiswrapper ndiswrapper -m
In particular, what is, IMO, the greatest strength of the "linux desktop", *thousands* of applications available in a simple and easy manner. Advertise THAT. Easy updates, tons of apps. Windows per se, the company, has nothing like that, not even close. And they barely update their own stuff, let alone the various applications people like to use. I honestly don't think most joe and janes out there are even aware what you can get with a default install of any of the major linux distros. With a few tweaks and sops to economic and social reality (MP3s are not going away anytime soon, costs money, the distro vendors need to suck it up economically and politically at this time, same with DVD playback, etc), primarily in the media playback and games areas, you'd have a winner IF it came preinstalled from joe large company with significant public nameshare. The rest would follow, even the largest companies are herd animals.
Just needs some *advertising*. FF proved that it is possible to go from relatively obscurity "to the masses" to a quite respectable showing in a short period of time.
That and some larger hardware vendors actually offering it on their machines. AOpen has just announced their miniPC will be shipping with Linspire Linux as the default installed OS. It's starting to crack-slowly. XP is a hundred dollar extra option. And IF the big vendors shipped it installed, their usual bundle offerings would work out of the box, eliminating a lot of that obscure driver foobar nonsense. Acme & Odds, ltd. videocards and printers and USB widgets of various kinds and assorted add-on whatnots WOULD listen to some folks like Dell and provide drivers that work, and work well, facing a potential order in the hundreds of thousands or millions. No fooling around then, they would get to work and do the coding. Right now it is "meh, who cares?" with them guys,they throw some slop out and have a tiny webpage and forum someplace on their site for linux...mostly to shut up the whiners, in their opinion.. pretty obvious really. This is because there's no established market, a chicken and egg deal so far. Someone *big* has to blink and blink hard and make a move, a normal high stakes semi-risky business move. Joe Bob at the local whitebox shop buying 5 machines doesn't show up on their radar. It needs to be the large hardware vendors and game company devs push this, they are the only ones who can really crack it, it is NOT, repeat NOT, going to come primarily from the software side of the equation by some billion humans deciding one morning to "tryout this li-nux thing" they heard about. Ain't happenin'. Not any time soon, anyway. You aren't going to change human reality that people run what comes pre-installed. It just "is" is all.
A multimillion dollar ad campaign. That's what it'll take.
Now please, for fuck's sake, let's talk about something else. Every couple of weeks there's another damn article whining about how open source is soooooooo close to succeeding as a mainstream desktop alternative and asking what's keeping it from taking that final step, and everyone always answers "consistency" or "usability" or "accessibility" or "pictures of naked ladies", but the real issue here is that Grandma doesn't know what the fuck Linux is because she doesn't see ads for it on TV.
Goodness I'm bitter today.
Little things like real replacements for common programs... not half assed attempts.
;-)) of the flagship Intuit product.
There are applications that I NEED on a daily basis that cannot be emulated and there is no equivalent. Don't get all bent out of shape about this, it is just a fact that noone as yet has been able to 1:1 replace Dreamweaver.
Also, if you are bound to some program for your business or home use you kinda have to go with what works.
3 simple examples:
Dreamweaver - NVU is not a replacement by any means... and the sad part is NVU is about as good as it gets in the OS world.
Quickbooks - No, emulation is not an option... it is sluggish even in windows. And don't try to tell me that the dozens of disparate accounting projects on sourceforge or freshmeat are going to come anywhere near the simplicity and dependability (damn straight it is dependable... lots of backups
Radmin - Remote Administrator by Famatech will not work on a *nix box with emulators... some forums have some well meaning people saying "So what if the keystrokes don't work, you can copy and paste text instead" - Thanks... but no. I have a hundred clients with radmin licenses and when compared directly to Radmin I would rather eat glass then install the latest VNC variant. If I had started out with VNC it might be different, maybe if I started out with VNC I wouldn't be agonizing over trying to switch to *nix.
The plain fact of the matter is, there are many programs that are not directly replaced. I have been trying to switch to a distro for 5 years. I install a new distro on a relatively modern laptop everytime one comes out. It sits on my desk and I genuinely try to use it. So far I really like the debian distros (Like Ubuntu minus the constant sound effects), where 2 years ago I would have been hard pressed to use anything but SUSE.
Everytime I make a legitimate attempt to change over I run into a half dozen piddly little sub $100 applications that I cannot emulate or replace.
I have seen Linux make great strides in the last 5 years I have been following it. I have moved most of my servers to linux and bsd (web and mail), I even replaced my SBS2k3 server in favor of ClarkConnect Home 3.1 (so sweet) to run my windows domain.
I used to get my ass kicked trying to install an HP JetDirect printer, then CUPS started coming preconfigured in the distros... man was I excited! Then I would find that dual displays were troublesome... again that has changed for the better.
All I have left are DreamWeaver (not just wysiwyg, but the templating and ftp site synchronizing) and Radmin since all my windows clients have it. I manage 50 or so client computers and a dozen windows servers in Alaska from Seattle with RADMIN... so it is kinda important to me. If I can figure out those hurdles then I am on the way.
Sometimes it is just the little things.
Boredom's not a burden anyone should bear.
We don't need a new distro every week. Too many people are wasting their time reinventing the wheel. They need to cooperate and consolidate. Until then, a mainstream desktop will always be 2-4 years away.
Well, the internal database cache speeds things up enormously, and the ability to completely customise and script how the playlists and other interfaces are shown. Sure, you can edit the .glade files for Rhythmbox, but there's only so much customisation you can achieve with that; other things have to be done at the application source level.
'Yes, firefox is indeed greater than women. Can women block pops up for you? No. Can Firefox show you naked women? Yes.'
1) Drivers
2) Drivers
3) Drivers
From a High Level, to convince people to switch systems, you have to offer them something an order of magnitude better than the system they are invested in without giving up any substantial functionality. This means:
The computer should do exactly what the user wants, whenever the user wants, without the user having to think about it. This means extensive end-user testing and brutal simplification. The user should never see anything unrelated to exactly what they want to do. The folder heirarchy they see on the drive should ONLY contain things relevant to their activities. They should be able to re-arrange everything on their disk and still have it all work. They should never have to edit a config file. They should have to wade through "interface spam" of a million options which one in a million users will ever actually use. And yes, this means extensive high level architecting of everything that goes into the system, something OSS isn't traditionally good at doing.
The computer sould be able to replace legacy systems. That means being MS Office compatible, not a small feat. Not just word, but scheduling, and Excel macros must be readable in the new system.
The computer still needs a killer application or usage that makes everone want to switch to it. Apt-Get is pretty killer for me, but command line functionality will never reach the average desktop user. What else can the nature of Open Source provide? How can we use dynamic re-compilation to do something amazing that retail software can't provide?
No offence, but Linux as a desktop OS is still pretty hacky. There are a million unnecessary (to me) files hanging around when I'm just trying to do something, dozens of different ways to try to do something but four or five of which will work, command line still being integral to anything fun on the system (and even some baseline functionality), etc. My feeling is that the current state of Linux isn't the way to get there, any more than Dos should be the way to get to Windows. Perhaps it is time to throw our collective weight behind SkyOS, Zeta, or another upcoming Desktop-oriented OS, and refocus Linux on being the kick-ass server OS we all know it to be.
The ______ Agenda
I was trying to lay out the situation as I see it.
On the consumer side, we are stopped by the fact that all of the major vendors are bullied by MS, perhaps more implicitly than in the past, but bullied none the less. If any of them were to support Linux on an equal footing with Windows, in catalogs, on the front of the websites, etc, the would risk losing their OEM license or getting their rates changed. When they sell cheap commodity boxen in a very competitive space, that can easily kill a business model. All the big players are profitable and have boards to report to, so throwing caution to the wind is unlikely.
On the business side, we are faced with a generation of tech workers raised to develop applications for the Windows platform, administer NT/2000/2003 servers, support XP users, get MS certifications, etc. All of these people, right or wrong, have staked their careers to the MS platform and are unlikely to see the incursion of UNIX systems into the workplace as anything less than a threat to their jobs. The amount of corporate infrastructure currently geared towards the Windows platform, from helpdesks, to workgroup servers, desktops, custom applications, macros, Exchange email, etc is staggering, and there is little chance that CTOs will wake up one day and say "Oops. Let's try this other thing."
What I'm trying to get at here is that there is more than simply applications keeping Linux/UNIX from the desktop; that is mostly there, at least in the form of reasonable substitutes. It is the inertia that has formed around Microsoft's platform that keeps open source from becoming mainstream.
If you get a chance, check out the article. :)
Given all this, what is our next move?
"It is not advisable, James, to venture unsolicited opinions."
That does sound like a pain, but to cope with it, you could setup a cron strip to check if it was wrongly changed and replace the file with the config-settings you want. Simple fix to deal with a problem.
I was told this by a linux friend.... and after 12+ hours of him giving me instructions about how to get my netgear card working on ubuntu it finally worked. Restarted... gone again. I ran home to windows and realised that windows has never actually failed me in anything. Haven't started it since.
Simply put, 95%+ of all users don't care at all, and see no reason to change.
Slashdot - where whining about luck is the new way to make the world you want.
I think that it's too hard to install a lot of software under linux. apt is great under Debian or Ubuntu if your package is in the repository, but for things that aren't in the repository, installation is pretty hard.
.dmg file, which is automatically mounted, and then you can just drag an icon into your Applications folder. That's really cool. Most OS X users don't even know that the app icon that they're dragging is actually a directory.
In OS X, you can download a
For me, some of the really cool gnome and/or mono apps (like beagle) are great examples of what's simultaneously right and wrong. The apps themselves are pretty great, but they're incredibly difficult to install on many systems.
I have been using Linux on my desktop since 94, so I can relate to this feeling.
.... *application promptly dissapears*
I tried to rip a music CD on my Fedora Core 4 system the other day...
I put it in and ran Sound Juicer... it saw the CD and loaded all the track info for it. So far, so good.
I wanted to set it to rip me an OGG at quality 6, the same as all my other ones I ripped in windows. It would let me choose between OGG and FLAC (no MP3), but there was no quality setting. An audio ripper with no quality setting?!? Impossible I thought...
I looked in the help file, and it said nothing. Though the help file mentioned if you wanted MP3, you could use something called 'gnome-audio-profiles-properties'. There was no link to run this in the program, and I can't find it in my Gnome menus, so, being the guru I am, I ran it from the command line...
This is a GUI which has a text field to type in a GStreamer pipeline!
"audio/x-raw-int,rate=44100,channels=2 ! vorbisenc name=enc quality=0.5"
Like anyone (especially grandma) is gonna know how to fill in that!
Anyhow, me being the guru I am, I fish through it and see the quality setting... I want OGG quality 6... so what does "0.5" mean in OGG terms? Well, let's look in the help file...
*clicks help button*
No "this help file does not exist" dialog, no stack trace, nothing. *Poof* Gone.
This is why Linux still isn't really ready for my desktop.
Uh, yeah. That guy was right and you're wrong.
There's this persistent delusion that people will all switch to Linux once there's not a reason not to. Except for (as that guy had noted) a handful of tinkerers and rabid Microsoft haters, users will change because there's a reason to change, not because there's not a reason not to change. When open-source desktops provide one, the users will be there.
(Incidentally -- 5%? Where the hell do you get a consensus for 5%? It's under 1%.)
What I'm listening to now on Pandora...
Yannoe what I think it needs? Linux specifically, anyway: a GUI for _everything_. Joe Average Outlook-user is terrified by the command line. Fix it so that he can pretend it doesn't exist, and he'll be much happier.
"If any of them were to support Linux on an equal footing with Windows, in catalogs, on the front of the websites, etc, the would risk losing their OEM license or getting their rates changed"
Sorry, but this old excuse just won't fly any more. MS has already been threatened by the government since the monopoly case for doing much less than what you suggest. At this point they'd probably be fined over a billion for pulling that sort of stunt.
When you look at the kind of settlements they've been paying out the last few years it's clear they don't have the stomach for going that far.
There. I've said it. I run both WinXP Pro and Linux (Redhat 9.0 and KDE) Linux/KDE is clearly slower starting aps than WinXP. This is a fairly big deal in a business environment.
"Eve of Destruction", it's not just for old hippies anymore...
We need to have an app that will interface with MS Exchange and take care of all the things that Outlook does. Then the corporate desktop will be a much closer target, especially since it will mean saving the cost of an Office license for each seat.
Currently the choices are Outlook, Outlook Express and Outlook Web Access.
A list of Hardware, the driver it has and the latest driver available.
Why is this so hard?
The reason that I got into Linux in the first place was because I wanted something "different." I kinda think that that's the reason that most people got into Linux was for something different. The problem is that all of those differents were just a little different than all of the other differents.
There are bazillions of technical things that can be done to improve the overall quality of the Linux system. The problem isn't quality, it's the confusion of usability. (i.e. Which is better, GTK or QT? Do I want a DEB based system or RPM? Can I use both? etc, etc, etc).
One of FOSS's many beauties is that if I don't like the way something work I'm free to go and change it to my hearts content as the drool rolls past my chin. There's only one minor problem with that: my code sux... a lot. Talented coders are a rare breed, even rarer when the word "Visual" is not the prefix to your coding tool-kit. Think that the career options are rough now? Imagine all of your end users able to whip up a new interface model on a whim! Hell, I'm a lowly net/sys-admin and the idea of that gives me the shivers!
Anyway, back to the topic...
If I had a chance for recommendations to be seen by The Powers That Be they would be the following:
1) Separate your code from the Graphic Widget kits more. K3B is enough reason for me to keep the kde libs loaded on my machine, but konquer on my desktop is a little to kandy for my tastes (i know that's flame bait there, sorry). Build packages that can use either graphcial set with just some minor modifications, a-la "openoffice.org-gnome" in the debian repositories. That may not be the best example, but as I understand it, the widget sets were meant to give coders the ability to easily use a readily recognizable interface for the end user, not to be a staple in an application who's purpose is not to render pretties on my screen. I know this is nowhere nearly as easy as I think appropriate in this paragraph, but it's worth thinking on. Perhaps it's up to the interface writers to perhaps standardize their calls for said objects: make a GTK button the same call as a QT button, or some-such-thing. Of course there will be differences and additions (there always are) but the basics can be covered, I'm sure.
2) Packages... make up my mind! RPM, DEB, PKG, source... they've all got their beauty points, but having some only available to certain users burns me, and I've been using Linux in one flavor or another for 10+ years now. Recently, I ran across a program called "checkinstall" that has saved me from dealing with install-hell when a package is only available via source. This program allows me to take a source package and build either a DEB for my desktop or a RPM for my laptop. Again, (refer above) my coding sux, but wouldn't it be possible to expand on to expand on this, mix it with "alien" and allow it to handle conversions between types of packages, then build this into YaST and Synaptic and (dammit, I forget what RedHat and Mandriva are using now)? Imagine if SuSE, Debian, RedHat, and Mandriva got toghether on this idea, and got all of their installers able to handle any of the packages and installation methods in *gasp* one install database (per system of course)? Even with the extra overhead, it'd still be faster than, say, InstallShield(R) packages! Convince maintainers and developers that a properly formatted hash-style format can be used to cover the "provides" and "depends" stuff is in the source packages, much like they already should be in the various packages, and we may have a winner here. It may also be handy to let the Installer ultimately decide where the files of certain types go, so the Debian people and the Mandriva people don't start throwing their peanuts over the bar at one another any more.
3) Unified system management: On this, I'm thinking along the lines of something like webmin, where add-ons are constructed and placed in the framework for controlling certain aspects of my system. Being a Debian user,
--- no sig to see here... move along.
As far as his comments about user lockin go, many Universities (colleges in America) have already switched over to Linux often as a dual boot option.
Inviting students to tinker, schools move farther and farther from MSoft each year and need subsidies to keep them on board this will do more to put linux in people's education than the tiny confrences Microsoft is able to organize.
The thing holding linux back is it's inability to take risks regarding legal issues (Drivers mp3 codecs etc.) and the difficulties involved in it's use.
What's to speed up? It's an audio player. I can't remember ever having speed problems with my audio players.
By the same token, I can't remember ever feeling the need to script an audio player. But maybe you have more time to waste than I.
This space intentionally left blank.
I'm sure I'll be flamebait, but i'll have to file you under "you still don't get it". (that and shamlessly promoting your blog.)
/etc. Thats playing with fire.
Not only don't you get it, you won't listen. Sure - argue that I'm wrong. Willing to take a bet which one of us will still be right in a year? Two years?
You said: "Instead, the problem with Linux is that not enough people are getting it preinstalled on their computers."
Bzzzt. Wrong. Do you think everybody would automatically love Linux if it was dumped on them? No. Not by a mile. Remember, Linux is FREE and can't compete on the desktop with stuff that isn't. Stuff that is generally either put out by an evil empire, or a fruit company
I stand firmly behind JWZ's statement: "Linux is free if your time is worthless".
99.9% of the people out there use their computer as a tool. They aren't interested in tinkering with it, or even worse, fighting with it to try to get something done. I don't want to have to screw with my computer for simple things any more than I want to screw with my car just to get to work. THESE ARE SOLVED PROBLEMS.
If you want to get Linux accepted on the desktop, I suggest you take a good hard look at trying to do even the most simple things that people do on other platforms every day.
-play music
-configure a printer
-move files around the network
-play games
-adding/changing hardware
-etc
These are simple and linux still has plenty of problems with them.
Play sound? ALSA or OSS or something else. Barf. And doing something like adding a new graphics card? *shudder*. The ensuing Xwindows pain in the ass makes me sweat just thinking about it. "You've bought the latest wifi card? Oh sure, you just need to grab this patch, patch a kernel module, recompile, reconfigure, modprobe, and you're set!!" OH COME ON!
If the only way to set something up is to edit a config file, YOU FAIL. Period. This is not open to discussion. You will not win on the desktop. When my mom/dad/sister/grandma calls, I can walk them through GUI's to change settings. I can't/won't dare have them editing some random file in
I contrast this with windows. A recent event for me: the onboard firewire port on my laptop died. I bought a firewire PCMCIA card. I plugged it in. Windows detects it, finds the drivers, installs them, and is done. I plug in my iPod and up pops iTunes, it syncs, and everybody is happy. IT'S SUPPOSED TO BE THIS SIMPLE.
And i'm not even going to go into the KDE/GNOME/whateverdesktopyouchose wars and inconsistency nightmares that are going on.
(And even more, i'm not going to go into the incompatibility issues with the different distros and system configurations. AIEEEEEEEE)
It's not an exposure issue. Linux still isn't there yet. If you think it is, I suggest you go over to an usergroup like mythtv-users for people setting up that particular media server. You wouldn't belive the fucking nightmares people have to deal with for things like getting digital audio working out of their box.
THIS IS ALMOST 2006! PC AUDIO HAS BEEN A SOLVED PROBLEM FOR A LOOOONG TIME FOR THE REST OF THE UNIVERSE.
To replace windows Linux doesn't need to be as good, it needs to be BETTER. I'll tell you I'm perfectly willing to pay $100 for windows/MacOSX for the shear amounts of headaches it saves me vs. trying to run linux on my desktop (and yes, I have done it). If you think paying $50-150 (arguably, whatever the OEM cost is) for windows over a 3-5 year windows product cycle vs. the amount of trouble it saves you is too much money. Well, I'll just have to say we must live in two very different worlds.
I am soooo not a microsoft fan, and I think a lot of their user interface work is junk, but they **still** beat linux's ass when it comes to general usability.
What's to speed up? It's an audio player. I can't remember ever having speed problems with my audio players.
Then why does Winamp sometimes take 10 seconds to launch on my windows machine? Startup time and time spent searching for a specific song can and should be optimized.
Hi Furious,
You certainly don't come across as flamebait, and everything you said makes sense. An incidently, yes, I was pushing the blog a little, but more because the fist 20 or so posts were just gut reactions to the headline instead of thoughtful responses to the article.
I agree that Linux needs to come a long way before if gets even simple computer needs to work flawlessly like they do with Windows. The lack of a non-hacked way of getting USB WiFi cards working drove me crazy for many a night, and sent me back to Microsoft's arms several times.
As for the pre-install not being the cure-all I make it out to be, I agree only partly. One would hope that the vendor of the PC would get all of the driver stuff sorted out before shipping, so our respective horrors would be unlikely to be repeated. The application support would give new users pause, but I think that between OO.o, Evolution, and Firefox we have the basis for a good mainstream system.
The conclusion in the article, however, is that the open-source community should think about supporting ReactOS (in a substantially more mature form) as a way of getting onto the desktop rather than try to shoehorn Linux into the existing Windows-friendly infrastructure. That is that conclusion that I haven't seen anyone mention in this forum, and why I asked them to read the article. :)
"It is not advisable, James, to venture unsolicited opinions."
The drivers are great if you buy supported hardware.
"Linux isn't free; it costs $150 to replace my scanner." Those who rely on donated hardware cannot choose to obtain supported hardware over unsupported hardware. Those who are switching from Windows to Linux on paid-for hardware have similar problems.
There's a lot of hardware that won't work well on Mac either, but you won't find it in Apple's store.
For peripherals compatible with Macintosh computers, I can look on the front of the box for "Compatible with Macintosh computers" or (better yet) a certification mark on the front of the box. I haven't seen any penguins on home PC peripherals available at Best Buy or Circuit City. Which stores in Indiana should I be trying?
Perhaps you need to switch to linux and try VLC? or maybe even Amorak for the same speed.
Calm down people, its a religion not an operating system.
As someone who's not particularly adept with Linux, but has attempted to use it many times over the years, allow me to say that this may be part of the problem. Linux is absolutely nowhere near the cusp of acceptance for mainstream desktop usage, and for good reason:
For starters, drivers. Rarely, if ever, have I installed any flavor of Linux (starting with Slackware back in '99, having since used Redhat, Ubuntu, Mandrake, Knoppix and SUSE, not necessarily in that order) and had everything work. You need to futz with obscure config files to get something as simple the mouse wheel working, much less buttons 4 and 5. Video drivers are rarely up to snuff; as I've had ATi cards for the past few years, I've yet to even play Chromium BSU. Sound? Forget it. Basically, and I think this is the single biggest issue, virtually anything requiring a driver in Linux is a hassle. No one wants to spend hours pouring over forums and HOWTOs to install a bloody driver.
Then, there was the package dependancy hell, which has been somewhat resolved by package management systems. However, my experience with these systems has been that they're unbelievably unintuitive, and have an awful interface. Take Ubuntu's system, for example: it's 2005, yet its interface (at least when I last used it, maybe 7-8 months ago) looks like a circa-1990 BBS.
On top of it all, there's the hideously outdated UIs. There's little, if any, consistancy between apps in appearance, and most of the default themes I've seen in the various Linux distros still look like a clusterfuck of a Win98 box. They don't even match up to WinXP's level of consistancy and polish, much less OSX's.
Linux really does have the functionality to put it on par with Windows and even OSX in a lot of cases. The problem is that Linux is, by and large, an OS developed by hobbyists and developers for hobbyists and developers. Its level of polish is orders of magnitude off from Windows, and not even on the same plane of existance as OSX. It's just a hassle to install and configure, and not particularly nice to look at. Sure, it's less of a hassle now, but it's still just not good enough.
At this point, I honestly don't see what point - other than being free of cost - that Linux on the desktop serves. Sure, more competition is always welcome, but Linux is already a phenomenal medium/heavy-duty OS - does it really even need to be on the desktop, too? And more importantly, without a serious overhaul by a group of artists and GUI designers, does it even have a chance? My guess would be, on both accounts, no.
Perhaps you need to switch to linux
Buy me a new flatbed scanner and I'll try it.
I use linux exclusively but there are huge features I miss from windows.
Note: Many of these gripes could be redhat specific, it's all I've ever used.
1. File associations
Every app seems to have it's own method of determining what to use when I want to open a file. (The three big ones for me are firefox, wine and kde). I'm probably missing something here, but I haven't been able to figure out a way to make them all cohesive.
2. Multimedia
2a. Video
Tried compiling mplayer or xine? And have them work with the majority of the codecs out there? And get it to run on your laptop display fullscreen without dropping frames? And get sounds from your IM app to play while you're watching something? WTF, this should not be this hard.
2b. Audio
It may just be my config, but I had to manually add a software mixer in alsa to get my IM dings when playing any sort of music or movie. Add to that, my work voicemail system uses a GSM codec that I had to write a shell script to filter through sox and then a player in order to even hear it. I wouldn't even know where to begin if I wanted surround sound from a dvd.
3. Corporate groupware
I hate exchange and notes as much as the next guy, but they or a [supported] comparable solution needs to be available without wine.
4. Firefox (or another browser) needs to be able to render any site IE can as IE does. If you're running linux exclusively and you need to, for example, do online banking, switching to IE 'really quick' isn't a viable option. It does not matter that those sites are in error by supporting IE only, that fact doesn't help me reconcile my checkbook any easier.
5. Wifi
This is just a PITA for me for some reason. Is there a simple app out there that will let me scan for APs and connect to one without going through 8 different apps?
6. Fonts (minor)
I miss in windows the wonderful Arial font that would always seem to be 1 pixel wide everywhere. In linux every font looks thick and just... I don't know, too thick. Java (jedit) seems to accomplish it even on linux, so I'm probably missing something to get it working in kde as a whole.
That's all I can think of at the moment. Games are a biggy, but that's always brought up. Some people seem to dislike oo.org but I don't have any complaints with it.
Linux may well be ready for a completely "average" user, who doesn't use any unusual hardware and doesn't need any unusual applications - but how many people do you know who run a completely average system with no special or out of the ordinary needs at all? Of these non-average users, how many are willing to learn how to hunt down libraries and recompile things to get their unusual hardware/software to work? Very simply, with Windows and Mac, you don't have to - they just work out of the box.
Cronjobs are never "simple fixes". 'chmod -w' is much easier. Anyway, the desktop is doomed as long as you have to edit anything (via a gui, but especially by hand) to get simple peripherals to work.
"I tried to rip a music CD on my Fedora Core 4 system the other day [...] This is why Linux still isn't really ready for my desktop."
:)
Well it saved you from being a pirate.
Now... I'm happy with SuSE/Linux and I cannot ever seeing myself turning back, but I'm a nerd that enjoys the occasional hw/sw challenge (something I've not had on a windows box in probably 3+ years). But for Joe Sixpack? We (as in the all of us, or the royal... take your pick) need to bring Linux's usability up past Win95, because in my opinion, that is exactly where (SuSE 9.3) Linux is currently at.
The thing is that the issues you mention have basically nothing to do with Linux and are really only solvable by increasing the number of Linux users. It's not like Microsoft actually writes the drivers for most hardware, the manufacturers do that and (in theory at least) test the results and maybe even jump through the hoops to get them included in Windows. The only reason they bother is due to marketshare. Things are changing though, just in the last year or two I've seen quite a few companies with at least some nominal support for Linux. Ususally the Linux support is pretty sparce, poorly tested, and generally assumes that the user knows a lot more than would be expected of a Windows user but it's a lot better than it was just a few years ago.
That does sound like a pain, but to cope with it, you could setup a cron strip to check if it was wrongly changed and replace the file with the config-settings you want. Simple fix to deal with a problem.
Talk about a "ductape solution". In this case, yast should be able to identify changes not made by it and either leave them alone or warn the user.
No sig
Perhaps Winamp sucks?
I use iTunes in Windows (on a relatively slow machine, too) and it is extremely fast in all aspects of operation, from startup to shutdown.
This space intentionally left blank.
This is the single thing stopping me from being able to roll out pure Linux desktops to many of my clients. A replacement for everything that exchange offers on the client and server side is what throws the wrench into the gears. Yes, there are servers that can replicate parts of what exchange does. Yes there are clients that offer many of the things that Outlook does when connected to exchange. But there just isn't yet a unified client-server solution for Outlook-Exchange that can be deployed to the common-desktop-plebe that allows them to start working and communicate on the day after orientation that Exchange offers. Any secretary can start using exchange to schedule meetings, send out emails on behalf of people, and do the generic office communication thing with Outlook-exchange. There just isn't a total-opensource solution that does this yet.
I rah-rah Linux and open source like the rest of you, but 'best tool for the job' trumps my ideologies every time. Give me Open Outlook and Open Exchange and I'll be your Champion...
(excuse my typos and ramblings, there's an empty bottle of goldschlager next to me and I think I might have my Exchange priv IS down to a manageable level again)
It needs mainstream desktop users.
-ducks-
Users need Linux to run the software they can't get for free, which requires commercial software developers to write software that'll run on Linux. It needs to be easier for more commercial software developers to support both Windows and Linux. Open source makes a very good desktop, and there is a heck of a lot of good open source software, but some types of software only thrive in a market environment. This includes games, niche software, and any software that average users are willing to pay for but average developers aren't willing to write for themselves. Suppose a very cool game will cost $1 million to develop. There is nobody who will need to play that game so badly that they'll spend $1 million to have it developed, so such a game simply can't happen if it has to be open source. But for a commercial developer, it's worth spending $1 million to develop it if they think 50000 people will spend $35 to play it. For most commercial developers, the cost of supporting both platforms outweighs the benefits, or they think it does. The only ways I can think of to get commercial software developers/publishers to support Linux are: 1) Reduce costs/difficulty of porting and/or cross platform development. 2) Increase market share (chicken and egg problem). And 3) Evangelism.
Course this would be moot if there were linux drivers, right? But as you also mentioned, thus far many Linux drivers assume that "you are at least this smart" to install it (my Linux ATI drivers, for example... So my latest kernal didn't come with the source code... how do it get that...). I suppose more users like me (nerds who enjoy the odd hw/sw issue) asking for Linux support then it'll get better, but there are things WE can do as well I believe. Starting with nicer application names (what the hell does "NDIS" mean anyway!? And "abcde" is a what? Oh, a CD ripper, I guess I should just know that, huh?).
"1984" was ment to be a warning, not a guidebook. You hear that Kim Jong-il!? BushCo?!
I want an easy to use programming framework integrated with the IDE. See MFC, ATL, WTL. I want linux to be faster than Windows XP on my PII 400Mhz. I want a community of code like CodeGuru and CodeProject.
My condolences. I know this doesn't solve your immediate problems, but you might want to consider a move to Mandriva. The paid packs include all the drivers (ndiswrapper, etc, and deal with them automatically), and the *drak tools offer all the convenience of YaST without over-writing hand-configs. Plus, I offer free email/IM support to all Slashdot-reading Mandriva users.
U.S. War Crimes blog. Email for free Mandriva support.
because Linux is just the kernel, and the main problem with the kernel is lack of hardware drivers, which Linux has more of than SkyOS and Zeta have. The killer app and Windows compatibility are also more likely to come on Linux because its user-base and corporate backing are already larger.
In desktop distros like Mandriva, the home folder is already the default view of the system, meaning everything the user sees is directly relevant to him/her.
So it sounds like your "Linux" gripes are really Desktop gripes, and GNOME is already moving in the directions you propose. If you think they need to move further, why not goad them in that direction rather than suggesting that people jump ship, which will only exacerbate the problems you mention due to dissipating the critical mass necessary for driver development to happen.
U.S. War Crimes blog. Email for free Mandriva support.
Not all games are open source (see topic). I do agree that we need more commercial games that are easy to install and set up.
Ant(Dude) @ Quality Foraged Links (AQFL.net) & The Ant Farm (antfarm.ma.cx / antfarm.home.dhs.org).
Seriously, why would you care what anyone else runs on their desktop? Without commercial vendor support at this stage, Linux is not a mainstream contender. Open source has all it needs to be mainstream: Mac OS X. Tons of open source components sitting alongside commercial offerings, with users having very little tolerance for shoddy craftsmanship. Trying to get Windows users to jump to Linux just to reuse bloody *hardware* is stupid. Shove them over to Mac OS X instead and *then* maybe you'll be able to convince them that 100% free Unix-y goodness is a worthwhile move.
Puts a lot of "normal users" off KDE. I prefer Gnome anyway, but still....
Can I ask a question... Could the linux people go through EVERY SINGLE HOW TO and see if they can replace it with a batch file?
Here are some basic things that are wrong with my very slick ubuntu system:
1. Sound doesn't work without massive fiddling. I think I happen to have gotten unlucky in this, but it took me a huge amount of effort to get sound working. I am not sure what I did to get it working. I think it was one of several support libraries that wasn't installed. Moral: don't unbundle. Throw in everything the user needs. If you are a moby geek, sure, go for a slimmable distribution, but if your target is the other 95%, it has to Just Work, out of the box. So sacrifice disk space for functionality. X has to successfully probe the monitor and correctly identify the modes that it supports, as well as its physical dimensions; when a new monitor is plugged in, X has to be able to cleanly identify the new modes that are available, and support multiple monitors, and all that crap that Windows and OSX Just Do, completely transparently. Because Xorg is so dependent on static configuration, if something blows away the magic config you put in xorg.conf (which happened to me recently), you're in for an hour of hacking on the part of a serious geek to get it working nicely again. Most people are simply going to wind up with a configuration that isn't optimal, and not know what to do about it. Impression: linux is ugly. It isn't if it's configured right, but it's hard to configure it right. My linux screen looks really nice now, but it took a lot of extremely geeky fiddling to get it that way.
2. UI is massively inconsistent, and massively clunky. You want a person's first experience of Linux to be "wow, this is a lot easier to use." If it's "wow, this is a little funky," then they're going to stick with Windows. The 5% that are running Linux are early adopters, and they're willing to suffer to be on the bleeding edge. Most people aren't early adopters; for them it has to Just Work. Say what you want about Windows - after it's installed you're going to be sorry - but it works out of the box, for the most part, and when it doesn't, it's a matter of downloading a few drivers that install easily with installshield. Linux is better technology under the hood, but the usability isn't there.
3. Consistency. My laptop moves around a lot, and peripherals change a lot. My trackpad doesn't work if I start X at my desk, because I have a trackball and keyboard at my desk, and these throw off the device probing. You hear a lot of stories like this. I put my machine to sleep, and nine times out of ten it comes up with a blank screen and I have to hard boot it to get it back; the other time, it works fine. Things sort of work, but they're fragile. If something works, it's got to keep working. This kind of inconsistency is just not something anybody but an early adopter is going to accept. It looks like the problem with X is that it's simply not probing APM correctly, which is because my system does ACPI, not APM. The X wizards probably already have a solution for this, but it's not on my running system, Ubuntu Breezy, so it's not helping me.
4. None of my data transferred over (I switched from Mac), except for IMAP email, because that uses IMAP. All my address book information is stuck on my Mac where I can't use it. My calendar is on my Mac too. There's no interoperability, nor even a way to transfer the data over once and leave it. Given that there are standards for exchanging this data, it ought to be possible.
5. The sights are too low, so even early adopters are underwhelmed. Linux doesn't try to do anything new - it just does what MacOS X does, only not as well. Under the hood it's about the same as OSX, and much better than Windows, but from the user's perspective it's not as good as either of these two competitors. It's hard to compete, because Windows and Mac are both single corporate messages, and Linux is a free-for-all; both its strength and its weakness.
You'd think that free word processing and stuff would make a difference, but people would ra
1) A single graphical user environment. No more fights between KDE and Gnome, lock the two sets of developers together in a room and don't let them out until they produce a single unified product.
2) An easier install process. Windows users just click on install.exe or setup.exe but Linux users have to fight their way through apt get, rpm package management and unpacking tar files using obscure command lines. There should be no need for a user to manually unpack then compile source code, this is the 21st Century.
3) Decent power management on laptops. The state of ACPI under Linux is a disgrace and the developers responsible should be taken out and shot.
4) MS Office for Linux. I hate the evil empire just as much as the next Slashdot reader but MS Office is the standard, accept it.
The more corrupt the state, the more numerous the laws. - Tacitus, 56-120 A.D.
A more intelligent dynamic linking system is good.
I would settle for being able to tell which libraries are optional and which are part of the OS proper.
On a GNU-Linux system, checking the OS version gets you a kernel version number plus some distro-specific gobbledygook-- almost meaningless. It doesn't even tell you if a GUI is present never mind what 3D capabilities exist. And Linux distros don't even identify to programs whether they are LSB compliant!
The lack of committment to a standard for desktop functionality manifests as strangeness and problems to coders and end-users in many ways, both large and subtle; They get scared away from a landscape that looks like rapidly-shifting sand. As much as I like APT and Debian 'universe' as a power user, it is not an appropriate software management model for a personal computer. RPM-Yast-Urpmi-whatever doubly so.
Just like websites check for "Mozilla 5.0" compliance, basic Linux GUI/game programs should be able to check ONE dependency like "LSB 3.0" and have done with it.
Why must my package manager check to see if it has to update/install some of the most basic components on my system everytime I install a chat program or a calendar??? That should not be happening! The install process for any user application should keep its nose the HELL out of my system folders. I don't care if its APT doing the legwork. Except for system updates intended as such, leave my system libraries alone dammit.
Maybe we approach it from the wrong angle.
In my opinion, most people aren't really qualified to be using computers at all. And most that buy them don't really know what to do with them beyond surfing the web and getting/sending email.
Certainly it's a very small minority who are actually qualified to maintain their own computers. Hence the rampant virus issues with Windows boxes.
So given that, a) only people who need/understand computers should be allowed to use them. b) people who need computers but don't know what they are doing should be paying someone else to maintain their setup. c) Everyone else can have remotely managed set top boxes or something to get their email and use Google.
In such conditions, almost like magic Linux is ready for the desktop. Viruses cease to be a threat. The vast number of people working tech support will be relieved of their stressful jobs. Electricity usage will go down. There will be a culture of "paying for support" so we can get our leisure time back instead of maintaining family member's computer's for free. We won't have to talk about "can grandma use it" we can instead talk about "who let grandma touch a general purpose computer?"
Who's with me?
-- John.
This is the way Xandros' Control Center modules treat manual changes to xorg.conf and other files: A warning box pops and and you have to supply the root password in order to change those settings.
1. File associations
/etc/mailcap or /etc/mime.types file!
Edit your
No! No! No! This is one of the biggest problems. It is absolutely unacceptable that one has to edit a system config file manually just to get some file association stuff to work! It should be possible by simple clicking, and NOT EDITING CONFIG FILES!
This sig does not contain any SCO code.
If a user has to go through HOWTOs to know how to edit obscure config files you know something is wrong. Golden rule: the user must not be forced to dive into config files. NEVER. Too often a simple question like "my printer (model XYZ) does not work!" ends up in "type find -name balau848$""U(" -rh [{\48 20} and then edit /etc/blah/abc/xx__jht/rtkjc, check lsusb, copy the XYZID, check in /proc for bus ID 409482....."
Provide a GUI for EVERYTHING. And provide a good, self-explaining GUI. Rule of thumb: if the user has to look in a manual, the interface design failed. An exception are very complicated applications - there, you need a manual. But then, write a GOOD one. manpages are *NOT* good for non-gurus.
Also, distros should have something like autopackage pre-installed by default, to m make decentralized, easy setup files for linux possible.
This sig does not contain any SCO code.
... that people have Linux on thier desktop? As long as people can accomplish thier tasks with any given OS why the need for change? If its for the better of the world me thinks a cure for cancer or AIDS is more along the lines, what people really need. So why is it so important that they have Linux on the desktop, and not Windows or Solaris or BSD?
ze dog has no nose
Pre-install only solves the HW problem up to the point where I plug in the new peripheral I just bought and discover it comes with a windows and mac install cd. Go to any PC store and count how many devices have a Linux (any version, never mind all versions) install CD. You'll be lucky if you find a single one. Windows doesn' support most hardware out of the box. They have a selected set of key devices that were available at the time that particular revision of Windows shipped. Everything else, and everthing new requires a driver and suppoting application software direct from the manufacturer. The Linux people have done an amazing job getting a lot of HW to work, but really the numbers are not on their side.
Microsofts Media Playe regulaly looked up my computer and I would consider that a serious perfomance problem.
To be mainstream it will have to lose many of the features that make it attractive to tech enthusiasts.
If you want a very nice and easy to use *NIX desktop, head down to the Apple store.
If the big vendors shipped linux pre-installed, it would sell. Right now, Linux distros are easier to install an use than MS Windows, but that's a big psychological hurdle for people to get over. If they install it themselves, it's too adventurous. If it comes pre-installed and preconfigured, then they know that what they have is how it is supposed to be.
It's not as much a chicken-and-egg thing as M$ would benefit from. M$ has, and still does, play hardball with the vendors to ensure that competing operating systems and even competing packages and services don't show up on brand new systems. The chicken-and-egg problem show up in getting the vendors to break loose from M$' grip. One or two alone won't or can't do it -- it'd be too easy for M$ to retaliate like it has in the past. However, if many decide to do it at once, then they'd all hop on the band wagon.
I'm not sure what the first step in that direction is. Prying the contents of the contracts and non-written threats between M$ and the vendors out of hiding and getting them into daylight for public scrutiny would be a coup. But years of court action hasn't made any progress in that regard, maybe there is another way to solve that problem. Personally, I'm expecting the MIT $100 notebook to break the ice, assuming the M$ political engine can't crush the project first.
Beta is broken and the link to classic doesn't work. Stop wasting our time or there won't be anybody left here.
Amen - Too often, 'mainstream' is equated with 'good'. This kind of lowest-common-denominator thinking has messed up many products:
1. Honda Civic, used to be very popular with motorheads for its easy-to-mod-and-customize frame, Honda saw how popular it was getting and changed the body design to try and attract even more buyers (middle market buyers) and ended up killing the features the hobbyists loved. They moved. Honda is now trying to remedy the situation.
2. McDonalds vs. Local Burgers - I live in Chicago. Show me anyone who would take a Quarter Pounder with Cheese over a ButterBurger from Carvers and I'll show you, well, a person with no taste.
3. Financial Services - one size does not fit all. Every business is different, all have different needs, suggesting that all require 'the same' corporate or financial structure is wrong on its face. IT is just as complex, if not more so, and requires a similar level of customization _in order to function properly_ (emphasis definitely added)
I use *nix. I love *nix. I want it to be popular, but I dread the dumbing down that would come with market share even approaching that of Apple.
Besides, who loses if *nix stays small? No one. Those who can use it, will. Those who can't will pay the price for their insistence on using a lowest-common-denominator tool.
Here's a track record:
We're just starting to learn this how to talk someone really wants thi instead of were it only better then maybe someone would come to want it.
Make Linux (or any other OSS alternative) look and feel like WIndows, same "plug and play" convenience, and people will buy it.
I know car analogies kink shashdotter colons, but it fits in this case: The average user wants a car that is simple and convenient to operate. They do not want to have to open the hood just to install a new air freshener or jack up the rear end to install windshield wiper blades. That might be fine for the car tinkerer, but not for the average user.
The average user is accustomed to Windows. You cannot change that, but you can change the way alternative OSS looks and feels. Make it more like a family sedan or SUV than a dirt track racer.
Ignorance is curable, stupid is forever.
...when you pry my amaroK from my cold, dead hands.
Have you used amarok? It's everything winamp, foobar2000, and itunes were meant to be. It doesn't just rock. It dominates the other audio players.
[an error occurred while processing this directive]
Games are the only thing I reboot for, but for most people it's not just games. Go to any computer store, you'll see photo manipulation wizards, home and landscaping design programs, genealogy databases, trip planners, video editors... it'll be decades before there are good open source equivalents to all the most popular commercial software, and until that happens the only thing that will make Linux usable for many people will be a Win32 subsystem that runs all that existing software out of the box.
.NOT Invented Here API and moves the goal posts again) to make Linux a mainstream OS.
Yeah, yeah, I know, "Emulation killed OS/2 somehow!" "All of the hordes of commercial software companies making Linux programs will stop, and I won't want to run Win32 or Winelib builds of their programs for some theoretical reason!"
People don't buy computers to run operating systems. They buy them to run programs, and if enough of the programs they want to run require Windows, then they'll choose Windows. I've been running Linux on the desktop since 1997, I'm posting from Fedora Core 4 right now, and I had a moment of false excitement when I got Starcraft working in a properly tweaked Wine install in 1998, but here we are in 2005 and I *still* have to reboot to play a years-old game like Baldur's Gate 2 without bugs. I'm not trying to complain about the Wine developers - the fact that they've been successful at all (Deus Ex now runs better than on Windows XP, hooray!) is amazing. But it needs to go further (and it needs to do so before everybody moves to Yet Another
OK, let's take these one by one:
OEMs -- Is supporting Linux on an "equal footing" going to be cheaper for OEMs? Does Linux do anything to sell OEM computers? Does it make it easier for consumers to rip MP3s or download video from the internet? Is it flashier or easier to use?
Business -- Is there anything about Linux that makes it easier to setup a filesharing network than Windows? Single logins? Calendar and 'knowledge' sharing? How about developers? Are there good RAD development tools? Is it easier or cheaper to build Linux apps?
See, the answer to all these questions is basically NO. You can blame other people's biases and intertia all you like, but until Linux provides some obvious upside to those customers, the intertia will continue.
It seems that the Linux world decided that the Linux Desktop was inevitable without ever bothering to figure out why it would be or even should be.
Whenever I hear the word 'Innovation', I reach for my pistol.
Quickbooks is already ported to OS X, so it's most of the way there already. That ought to be an easy choice. Who wouldn't want a business system immune to the viruses, worms and trojans out there?
Even if the counter that worms, viruses, and trojans will appear as the linux distros gain marketshare, early adopters still save. However, the internal organization of the operating system, especially privilege separation and least privilege, makes it much more resistant. The systems can be further harden by putting things into different partitions and mounting the r/w partions (where there is data) noexec, and mounting the partitions containing executables read-only. Anyway, I digress...
Most businesses won't miss the absence of games and will appreciate a low maintenance platform.
Beta is broken and the link to classic doesn't work. Stop wasting our time or there won't be anybody left here.
1. cmd blows. bash rocks. I can't stand the Windows shell. You can't get anything done, and it is not well documented.
2. There are certain applications which are just better on *nix. TeX to name one. MiKTeX has come a long way, but font installation is easier in teTeX. ghostscript, sed, awk, grep, etc. Gotta have those things. Cygwin is ugly.
3. netfilter is better than Windows firewall.
4. Not as many worms and viruses affect *nix.
5. *nix has much better backup capabilities. dump, tar, etc. How does Windows help you to make backups?
That being said, 97% of people:
1. Don't give a shit about shells because they are content to not do better than mediocre productivity.
2. Are ignorant about how bad their documents look in Word and don't know how easy LaTeX makes good document preparation. As Oetiker put it, "your hamster might have some difficulty grasping the concept of text markup".
3. Windows users seem to be content with bad security.
4. Are so stupid that they will execute any code attached to any e-mail.
5. Don't do backups.
OS X already does all of the things Linux advocates have been said we need to do for years. Look at OS X's market share. It just doesn't matter.
Agreed and agreed. Linux is not inevitable on the desktop, at least not in a timeframe that is meaningful.
The point of the article was that we should embrace one of the open source Windows environments (such as ReactOS, in a more mature form) as a way of establishing OSS on the desktop. This would provide value to the OEMs (they would have more leverage with Microsoft in OEM license pricing and would be able to lower the costs of PCs) and would give the corporate users an answer to Licensing 3.0 that doesn't involve paying MS for software on a subscription basis whether or not MS decides to ship something each year.
Of course, none of this works if the OSS Windows software isn't up to snuff. ReactOS's stated goals of device driver and binary application compatibility would certainly help the two biggest gripes I've seen here with Linux though.
"It is not advisable, James, to venture unsolicited opinions."
What Linux on the desktop needs most is for those building/using Linux to understand that if someone wants a program/device/setting/etc to work or be modified a laundry list of cryptic commands is neither easy, simple or appropriate.
Time and time again I will see someone point out a issue they are having with Linux and someone will respond with talk of editing (by hand of course) a number of config files and end it with "It's easy."
But of course, that's just my opinion.
Wiwi
"I trust in my abilities,
but I want more then they offer"
thats weird... I am running on ubuntu and I had the same problem... but now its works great. Contact me at no1kissfan@bellsouth.net if you ever change your mind :)
The answer is simple, try to get as much support for hardware as Windows has. I would consider myself at least a "power user" and I have yet to even touch Linux because I don't wanna have to look through a checklist everytime I buy hardware. I went through that crap years ago with windows 95 and I'm not about to go back when I can, alternatively, simply plug in my device on WinXP and it will work flawelessly after I install some drivers from a disk usually included with the device. If the Open Source community wants to get serious about open source OSes, they need to evaluate their relationship with hardware manufacturers.
Judges and senates have been bought for gold; Esteem and love were never to be sold.
1) Decision makers needs to be educated to make informed decisions. SuSE openexchange is equally as good as exchange. Lotus Domino has a way better track record on both features, interoperability and stability - the fact that exchange has equal marketshare despite being more expensive clearly proves that lots of people are not making informed nor smart decisions on platforms.
2) Tools for easy corporate roll-out and management of linux desktops needs to become mainstream (no-one with more then 50 desktops are going to go the manual route for everything). This can be done by going the novell route, but that negates most of the licensing savings you would otherwise get.
3) GUI's and wizards - sometimes a person who doesnt know everything about X, will need to change or setup X and then want to be able to call someone if something fails. Thus this person will have knowledge to figure out the gui, but not necessarily want to spend the time figuring out the app/systems specific config.
Fix those 3 things (and 1. is definitely the hardest) and corporate roll-out will speed up.
Applications missing from Linux which would enable desktop use are:
The concensus was that Linux is great for fixed-use applications like ATMs and airline kiosks. It is good for technical workstations (mainly used by nerds and geeks). And, it is good for transactional workers such as point of sale, bank tellers, etc. Linux is OK as a commercial desktop and it sucks for laptops and mobile professionals (poor power management, wireless, interfaces to mobile devices,
Check out the full survey results at:
http://www.zoomerang.com/reports/public_report.zg
Essay responses in the survey can be seen at:
http://developer.osdl.org/cherry/dtl/survey-repor
The desktop market is not going to be an easy one crack. I must say that I agree with most of the comments that innovation in the desktop space will win over any attempts to copy MS. Linux has been playing catchup in the desktop space for too long. Both innovation in the desktop and killer apps are needed to move grandma off of her Windows machine.
The Linux desktop needs exposure more than anything else. It's already got some pretty cool desktops with some cool features that I don;t see on Windoze desktops. But few people other than the adventuresome who delve into Linux actually know about them. Most people buy into the humdrum claptrap of "Linux ?? Great for the server but not the desktop". Or "It's really really great but still not enterprise ready".
Linux desktop needs exposure. Most people like it once they see it and get to try it. Some people even ask me "hey, how can I get windows to do this ??". Problem is most people never get to experience the Linux desktop.
Most machines come with XP (or Hollywood's Longhorn in the near future) pre-installed. That is all most people know. the use what the computer came with. When it breaks they throw the machine away and buy a new one (again with XP (soon Longhorn)). They never have any exposure to Linux. They have no idea what it looks like or what it can do. Linux need Exposure !!
"-play music" - no problem. Works fine. Most distro's play music fine right out of the box. Don;t see any problems with that.
"-configure a printer" - never had a problem with that either. HP, Epson, Canon, etc. All worked fine. Never had to manually edit a printcap file either. GUI config tools work right out of the box. network printing ?? No prob, works great.
"-move files around the network" - No problem there either. Flip Gigabytes worth of files across my network everyday and over the Internet via Linux. No prob. been doin it fine for years and years.
"-play games" - Plenty of em. Both Linux games and Windows games under Wine. Want more games under Linux ?? Convince the manufacturers to start porting em to other platforms.
"-adding/changing hardware" - a cinch.
"-etc" - even easier.
Don't see any problems here. Linux is batting 1000+
What Desktop Linux needs to compete are the following:
...and THAT'S IT. Why? Because if we had hardware support, then The Problem of Installing Linux would vanish into thin air, OEM Linux would be a reality faster than you could say "America Online," and we could actually write those "Wizards" and easy "Installers" everyone wants.
1. Hardware Support -- Which, of course, the developers of free software have little/no control over. Until a hacker can sit down at their workstation with the spec to I/O of every major chipset, device, or peripheral, Windows will win every time, regardless of ANY other improvement.
2. Creativity -- Stop copying. Stop trying to BE Windows. BE BETTER. It's happening, I know... but we need MORE.
ABCDE A Better CD Encoder....
To achieve desktop penetration, open source needs one thing. A better metaphor. A new paradigm for how data is stored. Something to replace the file system.
We've oozed eye candy and UI friendliness onto the OS for years. Its no longer what applications, what features or even what games we do or do not have. Linux is competitive with everything else out there for the desktop field. But we're not going to win anything by growing our margin. There's too many network externalities in play such that making the linux desktop "better" will bring it mainstream. We need to leapfrog the world if we are ever to achieve a meaningful victory. Factor ten engineering.
We need to go back to the core, to the heart of the OS and consider where we can truly evolve. What is the purpose of an OS?: secure allocation of resources. Currently all of our resources are files, some files are privledged with having a single function call, these being called programs. Others may be privledged as containers, they hold other files. Anything past this is simply a construct of the application, layers built on top of this base. And this construct is imaginary, because the user will still have to interact with the base layer at some point.
I do not think this is enough. It is not enough. Maybe you can achieve this level of dataware through gnome VFS and KDE's KIO (i think that is there VFS?), but how will this interoperate with a client running Windows or Osx? Can we compell all our applications to be built with GTK? No, we need to rethink from the ground up how data is stored on our systems. Our data is active, alive, by itself, growing.
Right now, we have everything they have, and even when we've built ourselves twice as much, we still wont be any different. Linux got where it is by being disruptive. The only way it will ever continue to excel is by causing further disruption, not by playing the same games everyone else plays. You cannot out Microsoft Microsoft. There are two different metaphors that linux users confuse on a religious basis; the desktop and the desktop. One metaphor describes the feel of the operating system, the other describes a form of market penetration. The success of our desktop environment no longer has bearing on desktop penetration.
Myren
But wrt: "Sure, VLC does everything winamp and media player does, but what an ugly, but fuctional, interface (I'm only using vlc as an example)."
:) When watching a movie in Linux, I like that VLC and (Kaffeine?) -- and I'm sure others, too -- have nice slider bars to locate one's position in a given file, too. Maybe some Windows players do, too, but not the random one that came on my Toshiba laptop; the DVD player is bizarrely branded by Toshiba, too. Huh?
...
Hah! When I watch a DVD on any Windows machine, I always wonder if the player interface (I can't recall the name of the players I'm thinking of right now) will be even gaudier than the last one I used. They look (ahem) a bit like certain Enlightenment themes sometimes (which is no slag on Enlightenment, if you're into the ultra-matrixy / brushed metal stuff -- it's just that I'm not), with tiny, hard-to-read buttons surrounded by meaningless, distracting goo. VLC is just the opposite
Themes for VLC would be nice, though, I agree -- I'd just like them to be mostly simple and clean (art in function) rather than controls tacked onto roccoco background fluff, like a lot of XMMS themes are.
We might disagree about "bling" in general, but at the moment I use daily both Linux machines and a Windows laptop*, strictly for "desktop use" (word-processing, simple games, web, email, etc). When I get home and switch from laptop to desktop, I am always glad to be in the nicer, cleaner, *less* blingy environments (just about any of them!) that I can work in on my Linux box; going back to Windows for my "day job" always feels primitive.
timothy
* School requires Windows -- so I bought a laptop running it. Better than Windows used to be, certainly, but still
jrnl: http://tinyurl.com/c2l8yr / foes: http://tinyurl.com/ckjno5