Slashdot Mirror
Linux on the Tipping Point
Reader stormcoder wrote to mention an article on Enterprise Linux I.T. in which the author posits that even though Linux is built on a legend, the reality of Linux outstrips even the myth. From the article: "..the fact that Linux has traditionally been compared to Microsoft's Windows brand products and not the other Unix variants will most likely lead the general public to perceive all this as Linux sailing on to new horizons while Microsoft stalls out. This perceptual shift should totally reverse the previous mainstream view that Microsoft and Intel were somehow at the forefront of high technology computing -- thereby pushing Linux over the magic edge of a social tipping point."
All people want to know is if it can run their word processor, their games, and their stupid movies their friend send them.
That page was riddled with Microsoft Windows advertisements...
Really! It is!
No, really!
*cough*
Most basic home or small business MS systems ship with XP home edition, and you are in for a rude surprise the instant you try to connect it up to a home or office LAN.
My rights don't need management.
Tipping Point is defined in the article. If you didn't read it, don't comment.
People have a preference, I'm not bothered if more people use linux. It's good for me because of what I do, I'm not desperate to get everyone else using it.
Oh yeah, every year for the last several years. Examples follow"
March 2003
July 2003
November 2004
December 2003
If you need web hosting, you could do worse than here
Perception - although cosmetic - is the foundation of ideas. Ideas rule - not politicians. They are the office-boys of government. This change of ideas is where the world changes. Isn't it great to be there while IT is happening and be aware of it! Matt
It means that, in the near future, the corporate IT department will plan to deploy linux systems and windows proponents will have to work hard to overcome the assumption that their product isn't up to snuff, not the other way around.
It means that the low-end systems you find in the store will have a variant of Linux and Open Office, not a low-end version of Windows (e.g., XP Home) and "works."
For every complex problem there is an answer that is clear, simple, and wrong. -- H L Mencken
Kickass 3d shooter games and support for devices.
It consists of a Linux kernel developed by Torvalds and his colleagues by radically improving an earlier open-source Unix released by Andrew Tannenbaum in 1987 If he is implying Linux was is based on Minix, he is incorrect, albeit, extreamly early versions of Linux did use the minix filesystem and minix to bootstrap. In design, though, Linux and minix are fundementally different. Also I am sure Tanenbaum would disagree that Linux is a radically improved version of Minix -- as he is an advocate of microkernels.
Is there a torrent for Tipping Point? Does it have the current nVidia drivers?
There are many varied reasons that Slashdotters hate Microsoft. But this brings up one of my pet peeves about the corporation, the perception that it's on the forefront of computing technology.
I remember as a young lad downloading a copy of Slackware on 12 diskettes, installing it, and having a revelation as I realised that there was something better than DOS on top of which we could build a GUI and desktop system.
It was a thing of beauty, to see this brand new thing called Linux which came with source code (gasp) and made DOS and Win3.11 look like the crap it was.
Imagine my devastation when I started reading history and found out that Linux was just continuing a then 20-year tradition of open source, stability, and multiprocessing. Then I had to watch the slow decline and shittification (to coin a word) of the industry as Microsoft became more and more powerful.
It was depressing to me to watch as person after person suffered through BSODs, memory mismanagement, corrupted data, etc. whilst I knew that sitting right there on my HDD, with no marketing clout, sat the answer.
Microsoft is navigated by some brilliant captains. But they're brilliant sociopaths, consistently destroying everything that is Right and Good about our industry.
I found myself apologising to users for the lameness of the software they were using, and unable to really provide them with any alternative.
It makes me a really happy person to look at Linux these days. Thanks to RMS, IBM, Novell, SuSE, RedHat, and others (the non-sociopathic brilliant people (fuck you, SCO)), Linux is looking really, really good today.
I'm actually finally considering migrating my wife off of Win32 (she types Chinese, and Chinese input under Linux was pretty useless up until this month) and onto Mandrake 10.2b3! Milestone.
Linux rightfully deserves the title of being on the forefront of technology. Microsoft? They were holding us back.
fifth sigma, inc.
No it can't be considered a monopoly. For the simple reason that Linux is a kernel and not a company in itself. And there is no single company that controls the development of Linux completely.
Linux is well ahead of the latest Microsoft products and so shines in such comparisons. It is not, however, remotely a leading edge system in the same class with the BSD family of Unix products and Sun's Solaris.
The BSD family of Unix is very large some may say it also includes HP-UX. Though if you include the open-source BSD's: Free, Open, and Net BSD. I would disagree that it is not even in the same class as them; as Linux has far more commercial support and technical features.
Also some may consider Windows NT based systems to be part of the BSD family, or atleast its TCP/IP stack.
Either Windows is the standard and its emulation the ultimate goal or Windows is junk to be ignored. Which is it Linux? The only people who have satisfactorily answered that question are the fine folks over at Apple. They've forged their own path while Linux strives to be "more like Windows".
as the article states - is the effect of the new powercell chip from IBM. I, too, am eager to get my hands on a workstation that will be up to 10 times faster than anything from intel. Apparently linux is the only major OS with support for the chip at this time. No support is mentioned by microsoft for the chip - and I wonder, had MS ever successfully supported a non-x86 architecture for any real length of time?
Rob Enderle's excellent new book: Everything I needed to know about Computer Science I learned in Marketing School
Linux acceptance has been driven by its perceived political correctness in the mass media, itself an artifact of the legend of Torvalds
/. that is)?
Linux is politically correct now (outside
Fundamentally, everybody likes to be leading edge,
Geeks do. Businesses don't give a toss.
most of those who did replace a few Windows servers with Linux soon found that the software's quality led to much bigger benefits in terms of operational stability, support staffing and the overall integrity of their information systems.
Well gee, that's new. The fact is, small companies install and run Windows, and put up with the problems, because they just can't afford a "linux guy".
Most people agree that products like Sun's Java Latest News about Java desktop don't have as many features as Microsoft's integrated office suites, but people willing to give up some bells and whistles are finding the open-source products fully functional and free of the proprietary limitations built into Microsoft's products.
Because Java isn't proprietary?
As a result, the fact that Linux has traditionally been compared to Microsoft's Windows brand products and not the other Unix variants will most likely lead the general public to perceive all this as Linux sailing on to new horizons while Microsoft stalls out.
No, the general perception is that Linux is arcane, and Windows is kind of annoying but "easy". I'm talking about moms and pops' perception...
"A door is what a dog is perpetually on the wrong side of" - Ogden Nash
You already have Doom 3, and Half-life 2 with wrapper software. Dont know what else you are looking for?
mnewberg.com
When and where does MS advert most? At their own place, where the "sales" are slightly less than 100% (interesting, in that)? Or do you advertise at the enemies place in hopes of casting doubt on the opposition?
I prefer the "u" in honour as it seems to be missing these days.
I think the best possible situation would be if Linux is used on office machines, since it's so much easier to lockdown and centrally administrate than other mainstream OSes. At home, people would use Mac OS X, because it's much better at providing peripheral support and simple means for end-user administration. Either that, or Windows.
:)
That's a world I could happily live in.
If you open yourself to the foo, You and foo become one.
"On all reasonable measures of performance, stability, and technical accessibility, Linux is well ahead of the latest Microsoft products and so shines in such comparisons. It is not, however, remotely a leading edge system in the same class with the BSD family of Unix products and Sun's Solaris."
Not leading edge? I guess all those monster linux clusters are legacy then. So much for Gentoo... and so many other bleeding edge Linux projects.
Ignorance in journalism? Never!
He thinks that people are advocating linux because of hte legend of linus as a robin-hood like figure against the evil forces of big corporations, while saying that its not really true, linux was developed using alot of ideas from unix, with the empethsis in the article being that unix-derivied OSes are more suporior then linux (and therefor windows)
p pa.html before, when linux started it was really meant to fill a void none of the other unix-like OSs were, and downplaying linux as offering no technical advantages over BSD and solarix is pretty insulting and biased
His argument is that while everything beats windows, linux in particular is only known above other unices because it has the quasi-religious loyality.....i guess people can't argue again that fully, if microsoft decided to revive xenix and made it all open source and changed everything we know about them, would anyone trust them as leaders instead of linus?
Its ashame he didn't read http://www.oreilly.com/catalog/opensources/book/a
"Linux is built on a Legend" - Slashdot.
That will be $5 billion please. Unmarked, and no funny business.
The real Ralph Yarro posts as Anonymous Coward. Anyone else is an impostor.
If users who get their computer knowledge from fashion magazines see a change, it will be from Windows to a marketed, hip, device. The Mac being currently marketed to such segments almost as an "internet attachment" to your iPod makes it the most likely candidate for the fashionistas who are tech-clueless.
The class of managers and marketers who are tech-clueless and perceive the PC as a fancy addition to their typewriter or calculator may very well perceive Linux as the new and better upgrade to the PC.
One thing that's frequently downplayed is that Linux will run on both Mac and PC hardware. Showing off Linux running on these two hardware platforms will go a long way toward telling the tech-clueless that Linux goes well beyond traditional personal computers: both in abilities and in their lack of turf wars and platform exclusive tricks. (And if it helps enlighten a few politicians, judges, or patent attorneys about what an OS is, all the better)
Linux is the tech of tomorrow that can replace the dodgier tech that's being pushed today.
This was exactly the same way for servers back in 1999, 2000, 2001. In fact, the only ones to get it right was IDG and Gartner, when they proclaimed that Linux would have less than 1% of servers on the internet by 2005. And we all know that Linux on the servers have not gone well. Right? Oh wait....
I suspect that we have allready gone over the tipping point. It is now just a matter of companies such as IDG/Gartner to point out that they were wrong on this as well. Of course, their own income will plummet.
I prefer the "u" in honour as it seems to be missing these days.
It can also be a sudden burst of mass insanity as mainstream public opinion gets caught up in a new myth like the "new economy" of the dot-bomb era. Such a one-sided definition totally misses the point of how or why such change (for which "paradigm shift" was the previous buzzword) occurs.
Slashdot - News for Herds. Stuff that Splatters.
Linux has it right. Apple will forever be relegated to a niche market, and if that works for them, that's fine. However, the prospect of Linux is much greater. In order for something to appeal to the masses, who have been spoon-fed one particular operating system, you have to learn to speak their language. Once you do that, you can gradually coax them into your camp, but I don't think you'll get many takers if you insist that they speak your language. People are generally lazy - they'll take the path of least resistance.
Doesn't it take two hands to type?
I predict the following: Linux will never threaten windows on the desktop.
People want windows because:
1) They want everything to just work instantly. Windows does this, as far as the average user can tell.
2) They want access to all programs out there, just in case they ever need them. Just having web, email, and a word processor is not enough, because there is always a small chance that they might need to install something exotic two years down the road.
3) They want to be able to play games. Even my mother wants this, and she has only played two games in the last 10 years.
Sorry, but something really extraordinary is needed to even threaten Microsofts dominance, much less overtake it. Linux can hope to become as popular as Macintosh, but even that is very hard. Apple stuff is easy, remember?
It is very likely, however, that Linux becomes as dominant as Windows is on the desktop for everything that isn't a desktop (or has a gui). And this in itself is pretty awesome.
Will code a sig generator for food
I work in a rural Utah computer shop, and every day, I'm seeing Linux make inroads.
I've got about a dozen customers fully converted to it for desktop use. The downside is that these customers are no longer coming in every couple months for spyware and virus removal or other Windows fixes. Go figure.
MadOgre.com
The Tipping Point" by Malcolm Gladwell - check it out at youir nearest public library.
This is not an endorsement - I didn't much like the book.
Take the 90-Day Challenge! http://rwmurker.bodybyvi.com/
the tipping point is like a see-saw. Once you reach a certain height, you continue over (assuming that upper weight slides slight inwards, and the lower weight continues outward). The tipping point for Linux is when enough ppl accept it, implement it, and lower their costs. Good example is Walmart. They were one of the first to implement Windows. By the time that competitors did so, Walmart had paid for the conversion and was able to lower prices while retaining profits. Others had to lower prices, but also profits. Now, they are quietly installing Linux. By the time that companies such as Sears, JC Penny, King Soopers, etc decide to do so, walmart will have paid for their conversions and will simply lower prices while having outlandoush profits.
I prefer the "u" in honour as it seems to be missing these days.
My opinion is that people who pooh-pooh Linux are going to wake up and find themselves on the wrong side of the learning curve. The fact that MSFT feels the need to sling mud at Linux speaks volumes in itself. The way Redmond has taken to squeezing more revenue out of their existing customers in order to shore up quarterly numbers. They're sure acting like a company on the defenisve.
Apart from MSFT, all the really fun stuff in IT is happening in OSS. There's no way a company the size of MSFT can be as nimble and OSS is a more efficient development model. Building on the shoulders of giants. MSFT can't compete with that.
You mention "several years" but only have examples from two.
That's our life, the big wheel of shit. - The Fat Man, Blue Tango Salvage
Every year number of users grows for more than double.
I call that year of Linux. "Year of linux" term is not defined by M$ still being biggest (or how big share it has), but with increase number of linux users and servers (and in last years with profit that companies make with linux).
p.s. if new users would keep comming with this rate, people would have to increase their sex activity just to provide new linux users in about 20 years.
Signature Pro version 1.13.2-3 release 83.5 beta3try7 after-breakfast edition
Everyone I know who ran Linux for scientific reasons switched to OS X because it was just more polished and easier to install and maintain. Or there were people who couldn't get SATA drivers(and still can't) and actually went back to XP(albeit grudgingly)! So when is this type of story going to stop being a yearly promise and actually become a reality?
User friendliness.
Cleaning up the kernel some.
Device manufacturer acceptance. (Too many manufacturers don't make drivers available like they do for Windows and Mac. Downloading them is okay in the short term, but this needs to change.)
Windows has detected an undetectable error.
Linux was certainly based on Minix as a concept, a unix clone for 386 PCs, as well as FS and bootstrapping. I'm more bothered with the "open source" label. Originally the source code for Minix was available only as part of Tannenbaum's (copyrighted) book on operating systems. What made linux great was the "free software" concept, not the technology behind it. I also think Tannenbaum has been quite humbled since the microkerneled "Amoeba" failed to catch on, along with Hurd, but the OS he did openly criticize has become mainstream.
Also note the idea that Linux is derived from Minix. Of course anyone who knows anything about kernel architecture knows that it may have *inspired* Linux but certainly did not provide a codebase which was "improved" in Linux.
Furthermore, the only technical advantage I can find for Solaris is that it is closely tied to Sun's hardware, and so runs quite well on expensive systems. Having used Solaris X86, I have concluded that the OS is featureless and difficult to use in comparison to Linux. Similarly with all the other commercial UNIX types, you still see the fact that the main advantage which is sold is its tight optimization for specific hardware.
I thought the article was a bunch of crap, but then I have never seen the ideas he says are being superceded.
LedgerSMB: Open source Accounting/ERP
Linux is in exactly the same position, and the free software world needs to recognize this. Unlike Windows, Linux's monopoly was not created to exploit its users; unlike Windows, Linux's monopoly is not -really- exclusive (it is just Unix). But Linux's chief selling point is still mere compatibility -- with both hardware and software -- and because of the power of this inertia, Linux can succeed without actually being better. This is what makes Linux the next Windows.
Unix was a great system in 1970. It was a far, far better way to manage a computer than the most popular approaches at the time, and became successful because of this. It was a great contribution to computing.
Today, however, by all rational measures it should be obsolete. Nobody designing an operating system today would make it anything like Unix, unless they wanted it to be compatible with Unix. I don't want to get into specific critiques -- if you disagree on this point, then just ignore me. If, however, you see the myriad outdated approaches in the design of Unix, then you will realize the problem here.
Systems like Plan9 or EROS use designs obviously superior to Unix, and are destined to fail because of this, not in spite of it. If we do not figure out the problem here and fix it, we will be stuck with Unix for as long as it took to get rid of Microsoft -- maybe longer.
Ken Thompson said it thusly:
I'd have to agree.. all my nerd friends are at least interested in linux, whereas 5 years ago they thought it was too arcane and annoying and stupid. Attitudes are definitely changing.
The phrase "tipping point" is getting old, really fast. It really is the new "synergy", the latest Dilbert buzzword, or if you will "another hollow phrase".
It really is an interesting theory and I also think it works that way, but I also think people are using it too often..
What's next? The tipping point where people are welcoming our new ant overlords instead of fearing them?
This is why building a browser into the distro, rather than just using the "best of breed" or offering a couple options during install, it completely foolish to me.
I hate Konquerer, although it HAS gotten better. But I just hate having HTML rendered in my file browser.
Damnit, why does the Linux community have to copy every crappy assed idea of Microsofts?
I miss the days when KDE ran FAST and didn't have a lot of "prettiness". Instead of the community focusing on building a Counterstrike Killer, instead it's all about anti-aliased fonts (fine, they are nice, but is it that big deal?), bloated User Interfaces and a billion different distros.
Did I mention I hate Konquerer?
Begin modding me down. I'm used to it. (opinions on slashdot that don't fawn over the subject usually get modded down as trolls.)
Oh my gawd, they killed kenny's mod points!!!!
One line blog. I hear that they're called Twitters now.
I don't know of anyone who considered linux a significant technological breakthrough. Its claim to fame has always been how it was "free." (even by people who don't understand the issues behind software freedom.) The only thing this article contributes is a tie-in with the latest jargon of the day "Tipping point." BTW I agree that BSDs should deserve more attention than linux, but I don't think articles like this will help achieve that.
Jenna Jameson will show us the way.
I converted one of my workstations over to linux as a means to broaden my horizions. Not too long after that I got a call to finx one of my friends computers. After 3 hours of cleaning up malware of all kinds I learned that 1) He had a fetish for hot asian teens, and 2) I would only get this call again, since he didn't want to learn to use a newsreader.
So I floated the idea of him getting a mobile rack and putting linux on a cheap 40 GB HD. I selected Mandrake since I had grown to loath Redhat.
6 months or so later he uses linux almost exclusively. I don't have to say why. And he's bragging to his friends as work how awesome it is to surf for porn on linux, and how him computer isn't crippled by shit.
But the masses, they don't hear the kind of cloistered evangelism taking place on Slashdot. It doesn't even exist in their world. But the dream of malware free porn does. It might not be the high road, but the way is there. And it has nothing to do with the rambeling delusions offered up in groupthinkgeek pieces.
Linux will reach critical mass in germany within the next 12 months. Everybody with more than 2 braincells and some IT knowledge has predicted this for years. For more and more partners I work with it isn't a question of wether or not one should use Linux, but how to apply it. It's actually a thing I bet my business on two years ago and to date I haven't regreted it. :-)
MS Windows is done with. People allways call me insane when I say this, but even the most notorious Windows users here say they will migrate to Linux when Win2K support ceases.
It could very well be that MS will pull a publicity stunt and start releasing their own Linux Distro, with DX9, NTFS and all. They'd have to admit having done a big mistake, they'd be 3 years late, but I guess with 40 billion on the bank it's not such a big problem taking hold of the 15% growth FOSS IT services market in something like 6 months flat.
Anyway you look at it, the next years are going to be interessting and probably lot's of fun aswell.
We suffer more in our imagination than in reality. - Seneca
I think we just found the ideal antonym for "beleagured".
Two things:
1. Half-life 2 will probably look like ass since you can't run it with all the Direct X 9 stuff (you must run cedega in Direct X 8 mode).
2. He was being sarcastic.
Note to self: No more arguing with the faithful.
I don't think it's about perception, it's about fact. I think the real solid hard core fact is that we are in the information age, an age that is centered arround the unrestricted flow of information. Linux treates unrestricted copying like a benefit, MS treats it like a competitive threat. How can it be more black and white than that?
Society needs that free flow in order to move forward, failing to do so has serious economic and personal consequences that can not be avoided. I renember. That is, I renember all the businesses that blew off Linux, wanted to blow off Linux, tried to blow off Linux, but eventually couldn't because of pure economic, flexability, and resource truths.
I honestly think too many businesses saw "copyright" more like a genuine free market commercial property right that the intrusive government regulation on how people can use information that it is. Linux aviods the whole problem, and is truely a blessing in disguise.
Its P0rn you fool! Actually Email has always been the "killer application." I think fewer people are satisfied with word processing without Internet access. P2P and ipod/itunes use seems to have some traction too.
Sure, but you'll have to recompile the kernel...
That sounds like coming from a mouth of someone who could be a model of "user" as Microsoft sees it.
I am putting myself to the fullest possible use, which is all I can think that any conscious entity can ever hope to do.
A lot of people on Slashdot have tons of experience typing with one hand.
Most homes and small offices are just fine with XP home. Plug in, DHCP from router and you're done.
Heck, to add a network printer, just browse the network and double click - the drivers generally install themselves. WAY better than Windows 98 and W2K.
IMHO, at $100 for OEM, Windows XP Home is a pretty good deal for the average home user and small business. IE and MS Office on the other hand can be replaced quite well by free OSS alternatives.
Are you kidding? Pretty much every company I've ever seen is always jumping on the latest management or IT fad. Java, XML, Extreme Programming, Six Sigma, CMM, you name it. What programmer hasn't been told by their management that they want to XMLify everything or convert all the in-house applications to jsp webapps, whether it made sense or not? Companies are even more bleeding edge than your average geek is. http://linux.slashdot.org/comments.pl?sid=141566&c id=11860211
People have lots and lots of data in formats that are only supported by proprietary Windows software. For example, my elderly neighbor has megabytes worth of genealogical material (digitized wills, etc.) on her Windows box in a format that she can only use with the proprietary genealogy software she bought
http://linux.slashdot.org/comments.pl?sid=141566&c id=11860401
I'm just wondering - what distro have you been using for your users?
I'm putting together a few systems for some relatives and I'd prefer to put something together for them where they won't be back with spyware and viruses for a while.
Thanks!
"...Well, there's egg and bacon; egg sausage and bacon; egg and spam; egg bacon and spam; egg bacon sausage and spam..."
Many BSD enthusiasts don't include Mac OSX or even Darwin as BSD. Definately not NT. the NT design team was led by David Cutler who also did VMS. http://www.jmusheneaux.com/WNT.htm
A lot of us can't run it because... /usr/share/baoeu/otehu/ -x -die. Pressing a flurry of keys might feel faster, but it isn't actually faster.
- the software support isn't there. No CATIA, no ProE, no etc. Can't be an engineer using Linux alone.
- it is still fucking slow. Hate to break it to you, but as a long time xfce4 user, XP is still faster. Yes, doing everything CLI is probably faster as far as system response goes. But it's slower for getting things done because double-clicking an icon is easier than typing
And so, until linux developers wake up, linux as an engineering tool will not progress. Engineers already have a million pieces of proprietary scientific software they have to learn; adding linux to that isn't worth it when there is fast, and stable Windows XP.
And it's stable enough as a desktop, though not as a corporate server. Maybe that's because, in at least one thing, MS has its priorities straight. They've focused on making an OS that's sane to use while attempting to make it stable. And, lo and behold, it works.
still haven't gotten octave working with gnuplot. fuck it, i've got more important things to do. i'll do it in matlab.
I guess "catastrophe theory" was too scary, so they had to come up with a more politically correct way of describing that kind of transition...
How about:
Command and Conquer Generals: Zero Hour
Battlefield: Vietnam
Battlefield: 1942
Need for Speed: Underground 1/2
Star Wars: Republic Commando
Grand Theft Auto: Vice City
Sacrifice
While I understand these are not all shooters, this is a random cross-section of games that I play when the mood strikes me. The only option is transgaming, which *might* work and costs money. Frankly, I'm not paying more money to play games I paid money for on the idea that it "might" work. I've found their compatibility list less than accurate.
I remember trying to run Black and White on transgaming. They said "works with some minor issues". That translated in to: "Works, but you have to sit through the long intro video, which you can't skip, and is going to play at 2fps". I also remember trying vice city. Transgaming didn't even have a working mechanism to put disk 2 in. You had to image the second disk to the drive to make it work. What a joke. Gaming on linux, easy? I think not.
Either Linux needs better emulation for windows software, or software manufacturers need to start developing a larger selection of games to be cross-platform.
BeauHD. Worst editor since kdawson.
In my experience no operating system does everything I want it to do out of the box, and Microsoft Windows and MacOS X only come close when I install a bunch of third-party software that didn't come with the system. Also, this entire framing of the debate ignores the far more important issue of software freedom (and many would, no doubt, cite the circular argument that this lack of debate on software freedom is proper and right because the mass audience doesn't know about software freedom).
People's choices are narrowed to favor things which proprietors can cater to. The framing of the issue excludes alternatives to proprietary software, effectively narrowing the allowed scope of debate. Free software OSes don't ship with many personal computers, and they aren't advertised to the degree proprietary software is, and they are often behind in features. But they excel in delivering software freedom--something proprietors cannot deliver at all. In many venues there is no discussion about what the free software community has been working on for the past two decades. Sometimes, if free software comes into play at all, it is just used to drive down the cost of some proprietary software (I recall reading a New York Times report which talked about a Microsoft memo that said something to the effect of "Lose no sale to Linux").
Perhaps if we did a better job of teaching the importance of valuing software freedom for its own sake, we would see this values reflected in more people's opinions (much to the ire of software proprietors).
Digital Citizen
I disagree. I think if they can pump out Mac Minis and can avoid the production problems that plagued the iMac, I think they can make serious inroads. The real test will be when we start seeing serious advertising begin, which leads me to think that A) they learned their lesson from the iMac debacle, and B) they haven't figured out production yet.
It's easier to fix a broken windoze computer with Mepis than it is to reinstall windoze. You might even retrieve your wife's pictures with it and you will surely not have to spend hours digging up browsers, spell checkers, paint programs and all the other software that makes a Winblows computer usable.
There are enough people like me willing to help Joe at no charge to make it much cheaper too. I teach a Linux for desktop users at a local computer club and can say that interest in free software has never been higher and the migration has never been easier. They are able to teach each other once I get them going. The progression is geometric and Microsoft will never catch up.
Game over.
thats not true. almost every major distro comes with a modular kernel, and devfs or udev (hotplug+coldplug). how do you think installer cd's manage to use your network card, and your friends cd, when neither of you had to recompile the kernel to make the installer boot?
Macs aren't on the shelf most places, and neither is Linux. there's your answer. Apple has always been stupid about marketing, and there has been little linux marketing to anyone except large coporate users. Home users run what comes on their machines, they have never even had a decent opportunity to try something else out. It's not like going to buy some other consumer appliance with a lot of choice staring at you right there in the store. Sure, a huge variety of hardware, but all of it pre installed with MS leads to joe home consumer only ever using MS, so they put up with it. They honestly don't know any better. yesterday I was in a computer store, a local whitebox place, talked to two other customers while I was waiting for service, neither of them had ever used an apple product (although they admitted they knew they existed) and neither of them had even HEARD of linux. Absolutely no idea what I was talking about. True story. Even normal places like walmart, although they sell linux online, don't have that option on the shelf retail, well, at least I have never seen it.
If/when some big box vendors start offering a choice at the local retail level, you might see it start to change more rapidly. Until then it's going to be very very slow. It's changing, but that is really why it hasn't changed faster.
I think it will speed up some now as MS gets more restrictive with updates, etc. People are very used to just slapping in a random MS OS disk and reformatting, etc, and casual piracy has been rampant. As MS makes this more difficult and people are kept being forced to drop 60$ every other month to drag their box to the store to get it cleaned and back to a somewhat functional state, eventually they will investigate matters a little closer.
I read comments on Slashdot telling me that Linux has better support for devices than any other OS, and already has all the kick-ass 3D shooters. Doom 3, UT2004, and so on. Now you're saying it needs both.
.NET or Cocoa, that replaces the kludgery that is X11 and focus on the best usability we've ever seen. Get rid of that crappy start menu and taskbar Windows rip-off. Get rid of all those hideous fonts, redundant menu items (KDE ships with "Control Center", "Utilities," "System", and "Settings" as seperate menu items). Everyone needs to start over and do it right the first time. Where is the binary installer/uninstaller API? Better yet, where is the drag-and-drop installing that OS X has? Nope, let's rely on conflicting console package managers for another ten years. That'll bring the mainstream developers to port their products.
What Linux needs to dominate the desktop is one single unified desktop environment, based on a single universal API akin to
When you have ten versions of the same functionality, it's not "choice," it's redundancy. Most OSS people see things in black-and-white where all choice is good. There is moderation for everything, and you have to look at situations on a case-by-case basis. The desktop split that forces me to install two entire desktop environments just to be able to run all the apps out there is completely insane, and if some other company like Microsoft was doing it, everyone would be all over it.
Based on the obsession with KDE and GNOME in the Linux world, I don't see changes happening any time soon, though GNOME seems to be the one not afraid to break the mold and actually try different things for usability's sake. People bitch about Windows, and then when GNOME does something different, and according to all usability studies better, than Windows such as changing the button order or having a spacial finder, people bitch that it isn't close enough to Windows. It's gotten to the point that I've decided people don't really hate Windows at all. They just blindly hate Microsoft but want a Windows to use.
Just my opinion. I'm tired of hearing about how Linux is "almost there" every year, and it never happens because the attitudes and focus don't change. As far as I'm concerned (and this is the recently-converted Apple zealot in me speaking), OS X already beat the OSS world to the punch five years ago when it comes to GUIs on top of UNIX. KDE and GNOME feel like shallow 1998-era wannabes in comparison once you've used OS X all day.
As much as I dislike Microsoft, if it wasn't for them, I think over two-thirds of Slashdotters reading this wouldn't even be into computers like they are now.
Hate the business practices, but don't hate the technology. Microsoft is a great software company; they were just stuck with the problem of DOS backwards-compatibility for ten years. I haven't seen a BSOD since late 1999 when they released Windows 2000 and began unifying all their Windows products onto that codebase.
And another thing for Microsoft-haters, get over Clippy! Haven't seen him since '99 either. And Microsoft Bob was in '94--11 years ago. It's time to live in the now.
Linux rightfully deserves the title of being on the forefront of technology
I don't think Linux is on the forefront of innovation at all. What it has popularized (it sure didn't start it) is community-based development. But it's based on a 30+-year old UNIX model, and its desktop environments resemble Windows in many areas. A prominent OSS guy whose name I don't recall said it best--"All the volunteer effort in the world, and what do we do? We clone UNIX, then we clone Windows on top of it."
If there was ever a top innovative OS, I would say NeXTStep. A lot of the features it had are just now making it into mainstream operating systems, OS X included. And yet, GNUstep/Openstep efforts flounder and get ignored while "Qt" and "GTK+" continue to get used. Sigh...
Anyone in an ISP who has to manage Windows servers already knows what a nightmare it is. Linux simply reduces administrative headcount and improves security by its very design. Whereas it's typical for a commercially managed Windows server to have 30 accounts in the Administrator group just to get the work done, Linux boxes tend to be far far far more tightly managed. They're just built to BE managed, as opposed to Windows which is more of a Desktop machine with a server tacked on to it. So commercial service providers, the ones with thousands and thousands of servers already know that to offer affordable services to their customers they have to do it with Linux, Unix, AIX or Solaris. Windows is always a customer choice not a vendor preference.
Dupe of Slashdot comments from the past seven years, that is.
Never believe "Everyone I know is" comments. I've heard so many over the years that didn't predict a damn thing.
READ THIS BOOK: Crossing The Chasm: Marketing High-Tech Products to Mainstream Consumers
Every one of you who reads this site regularly should read it. Linux is not yet at the "tipping point" of crossing the chasm. The past year has been enormous though. I give it 2 more years, personally. IBM and Novell are huge and will make it happen.
Berto
From the article: " It consists of a Linux kernel developed by Torvalds and his colleagues by radically improving an earlier open-source Unix released by Andrew Tannenbaum in 1987"
No, no, no... How many times do we have to tell these people that Linus DID NOT ALTER MINIX to produce the Linux kernel!!! When will these people get it right before blathering on?
The race isn't always to the swift... but that's the way to bet!
The thing about Linux is that, being a volunteer-driven entity, it can pursue many different development paths at once. That's even if I take the false dilemma at face value (is Wine a statement that Windows is a standard, or junk to be ignored? How about mingw32?).
Ethan
heh well if she's 8 and uses linux... not only that but gentoo and can install programs via emerge... well she wont have any problems with installing a printer or video card. also getting into college and doing well on the SATs...heh
The best education consists in immunizing people against systematic attempts at education. - Paul Feyerabend
That's not really the problem. It has more to do with Linux as a game development environment totally lacking any good tools and debugging. Game developers don't want to have to make these tools themselves, that only hinders development times.
I have to disagree. The lack is not in code or tools to produce code. Linux does not really lack for game code -- it is at least as rich as the Windows world in available game libraries, and vastly richer in codebases to crib from. Some very extensive projects have been conducted, and very little has been built on them (Crystal Space is a great example).
The lack is in content -- audio, video, graphics, and even well-written English text. Almost every Linux game project I can think of has a severe lack of content.
On the Linux Game Tome today, if I search for top-rated games, the top-rated is Battle For Wesnoth. That game at one point used cribbed graphics from a commercial game, currently has a very limited 2d sprite set that would have been par-for-the-course in the Super Nintendo's heyday, and has character facial portraits that are atrocious (one more comment about the prince looking awfully fruity, and I think everyone will go insane). And this project is known for having rather good graphics for the Linux gaming world. I don't mean to bash the Wesnoth people (hell, I've hacked on the codebase), as they have a fun game. But the limiting factor is very definitely graphics.
Another game I've enjoyed is lincity. Good game, all open-source and not just a clone of a closed-source game -- but the graphics are reminicent of bad MSDOS game graphics.
After Wesnoth on the "top rated Linux games" list is Freeciv. Freeciv has graphics that are reasonable...for a decade-old game. It does not improve on the game it is cloning.
Next is BZflag, possibly the most popular 3d-accelerated multiplayer Linux game. It has incredibly simple graphics (note that 2.0 has improved things a bit). DOOM is really more graphically complex than BZflag. Again, fun, but it's simply not remotely able to compete with modern 3d games when it comes to texturing and modelling.
Next is Neverball. While this game has decent-looking textures, it also has no more than a handful of textures all told, perhaps a tenth or a hundredth of what a commercial, closed-source 3d game would probably have.
Next on the list is NetHack. NetHack is a terminal-based game (not that I think that this inhibits gameplay, as I just finished a four-hour stint playing Tales of Middle Earth), with extremely simplistic tile-based graphics. There have been a few attempts to improve things -- Falcon's Eye is a notable NetHack fork, with music and alpha-blended graphics -- but still nowhere near modern commercial-quality graphics. Now, as the NetHack aficionados among you know, NetHack can be a lot of fun, and while long-term replayability depends more on game logic than graphics, anyone who thinks that graphics and sound don't play a key role in making a game enjoyable is simply not being honest with themselves (and I would suggest that they try watching a horror movie with the sound off).
I am not demanding that open source developers do differently. I hack on games for the fun of it, and would not be interested in producing graphics, because I am not a good artist, or someone that finds creating game graphics fun. Good coders like donating their time -- perhaps because they are in a position that currently pays well and lacks enough employees that they do not need to compete as hard, and can afford to give away work as gifts for the sheer enjoyment. Artists work in a rather more competitive world (there are simply more people that want to be artists than there can be funded artists), and do not seem to be able to enjoy the same gift-based culture.
Another consideration is degree of work commitment. Code is largely opaque to the user, and differences between programmers large
Then my work is done here!
The cake is a pie
One of my clients is moving their practice to a completely paperless system. A key part of that system is signature capture. Medical and legal documents are pretty useless without a signature.
I couldn't find one single signature capture device that could embed signatures into an open-office document.
Some of my clients are ditching tape backup for on-line, off-site, backup services. These services require the installation of a software "client" on the server that stores the data. Guess how many of these services run on Linux? Not so many.
These examples illustrate the problem most businesses have with Linux. Linux is great for stand-alone boxes where interoperability is not a factor (databases, web-servers, proxy servers, firewalls - etc). It doesn't look so compelling when you consider compatibility with lots of devices and services.
Businesses look at aquisition costs, and recurring costs of technology, but they also look at the cost of "incompatibility". Not being able to do something, sometimes, costs way more than buying software licenses.
The barriers to Linux these days are not technological barriers, they are mass market standardization and acceptance barriers. For Linux to really shine in the business world some things about Linux will have to change. It will upset the Linux "purists" but the business community will demand it.
Linux is about choice - but that's what the business community doesn't want. They don't want to have to choose a window manager, or a distribution, or even a web browser. The business community wants all of that choosen for them, and a reasonable level of certanty, that most vendors they want to work with, will support their standard configuration.
-ted
I am the "linux guy" for a few businesses in the area. I charge $50/month per computer. That's for software, basic desktop support, security updates, most everything. I support Linux on servers, firewalls, and desktops complete with MS Office and a couple other necessary Windows apps.
I want to be perfectly clear when I say this: "There isn't an actual, profitable business in the US that cannot afford me or someone like me." Many businesses spend almost this much on virus protection software alone. There's a Garner study somewhere that estimates the total cost of maintaining a PC at something like $3600 over three years. I'm nearly half that.
Beyond that even, much of the initial costs of Linux can be offset by planning ahead, purchasing equipment and software with Linux in mind. Due to it's open nature, Linux lends itself to a regime of continually reduced support requirements in a way that proprietary software does not. My clients will see the most benefit from switching to Linux years down the road.
I've taken businesses that literally could not use their networks because of the number of viruses and worms on their machines and upgraded them to Linux for less than the cost of installing Windows in the first place. After most glitches are worked out in the first year, maintenance costs are lower as well. I'm sure I don't need to tell you that the incidence of viruses has nearly disappeared.
Of course, I'm not saying Linux is for everyone. My clients are generally people who don't want to be bothered with maintaining computers and software, and want it to "just work." If they thought they could handle it themselves, I'm sure most of them would switch to OSX. The fact is, most people can't or don't want to fiddle with computers, and Linux is one of the only OS's priced low enough for people like me to make money fiddling with computers for them.
"I assumed blithely that there were no elves out there in the darkness"
Every year number of users grows for more than double.
I love when people just throw out random statistics with no sources. Everyone is a walking usage study chart on Slashdot.
- Manufacturers start releasing Linux drivers for their hardware along side of the Windows and MAC drivers
- Web sites stop building their sites to only work with IE on Windows
- PC manufacturers sell more systems with Linux pre-installed for the average user (which makes my job of playing everyone's Professional Geek easier).
- PC manufacturers reject the idea of hardware which *ONLY* works on Windows (read: winmodem, Windows only printers, etc.)
- It becomes easier to convince average users that yes, your little home PC really does work better (fewer viruses, spyware, etc.) with Linux
- Budget ISPs (like WalMart, BlueFrog, Nescape, NetZero, etc.) release dialers and connection utilities for Linux, so I don't have to custom configure a system if a friend wants to use a dial-up provider instead of spending $45/month for a Cable or DSL service he or she only uses once in a while
- At work, I don't have to fight with the IT group every time they try to re-image my workstation with the latest "Standard" XP image, or explain yet again why I really *DON'T* need to update my anti-virus software on a daily basis in order to keep their network free of virus threats.
I could go on, but these are a few reasons off the top of my head why I think it really does matter that Linux is continuing to gain traction on the desktop. There really are advantages to gaining momentum. Linux was my primary OS long before it was "Popular" or "Politically Correct", but I can see plenty of advances that have come about because Linux is now on everybody's radar screen.Do I think that 2005 will be the magical "Tipping Point" for Linux? I don't really think we are going to see a mad rush for Linux on computers sold at stores like CompUSA, Best Buy and WalMart, or even online stores like Dell, but there is certainly a smell of change in the air...
Your Servant, B. Baggins
Huh??? Which century are you living in? I used to build custom kernels all the time, but it's been several years since I had to compile a kernel. I suspect that the average user needs to recompile a Linux kernel almost as often as they need to re-compile their Windows kernel. Granted, most distros come with horribly bloated kernels with every module in the universe installed, and building a custom kernel is much more efficient. but it is hardly necessary.
Your Servant, B. Baggins
"...single-handedly invented a computer operating system..."
Not a soul helped him...
1996 called, they want their tipping point back.
In particular, the comments on the possible rise of the IBM Power architecture were interesting. I'm not holding my breath waiting for POWER to take over the World, but there's a good chance it could start to take a significant chunk out of the Wintel monopoly. Seriously, I doubt that it will take off in the desktop or home markets as the author seems to imply, but I could see it as a really exciting change in the server market!!
Your Servant, B. Baggins
You're lucky. I've lost windows installs to indiscriminantly removing spyware with ad-aware. Those guys are getting sneaky enough to do some damage on the way out.
The only option is transgaming, which *might* work and costs money.
Not so. Most games that work with Cedega, also work with plain-old-wine, with the help of a no-cd crack which can be found quite easily and doesn't cost a penny.
Non-computer people see the computer as a single entity - it's not quite Windows that crashes, it's "the computer" (but they do seem to know that Microsoft is to blame). They really don't give a crap about the distinction between the OS and the computer, so they may see it as more reasonable to get a whole new computer rather than perform what they perceive as "brain surgury" to put Linux on it (and which Linux should it be? Red Hat, Linspire, Gentoo, etc... kind of confusing for someone who doesn't follow this stuff and isn't quite sure what role Linux plays on their computer).
So here's the logic as I see it: now that a Macintosh can be had for a reasonable $500, and getting a new, "different kind" of computer seems like the best solution, the Mac Mini seems more likely to replace Windows on the average desktop than Linux. Everyone's heard of the Apple Macintosh, and Mac software does sometimes appear in retail stores (besides Apple stores). It seems to be selling well enough.
Now, we know that it's cheaper to install Linux on your existing hardware, but just try to put yourself in the shoes of average, don't-care-about-computers-they're-just-a-tool person and imagine how they see things.
We apologize for the inconvenience.
I rebuilt my wife's computer a while ago with Linux and installed Firefox. I even put the same desktop background on it that she had on her old Windows box. I told her, "Click here for the Internet." (She is not completely computer illiterate, and *DOES* understand the difference between the computer and the Internet.)
However, it wasn't until several weeks later that she even realized the computer wasn't running Windows.
(Now, if the CUPS project could get its act together so the freaking printing wasn't so unreliable, I'd be all set....)
Your Servant, B. Baggins
If you want a free (as in beer) corporate networking environment, then a lot of work has to go into linux. Where is the answer to Microsoft's Active Directory? Kerberos and Pam OpenLDAP? Ever set that up? It's different for every distro and it's dicey at best. Screw up your PAM files, and you're in a whole whack of trouble. NIS has been discounted as generally insecure. There is no NIS+ server for linux. The only thing that comes close (and beats the hell out of it too) is Novell's e-directory, but this is going to steer your distro choices. We need a clear, defined network transport, network authentication, and profile management tool, that allows multiple networks to be configured, a different form of network transport (based on user privelidge, not machine) and while we're at it, a graphical infrastructure that we can define services that will run on a server and use x-forwarding. If you have a strange app that needs a million dependancies that the workstations don't need, just compile it on the server, put a link in the gui, and the clients will query the servers for a list of applications, and then run the whole thing over an ssh tunnel, without having to type your password over again.
Give me that, and make it easy to install, and I'll convert 50 desktops tomorrow.
As usual, "Linux can already do that". :-)
Lesbian Gnu/Linux. Porn-get into it!
Live CD with links to porn sites! Have it able to play quicktime -- Does porn come on quicktime?
Religion is the main cause of atheism.
Tanenbaum was interviewed by a stooge from the " Alex De Tocqueville " "institute". The purpose of the interview was to get Tanenbaum to say Linus did not write the original Linux kernel. Tanenbaum though said other things. He said Linus wrote the original Linux kernel. Tanenbaum was critical of some of Linus's design decisions. Tanenbaum thought Linux should have done it his way!
Religion is the main cause of atheism.
Making a better product is about actually doing it, not talking about how it's going to happen soon.
FYI: Linux has become a better product since 1998.
1998 version of Linux VS 2005 version of Linux. 2005 wins as the better product.
Such an OS is a good idea as long as it includes a filter installed on this OS so that when I am surfing for porn I don't waste time because my search terms accidentally made me land on a site with stock quotes, sports statistics, technology news, and other useless crap.
I believe the same thing was said about the cube when it was introduced. Granted the mini is quite a bit less in terms of both cost and size, but it comes without a keyboard, mouse, or monitor - all those are extra. Once you get done adding it all up, it doesn't appear quite as attractive.
Ooh ooh yeah! A bit to left honey... NOT THAT LEFT!! Now youve ruined it!
Guess that means this time next year we'll be reading some more 'linux on the tipping point' stories :)
I can list dozens of things I love about Linux. There are also a few things I love about Windows that I don't find in Linux:
Windows has VBR packet writing for CDR and DVDR available (not via Microsoft, but the Windows kernel makes it possible...unlike the Linux kernel), and has since around 2000.
They have a C++ tool that can do RAD including a fully integrated editor and GUI builder.
They include multilingual, blind, deaf, and handicapped support.
There is a unified font system, and adding a font simply means copying it into a specific folder.
They also have tightly integrated apps, but I consider that more of a security risk than a feature...
Anyway, the point is that there are things that Windows has done well.
Mod me down and I will become more powerful than you can possibly imagine!
Kickass 3d shooter games and support for devices.
No, the biggest problem is that whenever Linux gaming is discussed all that is mentioned is "kickass" 3D shooters.
Time is what keeps everything from happening all at once.
Why is that so many people here assume all /. or Linux users (other than themselves) must share the same goals and viewpoints?
I mean appart from the AC, from which I expect no better, at least four other people thought the comment was "insightful". Or did they simply choose ignsightful because there is no "+1, pro Apple" modifier?
Anyway, for those who still not get it. Linux has no goals. It is a piece of software. Each person who work Linux has his own goals. Sometimes these goals are similar, and other times the means to achieve the goals are similar enough, that cooperation is possible. This is true both for the Linux kernel, and for the software that is assciated with Linux despite in general running on many different Unix platforms (including Darwin), and often also MS Windows.
What is cool about free software is that this is possible, people can work on different goals, and the those who manage to create the most useful code will win in the marketplace of ideas. So some people can work on making interfaces on top of Linux that make the computer look more like an MS Windows or MacOS box to the average user, while other people work on new and innovative ideas.
The marketplace will sort it out.
I have no doubt there is some validity in this theory, however I have not yet seen this happening. People always have said that, as FOSS software gains market share, it will start to attract more attacks.
Firefox is now at close to 10% of the market by all indications that I have seen. I have yet to hear of any major successful threats against it. I'm sure some will come, but only time will tell if the virus writers like IE because it's popular, or because it's so full of easy security holes...
Your Servant, B. Baggins
I thought compiling your own kernel and software was a Linux exclusive.
You don't have to do that in Windows or Mac OS.
Linux needs to jump over a lot more ease-of-use hurdles before it can make a solid replacement for tech-ignorant people.
Up, Up, Down, Down, Left, Right, Left, Right, B, A, START
I don't have a good answer, but I have come across a person that felt that way. I think it may actually have had something to do with the kernel although I wasn't interested enough to pursue the subject when this opinion was voiced. There are people, though, after all, who claim that "linux" IS the kernel, so it doesn't surpise me that someone might equate BSD un*x with a monolithic kernel.
Whoohoo! I had no idea anything like this existed. Fabulosity baby, I'm going plaid! (<- oblique Spaceballs reference for the clue-impaired)
It'll be fun to see what the correlation is between the Doppler Effect and my own spare time. Well, at least once I've finished doing my taxes, or I'll have a number of unhappy government types come calling...
"What in the name of Fats Waller is that?"
"A four-foot prune."
I read somewhere that Windows NT/2K/XP etc. uses the Mach Microkernel, or scheduler implementation, or something. Where was that...
/ evaluate/featfunc/kernelwp.mspx
Ah-ha: http://www.microsoft.com/technet/archive/ntwrkstn
Ok, so Windows uses the same microkernel as Mac OS X. Where'd they go wrong? Mac OS X kicks teh shit out of Windows. I agree Windows generally sucks, but this whitepaper is very informative and interesting, though a lot of it goes over my head. Reads to me like, Windows wouldn't suck so bad if they didn't run things "kernel mode" like fuckin' IE, the printing subsystem, and the whole GUI, etc. I'd like to see a Linux kernel hacker's views on this whitepaper.
grep -iw skynet
A lot of Linux vendors produce "Personal Editions" - to name one, SuSE, but in no way are they crippleware like Win XP Home or - Windows XP "Starter Edition" if you live in Asia - since all the packages are available via FTP.
While not convenient for the home user to be downloading the packages for GCC and whatnot, most of the people using GCC and/or compiling anything from source wouldn't be using a personal edition! My grandparents are about to change to Linux (they don't know it yet), since I'm off to Europe, and they won't have me to fix their Windows 98 problems.
Everything is just there, and, in ther personal editions (My Grandparents will be going on SuSE Personal with remote-assistance via VNC turned on just in case), what doesn't need to be there, isn't. Check list of roughly what they will be getting (will try to minimise the learning curve):
MS Office 97 -> OO.org
Outlook Express -> Evolution
IE 6 -> IE6 on Wine (possibly for banking sites) and Mozilla Firefox
Card Games in Windows -> Card Games in Linux, Tux Racer, etc.
WinAmp/WMP/Real -> XMMS, Juk, Helix (Real 10)
WMP/PowerDVD -> MPlayer and/or Kaffeine (with the blue interface thing)
Paint Shop Pro 4 -> Gimp 2 and DigiKam
I will hide (or at least move out of the way) things that they don't really need and things that could cause confusion... and only the stuff I (they) want installed will be installed. No extra crapola to deal with.
This is something I have come to like about Linux, and also there is the whole fact of 'if I don't like something, I can re-write it', or at least find a modification. Don't like the interface? Install a different one.
I really think myself that the Linux interface(s), the programs (especially the ones that are under continual development - which is a lot!) are coming of age.
My preference on the underlying System is Novell Linux 9 (a nicely polished product, but still room for improvement) or Suse Pro 9.2 - but Debian, Mandrake, Gentoo - hell even Fedora, Linspire Xandros and Ubuntu all have their places.
I've used most of these systems - used to like RedHat, but, like many others have moved on. I'm not saying I don't like them, I'm simply saying that my preference is different. When something better comes along (or when I get to Singapore or London and get a Mac), I will stick to what I know and like, but of course, I will never stop playing with other systems.
They all feel... less dirty... than Windows, and, in many cases easier to install and configure than Windows. My laptop says "Designed for Windows XP" on it - but XP requires drivers from the manufacturer. Suse/NLD in particular just works! How is that for a tipping point?
Founder & COO, Hayai India (hayai.in) / USA (hayaibroadband.com)
It already has "kickass 3d shooter games". What it really needs now is "kickass SOMETHING ELSE THAN 3d shooter games". Seriously.