Guest post by The title of the article seems familiar to you? Naturally it would, when you read something like this. But I do state the corresponding sentence isn’t even grammatically correct, thereby making it difficult for me to parody. I am sure that Linux is not close to extinction but is rather gaining momentum or at least holding its ground.
Now the following stipulations stand. First, in this article I will address the topic of software freedom or openness; it would be impossible to discuss Linux without this. Second, I will weight the pros and cons of the Linux way, the Windows way, and other “ways” as I see fit. Third, everything concerning these issues will partially be influenced by my own beliefs (anyone that tells you they are just describing the situation objectively without expressing attitude needs to take at least the following courses: psychology, sociology, literature, and – if time permits – genetics.
Linux’s success
Since you’re reading this article, I’ll assume that you frequent OSAlert. The volume of recent articles chronicling the successes of Linux may have led you to believe that it has a lot of momentum. This is true. The data in that article shows that Microsoft’s Windows servers’ revenue grew by 4.6% this year, Linux’s by 5.4%, and Unix’s declined by 1.7%. This means that Linux servers are still a more popular choice than Windows servers while Unix servers are becoming less and less popular. This of course, is expected with companies such as SGI and Sun more-or-less switching from Unix to Linux. Now I believe through all of the circumlocution in the previous article’s conclusion that the author’s overall vision was that Unix would eventually replace Linux where Linux now dominates. This is all nice and theoretical, but all figures indicate otherwise. Things fall when they stagnate, not when they are moving with all deliberate momentum.
You may object right now – “The previous article wasn’t about servers! It was about desktop Linux!” And you would be correct. According to this article, Linux’s desktop market share is 2.8%, an order of magnitude smaller than its server market share of 28.3%. I believe you can reason which is its more important market. Since its clearly the server market, to base an article about the future of Linux on data from the desktop market renders his point, at the very least, invalid.
The matter of OS choice
Windows has a much easier to use graphical interface than Linux for novices. Even if you’re an ardent Linux aficionado, you’ve got to concede on this point. Think about it: how many times a session do you drop back to your terminal to perform simple administrative tasks, etc.? If you still don’t agree, you probably don’t want to read on.
Okay. Now we’ve got that Windows basically monopolizes the desktop market and has a much easier to use GUI than Linux for novices (I realize Linux doesn’t have a GUI – by this I’m referring to things such as GNOME or KDE). Great. It always was a system for geeks. Think about it: you frequent OSAlert; tell me you’re not a geek. Okay, I also realize that many people are probably reading this on Windows or Mac right now, but a fair number of OSAlerters probably have Linux on some computer.
Additionally, Linux is not a company. With all of these recent news announcements concerning Novell or Red Hat or Oracle or Canonical, it’s easy to lose sight of this. Linux, in the common sense of the word, is a community – a community of geeks. And unless Microsoft coalesces some government into passing an anti-geek law, this community can’t die. Projects like Debian or Gentoo will always be around. In many ways you can think of this like V for Vendetta. Linux is an idea, and ideas never die. Hell, people still program in Fortran. Why? Because no matter how many companies out there release programming tools to compete with Fortran, it’s an idea that’ll never die.
Now to address the previous article, the author puts a tremendous amount of emphasis on the desktop, on WYSIWYG design (which really has nothing to do with the GUI but let’s bear with him). Now I would argue that the command line is usually far superior to using a GUI, as it allows you to work much faster and more flexibly. But for trivial tasks such as checking your mail or browsing the web, it’s often more convenient to have a GUI available. But think biology and evolution. A GUI is the natural result of a need to check mail. GUI’s are not Microsoft-invented concepts as the previous author implies; indeed many attribute the first GUI to Apple.
The previous author goes on to say that a GUI (or what he calls a WYSIWYG desktop) and the Unix way of accomplishing tasks on the command line are mutually exclusive. Next he mentions OSX, which is a GUI on top of a Unix system. I believe by what philosophers call a reductio ad absurdum, he’s just invalidated his own statement. But disregarding that, putting a simpler facade on top of a more complicated interior is commonplace. Think about your TV. Many remotes have a little slider that hides a set of fancy buttons. Now think about something like an IDE. It automates most tasks but lets you screw with the inner workings when you want to. So why shouldn’t we extend this metaphor to a whole system? We should, and we do.
The fair race
Now the author seems to think that Linux is simply copying Microsoft on everything. That’s not true. What Linux aims to implement on the desktop is simply the end of a rational series of progressions. Let’s say you have an error. How would you handle it? Now about telling the user and asking them what they want to do? Great. How do we do it? Dialog box! Great idea! That’s the human thought process. Drop a computer in the Garden of Eden during man’s golden age and Adam would’ve come to the same conclusion.
If the previous author had any further point in this section, it would be that Linux’s difficulty today is providing a Microsoft-like GUI on top of its traditional Unix-like system. What he really means to say is that Linux’s trouble is providing an efficient single-user experience on a system designed with a multi-user server-client model. I concede; this is why it’s taken so long for things such as Upstart to emerge. But simultaneously argue that this does not mean the Linux model is a failing system. Instead, it needs to update its traditional thinking, and this is absolutely beneficial to the community in general. However, the author seems to think that this would also incur a complication overhead due to more abstraction; this is simply not the case. Take a look at any of the new Sysvinit replacements; they’re all easier to use than the original.
Predictions
I’m not even going to address this section. I’m just going to point out that this author thinks ReactOS will replace Linux and leave it for readers to draw their own conclusions. (Don’t get me wrong. I fully support the ReactOS project, but honestly…).
Conclusion
Anyways, my whole point in writing this article was to refute the previous author’s work. Linux is progressing strongly and is not going away in either mine nor your lifetimes.
If you would like to see your thoughts or experiences with technology published, please consider writing an article for OSAlert.
This article means well, but you dignified with a response a piece of writing so willfully ignorant, so poorly written and so illogical that the best thing was to ignore it.
I second that. “Reductio ad absurdum” doesn’t even mean what he thinks it means. It means contradicting a hypotesis (on purpose, even if you know it’s absurd to do so), in order to show that the inverted alternative would be impossible and therefore the original hypotesis must be correct. It’s a mathematical problem-solving method which can help when you have enough data to contradict the negated problem but not enough to prove the normal problem.
He thought it means to contradict yourself, which it doesn’t.
Edited 2006-12-06 16:07
OK… no, this is not correct. There are enough flames in the comments, we don’t need some kind of article flamewar. This is the least we should expect from osnews.com
We already made clear our position in the other article, now we must let it go.
Edited 2006-12-06 03:07
I dunno, maybe if I could comprehend the bad grammar.
I’m just going to assume that this and the previous “war” article are equally pointless and stupid.
And yeah, we sure don’t need article flamewars.
Edited 2006-12-06 03:10
After reading that the internet is now over 16 years old not so long ago I realized that many of the most basic concepts we use in GUI’s today were released in the early 90’s and likely patented before that.
After these technologies hit the 20 year mark it won’t matter who started doing it or patented it first becasue patents are only good for 20 years.
That might seem like a long time but not when you consider that UNIX has been around since the 60’s and Vista is going to be released almost 17 years after Windows 3.0.
The GUI really has not changed much since Win95, the patents for which will likely mostly expire between 2009 and 2013.
Mmm, the Internet started as ARPANET in 1969.
The basic GUI concepts appeared at Xerox’ PARC in the ’70s.
These technologies HAVE hit the 20 year mark and as you can see, who started doing what and when still matters.
And you can renew patents, you know?
No you can’t. That’s the whole point of the patent system, so people will share their knowledge rather than make everything a trade secret.
I’m the author. Reading through my article again while considering the comments posted here, I see that a lot of stuff in it was badly written / overly caustic / flaming. I apologize, but I do stand by a couple things:
1. Most of the bad grammar is intentional for the purpose of parodying the original article (of course this purpose in itself is malicious and something I probably shouldn’t have tried in the first place)
2. Linux is not losing a war or even fighting one
3. Linux is not copying Microsoft
Again, I apologize for this poorly written article and will put more deliberation into future ones.
Now the author seems to think that Linux is simply copying Microsoft on everything. That’s not true.
Of course it is not. Linux is copying Unix, too.
Linux is not losing a war or even fighting one
Well it is not winning a war either, that’s for sure (according to the market share numbers).
You don’t win by simply copying the stuff and that is exactly what Linux does, no matter how hard you deny it. They simply copy everything from Unix and Windows, all the time. Great ideas should be copied, no problems here with that, but the thing is — they don’t do it well. Microsoft does it really well (“embrace & extend”).
I think you were right, with some corollaries.
Linux was essentially written with the goal of copying Unix. It’s like spaghetti was originally made with the goal of copying Chinese noodles. Now you wouldn’t call spaghetti a copy today would you?
Next, I don’t think Linux copies from Microsoft at all. In fact, I would argue it the other way around. Microsoft does not “embrace & extend”; it’s strategy is “copy & market until people don’t realize it’s not original.” I distinctly recall a Microsoft executive or someone affiliated with Microsoft claiming that the company invented the operating system. Pull a person off the street and ask them who invented the operating system; I guarantee the majority will tell you Microsoft.
Linux does copy a lot from Microsoft or general OS concepts. In fact there is hardly a kernel level innovation that is done by Linux and it has been mostly behind in implementing kernel level features as compared to solaris or Windows.
Example: M:N thread support, Well defined driver model, Asynchrnous IO in kernel etc etc. You should read “A tale of two kernels” by Mark Russinovich.
Next, I don’t think Linux copies from Microsoft at all. In fact, I would argue it the other way around.
Microsoft doesn’t have to copy from Linux. MS had its own Unix called Xenix some 25 years ago and it was the most popular Unix for desktop PCs, so I would say they hardly need Linux to copy from.
Microsoft doesn’t have to copy from Linux. MS had its own Unix called Xenix some 25 years ago and it was the most popular Unix for desktop PCs, so I would say they hardly need Linux to copy from.
If Microsoft and their lackies think 25-year-old technology represents today’s state of the art, no wonder the company’s products are so laughably execrable.
If Microsoft and their lackies think 25-year-old technology represents today’s state of the art, no wonder the company’s products are so laughably execrable.
Nothing stops Microsoft from offering both the good old stuff and the new cool thing, at same time. Besides, some things are just good the way they were 25 yeras ago and need not replacement. I guess you’re too young to know that. Just wait and you’ll see
Nowhere I said that 25-year-technology is the only possible way. You’re twisting my words.
Nowhere I said that 25-year-technology is the only possible way. You’re twisting my words.
No, you’re twisting mine. The fact is that if people stuck to 25-year old technologies (as you implied Microsoft can), they would be even farther behind.
But in fact even if Microsoft did base anything on XENIX in 2006, not only would it be obsolete, they would also be stealing, since they sold their XENIX product to the old SCO years ago.
The fact is that if people stuck to 25-year old technologies (as you implied Microsoft can), they would be even farther behind.
Again, you’re twisting my words. Nobody sad they have to stuck to 25-year old technologies. They can exist side by side with new technologies.
One thing you’re missing again: some things are fine today the same way they were 25 years ago.
But I already explained all that, kinda repeating myself. I think you know what I’m talking about.
But in fact even if Microsoft did base anything on XENIX in 2006…
Oh, come on.. I am not talking about literally copying and stealing the code, but about concepts, ideas, etc. For example, some Unix ideas/concepts are great and it’s great to have them in Windows, and vice versa, don’t you agree?
I’m sorry but I don’t agree. Linux may be indirectly modeled after Unix (Unix->Minix->Linux) but that’s as far as it goes. Linux has continued to evolve and Unix has for the most part stagnated. Not saying that Unix is bad, I think it’s a fantastic system and once you have it configured and running it doesn’t require much tinkering, because it’s a server OS; Always has been, always will be. Lets face it. If you throw a fancy desktop environment on Unix you may be able to get it to work on that particular model of machine, but try making it run on anything else. There’s a reason that you can’t get the newest desktop software for Unix/BSD…it’s not a flexible desktop OS.
If Linux were just a carbon copy of Unix, there is no way that it would have turned into the incredibly fast, secure, powerful, and flexible system that we have today. A system that is capable of being a mission critical server and the same time, desktop oriented enough for me to type this comment on.
You and I know it’s impossible to slap a UI over BSD, but those guys in Cupertino just won’t listen.
“You and I know it’s impossible to slap a UI over BSD, but those guys in Cupertino just won’t listen.”
I’m assuming that your being sarcastic, but anyway. I’m a huge Mac fan and have been using them for 19 years. OSX/Darwin at this point is only a distant relative of BSD. You would have an easier time comparing Darwin/XNU to HURD or Tru64 Unix. Besides, have you ever tried installing KDE or Gnome on a Mac. It’s possible but not the best way to spend a weekend. Aqua/Cocoa/Quartz were built specifically for OSX and that’s why they work. As much as I love OSX, it too is not that flexible, certainly not as much so as Linux.
“Linux may be indirectly modeled after Unix”
It is modeled directly after Unix and Minix.
“There’s a reason that you can’t get the newest desktop software for Unix/BSD…it’s not a flexible desktop OS.”
I must be imagining myself typing this on the latest KDE on OpenBSD then. And lets not even mention DesktopBSD or PC-BSD.
your a joke, you think linux did not “copy” NTFS, or reverse-engineer/copy the Windows API’s. ???
linux copied Unix so they could gain market share of Unix uners.
Once they did that, they changed their target to windows. as windows and MS are a successful model for an operating system, and application suite.
therefore, next thing you see, is linux “desktop”, OpenOffice.
Was it “just by accident” that you “.doc” format is exactly the same as what MS developed, by paying programmers ??
this is a joke, its pointless saying Linux does not copy the good work from everyone else.
and then use that stolen IP against the very company they stole it off.
Why else would you need or want NTFS in Linux ?? if it was not to take MS Windows users away from MS.
You saw something good, and popular, so you copy it, and give it away.
and wonder why the company you stole off, gets upset.
and then use that stolen IP against the very company they stole it off.
Reverse-engineering is *not* stealing IP. It isn’t in the US, nor is it in Australia. That’s a blatant lie.
Linux copied the desktop paradigm, but so did Windows and Mac OS.
Copying ideas (and improving on them) is not only legal, it is *good*! Why do you think Microsoft does it all the time?
Why else would you need or want NTFS in Linux ?? if it was not to take MS Windows users away from MS.
How about rescuing the data on an NTFS disk from a Linux system? How about sharing data between the two OSes on a dual-boot machine?
Are you going to say that the Ext2 and Ext3 drivers available for Windows are also examples of “stealing IP”?
BTW, Microsoft *itself* said that it wanted to improve interoperability between *nix systems and Windows…didn’t you get the memo?
I swear, can any anti-Linux troll actually post *factual* information once in while??
I swear, can any anti-Linux troll actually post *factual* information once in while??
It would certainly make a nice change, wouldn’t it?
There’s only one way to achieve something, that is why in Evolution eyes have evolved several times independent from each other, and why scientists think that on a planet with similar conditions like on earth a species with similar biology to man would again evolve as the dominant (intelligent? not so sure about this) one.
There aren’t that many ways to interact with a computer outside a console, and to implement a GUI if that’s what you want and it does not mean implementing one is copying MS or Apple.
BTW why do so many on here seem obsessed with market share? I always see these figures quoted but the reality is even only 2% of a global desktop market will quite likely be millions of machines which is significant enough and self sustainable. Linux type developers are clearly passionate enough not to ever let up coding what they care about without getting paid and distros like Debian or Slackware will never go away even though commercial Linux might one day.
I do not care about market share and please, people like NotParker, stop banging on about figures. They indicate trends AT BEST and frankly, the less Linux usage out there the longer we can continue to leave virii and malware to Redmond’s OS – no hard feelings.
For the people who like to use something they consider technologically superior or just more flexible and suitable to their needs and liking.
IT IS ABOUT CHOICE – NOT ABOUT WHO ‘WINS’.
Don’t apologize! I haven’t had such a good laugh in quite a while. Humor is a great way of combating FUD.
“1. Most of the bad grammar is intentional for the purpose of parodying the original article”
Well, that explains it. Maybe the parody is more obvious to those who read the previous “War” article (I didn’t, it looked pretty dumb just reading the title).
I would like to share the following argument against the new version of the GPL Open Source license.
I have seen something being ignored in all of the articles in the press…that the new version of the GPL (version 3?) threatens to undermine the software industry.
Now don’t get me wrong…This is not political at all but I think the hobbiests and the professionals can remain separate and the industry will still do the innovating…What alarmed me is when I read between the lines of a recent Bill Gates interview…he spoke of the new version and seemed not to approve.
Please…this is NOT political, but if the biggest player in the world has troubles over something like this…everybody and I mean all of you need to BACK OUT of this NOW. I mean at least wait maybe until 2008 when things have settled down.
If you still doubt me on this…think of the enormous impact Microsoft, Oracle, DELL, etc. have on the economy and in mutual funds. Pension funds, the list goes on and on. I am afraid too many of you people are playing with fire and are not respectful of the Intellectual Property of others. The GPL V3 has no place in any country, that respects all of the ip of the software ecosystem. Important figures like Steve Ballmer have also spoken out against the GPL…he also spoke out against Linux OS’s that have not paid up yet…but MOST OF ALL…he is enforcing a deal that I read that V3 will UNDERMINE! Seriously, people…you cannot take on risks like this…Wake up!!
Do you take any statement made by Steve Ballmer seriously?
http://video.google.com/videoplay?docid=-3446931931514285011
Just watch this video.
Now do you consider him an important figure?
ah. steve doing the monkey dance. a classic.
honestly, any other average joe would have been locked away in a room with padded walls for doing soething like that… at microsoft you get a well-paid exec job for it.
’nuff said.
rofl. This is the funniest bit of sarcastic parody that I have seen in a long while. Jolly good show.
“rofl. This is the funniest bit of sarcastic parody that I have seen in a long while. Jolly good show.”
Please keep it civil…no politics…just stick to the issue. You are for GPL3 or for the economy…and I dont appreciate being marked down by cultists…my post addresses the main issue BEHIND the article…mabye to subtile for some ppl.
“and I dont appreciate being marked down by cultists”
Because calling people cultists is very civil and “sticking to the issue”.
By the way, your use of “cultists” is a dead giveaway, NotParker.
By the way, your use of “cultists” is a dead giveaway, NotParker.
You can’t prove it…anyhow it doesn’t matter…I think his term matches those who oppose him. Although jihadies might work to.
@raynevandunem: No one can deny Steve Ballmer is important…also what does a video mean? ppl criticized old Bill Gates testimony also — so what?
Edit: Anyway, I’m signing off for now…I still have the hope for ppl…maybe somebody needs to do a Get the Facts thing but for GPL3 not just LinuxOS.
Edited 2006-12-06 06:07
No, we can’t.
It’s already apparent.
That was an actual video from a Microsoft developers’ conference in 2001.
Plus, what kind of CEO would vow to “f–king kill Google”?
http://www.smh.com.au/news/technology/microsoft-ceo-im-going-to-fin…
And its funny that he has his kids “brainwashed”: “You don’t use Google, and you don’t use an iPod.”
http://money.cnn.com/magazines/fortune/fortune_archive/2006/04/03/8…
It’s not that Stevie-B is important, as he is the CEO of Microsoft; it’s just that everything that comes from his mouth to the media is – let me emphasize this – f–kING RETARDED.
Please lie in confusion concerning Ballmer (aka “The Juggernaut, Bitch!”) no longer:
http://en.wikipedia.org/wiki/Steve_Ballmer
By the way, your use of “cultists” is a dead giveaway, NotParker.
The meme is spreading! Yay!
But … it wasn’t me. Just my ideas.
I tried swearing off confrontational speech, but I keep getting modded down by archiesteel and his posse of accounts.
I may well go back to my previous confrontational style since I get modded down either way.
Cultists don’t care whether you are rude or polite and factual … they just hate it when people don’t BELIEVE in the 3 freedoms and an obligation! They mod you down if you don’t kiss their *ss.
Edited 2006-12-06 06:18
I tried swearing off confrontational speech, but I keep getting modded down by archiesteel and his posse of accounts.
If you’re going to continue making the false claim that I have multiple accounts, then I’ll just claim that apolitical1 is one of your sockpuppets.
http://en.wikipedia.org/wiki/Sockpuppet_%28internet%29
BTW, if you don’t want to get modded down, try to stay on-topic.
It’s nice to see though that moderation the only reason you tried to rein in insulting those who disagree with you (what you call “confrontational speech”, which I’d call “incapacity to present counter arguments forcing a recourse to ad hominem attacks and strawman arguments”…though I admit your euphemism is shorter…).
I don’t know, it could have been because, well, insulting those who disagree with you is not a valid form of debate. But I guess you were unable to come up with rational arguments, therefore you’re going back to thoughtless provocation. Makes sense.
The meme is spreading! Yay!
How about two more examples of things that spread: First, the (untrue) meme that War and Peace was originally meant to be called War! What is it Good for?!”
Secondly, cancer.
Neither of these is a good thing, and by analogy neither are your interminable and interminably diuretic posts.
@apolitical1 your off topic its your first post. I am too I hope people do the same to me.
The argument is quite simple. GPL needed an updated license. people are now siding between v2 and v3.
v2 which allows developers to bend the rules of the philosophy behind the license *may* mean it that it gets greater adoption and recognized, and those benefits will filter back to the original developers.
v3 plugs the holes in the license and enforces the philosophy behind the license, which *may* mean less adoption and less recognized, but will allow users to enjoy the freedom of use on devices which does use it.
Pick a team.
GPL v3 was the reason why I abandoned Linux. I don’t like the trend, and I bailed out rather sooner than later. For me, Linux is history.
DG
Linux is licensed under GPL v2. Linus Torvalds has made it quite clear that he would not re-license it under v3.
If that’s the reason why you abandoned Linux, then you did so based on erroneous information…
Please let I never see another OSAlert article.
Linux is not the kernel in this instance its a reference to an end-to-end desktop solution which is supplied through a *distibution*.
Point 1) Definition of Linux on the Desktop.
Microsoft’s main Market is the OS with a stack of bundled utilities and Office with a stack of bundled utilities. Its not an easy split as Microsoft tries to take over every profitable part of the Desktop.
So Lets start Linux + X + Desktop Environment + OpenOffice + Firefox + Applications vs Vista + Office 2007.
My head is already hurting. I wouldn’t even want to try even to compare the two enviroments. It has 10 flames all over it.
Point 2) Ease of Use
Seriously break this one down for me, are you referring to my mplayer/xmms combo vs WMP or are you referring to OpenOffice vs Microsoft Office, or Nvu vs Dreamwever. or that clicking a K is much more easy than clicking start.
I suspect what you are trying to compare editing xorg.conf to clicking on a dialog, or writing a bash script to convert your jpeg to set size.
This is without responding to stuff like “how easy is it to fix an intrusion on a Microsoft vs Linux Desktop” or How easy is it to keep a current desktop with uptodate applications.
The reality is this is hard to do. What you can say is that Windows presents a familiar face to its users…and that is priceless.
3) CLI vs GUI
What eh!? How did you get to this dubious link. Regardless of the mix you put in this environment. War/Death of Linux whats the relevance to this. In fact your are talking about the CLI shell vs Graphical approach. Thats something completely different, and very little to do with the linux Desktop, and a whole argument that has been done better elsewhere. What I’m confused about is Linux allows you to use a mixed environment, you can have it your way. You can burn a CD from your command line or your desktop. You can even never use the command line shell..or use a graphical enviroment, and have many choices of *both*
Apart from the usual throwaway phases that I think are rubbish “a computer is a tool”(sic), “linux is a community”(sic).”Linux is for geeks”(sic); “Windows has a much easier to use graphical interface”(sic). Other than Microsoft PowerUser(sic) phrases what does this..or the last article contain.
Edited 2006-12-06 05:00
I want to justify myself here, but I don’t understand your first point. For your second one, I’m referring to something like changing TCP/IP settings. In Windows it’s trivial. In Linux it usually requires editing files by hand. For your third point, that’s exactly what I said in my article, so I don’t think there are any discrepencies there. In other words I’m agreeing with you.
For your second one, I’m referring to something like changing TCP/IP settings. In Windows it’s trivial. In Linux it usually requires editing files by hand
Err…From what year you coming from? As far as I know, most Linux distros default to DHCP by default, and things like SuSE has Yast which allows you to very easily and conveniently change your network settings. I don’t know about Fedoras or such, but I’d assume they have that too.
“Err…From what year you coming from? As far as I know, most Linux distros default to DHCP by default, and things like SuSE has Yast which allows you to very easily and conveniently change your network settings. I don’t know about Fedoras or such, but I’d assume they have that too”
Its important to remember that often Microsoft PowerUsers(sic) often try a Linux as they hear good things about it. They often give it a go…but there is something they don’t like X piece of hardware doesn’t work, or no alternative to X, or in a lot of cases *expected* things to work like they are familer, and simply were not as *good* in this environment, becasue things simply work differently.
Its also difficult to remember how rapidly Linux is evolving. If you worked with windows XP for the last 5 years *nothing changed*. Linux is still having a revolution. Its pace has been frightening. What was true of the desktop 6 months ago isn’t true now. Vista as it is is not just competing with Linux now, but the linux after that, and the one after that. If its 5 years again, thats 20 revisions of Linux; 10 of xorg; 10 of Gnome; and multiple revisions of the other applications that make up your distribution.
SuSE has Yast which allows you to very easily and conveniently change your network settings. I don’t know about Fedoras or such, but I’d assume they have that too.
Not just that, there is at least one cross-distribution solution with the GNOME Setup Tools, which is available on most generic distributions.
I always get a little irritated when people take it for granted that Windows has an easier GUI environment than Linux. There is no area in which Linux is making faster progress, so your knowledge and experiences from yesterday could already be outdated again. This becomes especially obvious when examples like the above are mentioned, which clearly aren’t valid anymore.
agreed.
for normal, basic work, KDE 3.5.x absolutely doesn’t offer a worse environment compared to Windows. in many area’s it’s actually better, and indeed worse in a few.
system maintenance is much easier in linux, esp if you have tools like yast. even without yast, you don’t need a4us, antispyware or firewall. driver installation is very rare. updates/upgrades are much easier. so in some area’s, linux might be a bit harder, but in most, it’s as easy, or easier, compared to windows.
MS Office is a great piece of work, and kicks Openoffice and Koffice’s ass in most (but not all) area’s. but the other office’s are improving, OO.o slowly and Koffice at a crazy speed. so this might get fixed in time.
Overall, a ‘normal’ user can work as easy with linux as with windows. he’ll need a ‘geek’ for certain things anyway, and that geek will need less time supporting his linux users, compared to the windows users…
In many ways you can think of this like V for Vendetta. Linux is an idea, and ideas never die.
Any OS News Op-Ed that references Alan Moore is headed in the right direction, IMHO.
The previous author goes on to say that a GUI (or what he calls a WYSIWYG desktop) and the Unix way of accomplishing tasks on the command line are mutually exclusive. Next he mentions OSX, which is a GUI on top of a Unix system. I believe by what philosophers call a reductio ad absurdum, he’s just invalidated his own statement.
Yes, thank you. I’m typing this from my OS X machine at home and though the terminal is there if I want to use it, I haven’t had to use it in over 3 years. I finally took the terminal out of my dock about a year and a half ago.
What he really means to say is that Linux’s trouble is providing an efficient single-user experience on a system designed with a multi-user server-client model.
And OS X shows that it can be done, and done well. But that’s partly because OS X is controled by a single entity and it’s easier to steer the bus.
Linux is … frankly, I’m amazed anything happens well or in a timely fashion given that the whole thing is a giant exercise in cat herding, and the cats are mostly part-time unpaid volunteers.
Linux is getting it done. But it’s going to be done at a different pace and in a different way is all.
Edited 2006-12-06 05:19
This session was pretty amazing. Not only did the computer scientist in me delight in hearing the details of how the Windows kernel does its job, but it was also very interesting to see the strides Linux has made since what I studied in college (it wasn’t that long ago, but we weren’t exactly up-to-date on our information).
The session was presented by Mark Russinovich of Winternals Software. My (unedited) notes from the session follow.
Slides reviewed by key players in Microsoft and Linux kernel development, with information about their comments to follow.
The history of Linux:
Starts in 1969, ken Thompson developed first version of UNIX at Bell Labs.
Three major branches of UNIX: UNIX System III from Bell Labs, UNIX Berkeley Source Distribution (BSD), and Microsoft’s XENIX.
In the 1980s, Microsoft had the largest deployment of UNIX systems with XENIX. Xenix was sold off to SCO later.
Linux 1.0 released in Marh 1994.
The History of Windows NT
Originally began at Digital in the 1970s when the VMS system was designed by David Cutler and others. Culter moved to Seattle to open DECWest and worked on pre-Alpha systems.
Gates hired Cutler in 1989 along with 20 other digital engineers to work on NT OS/2, which would be focused on OS/2 API.
In 1990, Gates refocused the project on the Windows API.
Windows NT 3.1 released in 1993 with the version number synchronized to the windows platform release.
Linux 1.0 released in Marh 1994.
Linux was in usable form well before the 1.0 kernel release. I started using SLS Linux just before the 0.99 kernel release, for example.
Gates hired Cutler in 1989 along with 20 other digital engineers to work on NT OS/2, which would be focused on OS/2 API.
To clarify, the only OS/2 API that Windows NT supported was the 16-bit VIO (text mode) API. I seem to recall that 16-bit PM (Presentation Manager) support may have been an optional add-on, but I’m not sure.
Comparing the Architectures
Both Linux and Windows are monolithic kernel cores with all core operating system services run in a shared address space in kernel-mode.
All core operating system services are part of a single module:
Linux: vmlinux
Windows: ntoskrnl.exe
Windowing is handled differently: windows has a kernel-mode Windowing subsystem, Linux has a user-mode X-Windowing system.
Windows Process Management
Process:
Address space, handle table, statistics and at least one thread (threads are the actual execution location for code)
No inherent parent/child relationship: When windows creates a process, the parent process gets the child identifier, but no further messages unless specificly requested. Likewise, if the parent dies, the child continues.
Scheduler
Two scheduling classes
Real time (fixed) – priority 16-31
Dynamic – priority 1-15
Higher priorities are favored
Priorities of dynamic threads get boosted on wakeups, but are not increased when they are CPU bound threads
Thread priorities are never lowered below the original priority level
Reentrant and preemptible scheduler
Linux Process Management
Process is called a task
Basic address space, handle table, statistics
Parent/child relationship
Basic scheduling unit
Threads
No threads per-se
Tasks can act like Windows threads by sharing handle table, PID and address space.
Pthreads — cooperative user-mode threads
Sceduling
Has 3 scheduling classes:
Normal (100-139 priority)
Fixed Round Robin (priority 0-100) (time sliced)
Fixed FIFO (priority 0-100) (runs until it releases the CPU)
Lower priorities are favored
Priorities of normal threads go up (decay) as they use CPU
Priorities of interactive threads go down (boost)
Reentrant and preemptible.
Windows Scheduling
The thread timeslice (quantum) is 10ms-120ms. When quanta can vary, has one of 2 values. Depends on the OS version, and a dialog setting.
Supports symmetric multiprocessing (SMP)
Up to 32 processors on 32-bit windows,
Up to 64 processors on a 64-bit Windows
All CPUs can take interrupts
Supports Non-Uniform Memory Access systems
Scheduler favors the node a thread perfers to run on
Memory manager tries to allocate memory on the node a thread prefers to run on.
Supports Hyperthreading
Scheduler favors idle physical processors when it has a choice
Doesn’t count logical CPUs against licensing limits.
Linux Scheduling
The thread quantum is 10ms-200ms
Default is 100ms
Varies across entire range based on priority which is based on interactivity level
Higher priority threads have a longer quantum, CPU bound threads with low priority have smaller quantums
Supports SMP systems
No upper CPU limit: set as kernel build constant
All CPUs can take interrupts
Supports NUMA systems
Scheduler favors the node a thread last ran on,
Memory manager not NUMA aware.
Windows Memory Management
32-bit systems split user-mode/kernel-mode from 2GB/2GB to 3GB/1GB. User-mode memory is allocated in the lower end. System memory is kept in the upper memory range.
Allows easy communcation between user-mode and kernel mode processes
Supports 32-bit or 64-bit systems, copy on write, shared memory and memory mapped files.
Per-process working sets: Working set tuner adjusts sets according to memory needs using the “clock” algorithm, removing things that the process accessed least recently.
No swapper
Linux Memory Management
Splits user/kernel from 1GB/3GB to 3GB/1GB.
2.6 kernel has a 4/4 split option where kernel has its own address space.
Supports same modern features of memory manegers like Windows.
Global working set management uses “clock” algorithm. This means there is no set cap on a given task in Linux. This means that it uses the least-recently-used memory page on the entire system, instead of just the local task’s memory set.
No swapper (the working set trimmer code is called the swap daemon, however).
Windows I/O
Centered around the file object
Layered driver architecture throughout all driver types. Gives the ability to write drivers that can insert above or below other drivers dynamically. This is taken advantage of by on-access virus scanners, which insert a new driver above the file IO driver, allowing for on access scans.
Most IO supports asynchronous operation.
Internal interrupt request level (IRQL) controls interruptability. Device driver handles initial request with all interrupts masked, but then returns control to the kernel before doing more intensive operations, so that interupts can be enabled while it is processing.
Supports plug and play.
Linux I/O
Centered around inode
No layered I/O model – on access virus scanning on Linux would be through some sort of a hack.
Most IO is synchronous: Sockets, Direct Disk I/O are the exceptions. V2.6 kernel adds new support for asynch IO, but not enabled in most places. This is here for benchmarking tasks like databases.
Internal interrupt request level (IRQL) for interruptability
Interrupts are split between ISR and soft IRQ/tasklet.
Supports plug-and-play.
Windows File Caching
Single global common cache
Virtual file cache, caching is at file vs disk block level. Cached at the file level, offset, etc instead of the disk block level.
Supports zero-copy file serving. Can send data directly from the cache without in-memory copy to internal buffers.
Linux File Caching
All identical to Windows. Previously it was block-level caching, but in 2.4 and 2.6 they have it all moved over to virtual file cache.
Windows Security
Based on aCLs and very flexible
Users are defined with priviledges grant rights, member groups.
Security can be applied to any object manager object: files, processes, synchronization objects, etc.
Supports auditing out of the box.
Linux Security:
Two models: Standard UNIX model (U/O/G/E), v2.6 adds Access Control Lists (SELinux)
Users are defined with capabilities and group memberships
Security is implemented on object-by-object basis
As no built in auditing support
Version 2.6 includes Linux Security Module framework for add-on security models. Lets 3rd party security models be written to provide any type of security layer, hooks throughout the kernel to support this.
Linux’s Evolution Towards Windows:
I/O Processing:
Linux 2.2 had notion of bottom havles for low-priority interrupt processing, with fixed number of BHs, only one BH of a given type could be active on a SMP (or a single proc system)
Linux 2.4 introduced tasklets, which are non-preemptible procedures called with interrupts enabled
Tasklets are the equivalent of Windows Deferred Procedure calls
Kernel Reentrancy
Much of 2.2 Linux was not reentrant. This means a single CPU can be in the kernel at a time.
Ingo Molnar stated in rebuttal: “his example is a clear red herring.”
A month later, he made all major paths reentrant.
Kernel Preemptibility
Through the base release of 2.4, Linux was only cooperatively preemptible. There are well-defined safe places where a thread running in the kernel can be preempted. The kernel is preemptible in v2.4 patches, and in v2.6.
Windows NT has always been preemptible.
Per-CPU Memory Allocation
Linux 2.4 introcued per-CPU kernel memory buffers. Windows had these in NT Service Pack in 1997.
Scheduling
The 2.4 scheduler is O(n). If there are 10 active tasks, it scans 10 of them in a list in order to decide which should execute next. This means long scans and long durations under the scheduler lock.
Linux 2.4 with patches from Ingo Molnar created a O(1) scheduler, using ordered lists by priority. Has per-CPU ready queues where the tasks are presorted.
Windows NT has always had O(1) scheduler based on pre-sorted thread priority queues.
Server 2003 introduced per-CPU ready queues. XP and previously Oses do not have this ability. Linux load balances queues whereas Windows does not.
Zero-Copy SendFile
Linux 2.2 introduced Sendfile to efficently send file data over a socket. Sendfile API actually did a copy operation to the network buffer before the send.
Linux 2.4 introduced zerocopy version.
Windows NT pioneered zero-copy file sending with TransmitFile, the Sendfile equivalent, in Windows NT 4. This was the API from the secert MS-Netscape meetings discussed in the Antitrust suit
Wake-one Socket Semantics
Linux 2.2: Thundering Herd or overscheduling problem: In a network server application there are typically several threads waiting for a new connection. In 2.2, when a new connection came in all the waiters would race to get it, but only one could.
Ingo’s response: 5/2/99 “here he again forgets to _prove_ that overscheduling happens in Linux”, 5/7/99: “as of 2.3.1 my wake-one implemntation and waitqueues rewrite went in”
Always been in Windows NT.
Asynchronous I/O:
2.2 Only supported few async I/O.
2.6 adds asynchronous I/O for direct disk access, AIO model introduced with efficent management of async I/O. Also added alternate epoll method.
Windows I/O is inherently asynchronous. Windows has had completion ports since NT 3.5. More advanced form of AIO. Completion ports are patented, does AIO infringe?
Light-Weight Synchronization
Linux 2.6 introduced Futexes: There’s only a transition to kernel-mode when there’s contention. Futexes = Fast Mutexes.
Windows has always had CriticalSections which have the same behavior.
Futexes go further, allowing for prioritization of waits. Works interprocess as well, where CriticalSections only work in the same process.
What do the benchmarks say?
TPC-C, SpecWeb99, Netbench were looked at. Big caveats:
Getting competitive published results is expensive.
Benchmark results reflect the performance of the hardware, operating system, and how well an application takes advantage of the scalable CPU, memory, and I/O mechanisms of the OS.
Look at the Future:
The kernel architectures are fundamentally similar
For the next 2-4 years, Windows has and will maintain an edge:
Linux is still behind on the cutting edge of performance tricks
Large performance team and lab at Microsoft has direct ties into the knernel developers
As time goes on the technological gap will narrow
Linus Rebuttal:
Fast Process Creation: Linux is faster, but not such a big deal (?) in Enterprise computing, IIS vs Apache,
Generic File Descriptors: (ie pipes and network connections that act like files)
Remote and Scriptable Management
True Network Transparent Window Management
Multi-filesystem FS layer with sane locking: NT does have a big issue here. Is this really important to support dozens of file systems?
Good directory lookup caching
See how much linux is like windows and how much catching up it does?
Now with Vista, Linux is again falling behind in GPU functionality, networking etc etc. Once Longhorn server is out, i bet people will see new Era of Windows.
Maybe you could have linked a webpage instead of posting 4 damn posts?
Maybe you could have linked a webpage instead of posting 4 damn posts?
Yes Mr Smart Ass aka Soulbender i would have but this article is saved on my computer and i don’t think it is available online, that’s why i pasted it here.
BTW it is funny to see how GPL fanboys are feeling chilli in their ass seeing linux kernel doing catch with NT:)
Edited 2006-12-06 07:59
“Yes Mr Smart Ass aka Soulbender i would have but this article is saved on my computer and i don’t think it is available online”
Never heard of Google? Searching for “Mark Russinovich” “A Tale of two Kernels” yeilds quite a few hits. So instead of a annoying people with 4 not very readable posts you could have linked to an easy to read and follow article.
I’m not a fan of the GPL nor a Linux user, btw.
Find me a link which has this complete information on the article from Mark and then open your mouth.
http://www.schrankmonster.de/PermaLink,guid,b543be1b-068a-4161-8e83…
First hit.
It doesn’t have any details which shows how linux is catching up
The link you point misses what i wanted to point out. So now get back to google and get me the result.
http://ryangregg.com/archive/2004/05/24/871.aspx
It doesn’t have any details which shows how linux is catching up
The link you point misses what i wanted to point out. So now get back to google and get me the result
What you pointed out is a very well known article unfortunately for you.
It is known and was proved that :
– the author is/was a Linux ML troll that got burned lots of time
– his papers taste more like a vengeance
– most of what he said concerning Linux was outdated when the paper was out
– it’s completely outdated now
He wanted to say that Windows kernel is very good in theory, and that’s all.
The problem, that he always try to avoid, is that even in 2.2, Linux was running circles around the better Windows kernel in most loads.
In 2.4, Linux completely left the better Windows kernel behind, performance wise.
Now with 2.6, they don’t even dare compare, MS prefers referring to (wrong) settings using kernel 2.4.
So I think you are completely mistaken as to which kernel has catchup to do.
A lot of people will tell you that “theoretically” Windows kernel is better, but practical evidence shows this doesn’t hold up in reality.
CrazyDude0,
The industry copies each others innovations very often. Microsoft copied Linux’s Sudo in UAC, Mac OS X widgets; Microsoft even used some of FreeBSD’s code in Vista’s new TCP/IP stack.
So…
Linux will do the same but much faster. I figure Linux will rival Vista (in terms of kernel security features, performance improvements, etc.) within several months. Desktop environments will also be on a level playing field when KDE 4.0+ and Gnome 2.20+ are shipped.
GNU/Linux is already far more advanced than Windows and Mac OS X when it comes to GPU-utilizing eye-candy GUIs. This trend will only continue with upcoming releases of Gnome and KDE.
Xaero_Vincent: This is what you believe that Linux will be on par with Vista. It is ok as it is ok to live in fantasy world. Linux is no where close to Vista in GPU utilization, can you give some examples or it is just fantasy again?
BTW the claim for FreeBSD in Vista is pretty lame and fake. Vista has a brand new TCP/IP stack with full IPv6 support not something you will find in BSD or Linux.
Neither will you find cool stuff like TCP offload engine, RSS or other advanced hardware technology support in BSD or Linux.
You are mixing sudo with UAC when they are not same. Tell me how you can run a process in Linux under a less privileged user automatically if you are running as admin?
Sudo actually allows you to run as admin when you are running as normal user. Windows 2000 had that too in RunAs so no it is not new and *nix had similar concept with s bit as well.
“Vista has a brand new TCP/IP stack with full IPv6 support not something you will find in BSD or Linux.”
This is a pretty lame and fake claim too. *BSD (not sure about Linux) has had full IPv6 support in their TCP/IP stacks for quite some time.
“neither will you find cool stuff like TCP offload engine”
More fake and inaccurate claims. Many *BSD (dunno about Linux) drivers supports TCP offloading when the NIC hardware supports it. This has been available for a long time.
“RSS”
Uh, do you know what RSS is? It’s an XMl format and has nothing to do with hardware. Unless you mean some other RSS that I’ve never heard of.
“Tell me how you can run a process in Linux under a less privileged user automatically if you are running as admin?”
For one noone runs under admin/root in Linux by default and even if they did it would be simple to make the Firefox/Konqueror/Whatever shortcut do a “sudo -u <username>”.
“Vista has a brand new TCP/IP stack with full IPv6 support not something you will find in BSD or Linux.”
This is a pretty lame and fake claim too. *BSD (not sure about Linux) has had full IPv6 support in their TCP/IP stacks for quite some time.
so bsd has it but it is not on by default thus i assume not well tested. Linux doesn’t have it. So i am at least 50% correct.
“neither will you find cool stuff like TCP offload engine”
More fake and inaccurate claims. Many *BSD (dunno about Linux) drivers supports TCP offloading when the NIC hardware supports it. This has been available for a long time.
Wow so you claim that BSD has a well designed TOE support in their TCP/IP stack? Something which is actually patented by others is in BSD? Or do you mean that it is upto individual drivers to hack TOE in? Vista has a well designed and well defined TOE support in the TCPIP driver. All big network card vendors like Broadcom, Intel will support that. Show me that on BSD or Linux please?
“RSS”
Uh, do you know what RSS is? It’s an XMl format and has nothing to do with hardware. Unless you mean some other RSS that I’ve never heard of.
No it is receive side scaling. It is used so that a network card can direct interrupts to different processors based on TCP connection information.
“Tell me how you can run a process in Linux under a less privileged user automatically if you are running as admin?”
For one noone runs under admin/root in Linux by default and even if they did it would be simple to make the Firefox/Konqueror/Whatever shortcut do a “sudo -u <username>”.
According to you this is UAC?
Edited 2006-12-06 08:41
“so bsd has it but it is not on by default thus i assume not well tested.”
It’s on be default. The major Ipv6 reference work (KAME) is bsd. Microsoft hasn’t been all that involved in the creation of IPv6.
“Wow so you claim that BSD has a well designed TOE support in their TCP/IP stack?”
Hmm… I was thinking of TCP checksum offloading. My bad. However:
http://linux-net.osdl.org/index.php/TOE
http://www.usenix.org/events/hotos03/tech/full_papers/mogul/mogul.p…
TOE is no magic bullet.
“No it is receive side scaling”
AH, that RSS.
“According to you this is UAC? ”
You didnt ask for UAC, you asked how it could be done.
Edited 2006-12-06 09:01
Sorry to disapoint you but windows was always using a crappy BSD ripoff of TCP stack. Vista may have it rewritten (and man was that shit in need of that) but I highly doubt windows will ever get to the level of stability performance and general quality considering networking as freeBSD.
You’re comparing a Trabi to a tank here. Windows is simply not a networking specialist.
so bsd has it but it is not on by default thus i assume not well tested. Linux doesn’t have it. So i am at least 50% correct.
So Linux doesn’t have IPv6 support? Errrrrr, right. Whatever.
And a totally untested, untried and totally new network stack that was outsourced to India is a good idea……..because?
Vista has a well designed and well defined TOE support in the TCPIP driver. All big network card vendors like Broadcom, Intel will support that. Show me that on BSD or Linux please?
Support was actually written for TOE in Linux some time ago, but the patches were rejected simply because TOE has very little merit. I’m sure Microsoft thought it was a really great idea to rewrite their network stack to hack that in, but, well, most of us prefer stuff that actually works ;-). Reasons:
1. System hardware always catches up to TOE speeds in no time at all.
2. The Linux network stack is proven, people know how it works and it is consistent between systems. Handing that off to a piece of hardware that will more than likely not work as it should, will have quirks between different hardware and will need endless firmware updates like other pieces of hardware today is a really bad idea.
Maybe Microsoft should just write themselves a decent network stack?
3. TOE requires more hardware programming to set up and destroy connections. Connection intensive protocols often underperform.
4. TOE NICs simply do not have the memory or CPU available to them to work effectively under very heavy loads. Under 100% load a TOE NIC will fall back to the software stack, defeating the object of using TOE in the first place.
5. Resource based denial of service attacks on individual and poor performing NICs. The NIC will either get flooded totally or bounce back to the software stack, making the DoS attack even worse.
6. No guarantee whatsoever that any given NIC is RFC compliant. Linux’s network stack is.
7. Netfilter, packet filtering and QoS software in Linux is very well proven. This would have to be rewritten to support TOE and vendor specific hardware for no benefit whatsoever, as do all the tools in Windows, which again are unproven.
TOE is all about vendor lock-in using hardware again, and it will most likely have many people scratching their heads over inexplicable network problems caused by the vendor specific implementations in the card. It just makes a NIC more expensive to produce with no appreciable benefits and massively increased support costs.
TOE is a disaster waiting to happen, Linux was right to reject it and for you to proclaim it as a holier than though benefit is so laughable it isn’t funny. Now that you’ve mentioned this I’m less keen to use Windows and the Longhorn Server for anything network related.
No it is receive side scaling. It is used so that a network card can direct interrupts to different processors based on TCP connection information.
Funny. Very, very funny. RSS lets the TCP stack run on multiple processors, which Linux has been able to do for years. It may be Earth moving for you in the Windows world, not so for the rest of us.
Tell me how you can run a process in Linux under a less privileged user automatically if you are running as admin?…….According to you this is UAC?
UAC. Oh, you mean that piece of unproven crap that everyone has complained bitterly about with Vista?
If you don’t think Linux or a Unix system can run a process under a different user when running another as root, you haven’t the faintest idea what you’re talking about. It’s been one of the biggest problems with Windows over the years, because it will simply not reliably run well when logged in as an ordinary user and ‘Run…as’ is simply a poor implementation of su and sudo that doesn’t work a lot of the time.
You’re full of it, I’m afraid.
– So Vista stack is developed in India. How do you know? How about you acknowledge that you are just lying because i know it is developed in Redmond.
– Wow so according to you TOE is bad, let us enumerate the cases:
1. System hardware always catches up so you want your general purpose CPU to do all the processing when you can have a hardware specially designed for networking. Nice try but it falls flat because TOE is in demand.
2. Hardware netowkr stack is not reliable, that is just your assumption, no proof, pure FUD. BTW most of the TCP stack handling like connection management is done in software. Hardware NIC only does sending packets, receving them and assmebling them.
3. TOE is bad for connection intensive protocols, ok which protocol is connection intensive? I think you meant connection intensive workloads or short lived connections, right? Yes TOE is not for webserver type workloads where data trasfer happens in small amounts but database server hell yeah.
4. So just because you can’t have all the connections fast, does it mean you don’t make the ones faster that you can? Again same flawed logic. Think of TOE in enterprise database servers, i think TOE is pretty useful.
5. Why will it make DOS attack worse, if a NIC doesn’t have TOE how does that help? Are you spreading FUD?
6. RFC compliant…point … MORE FUD.
7. So what if they needs to be re-written. Linux kernel is updated so many times that it breaks binary compatibility and then you think it is ok? With your logic, Linux kernel itself is not stable.
TOE is not vendor specific lock-in because it is a specification and any hardware vendor can implement it. Sorry if Linux doesn’t have any equivalent, it is not Microsoft’s problem.
– RSS – I am not talking about procesisng packets on multiple processors. I am talking about a NIC itself indicating packets on different processors. Think about it, if the NIC doesn’t do it, then you would have to do that in software. You would need to store TCP connection information and distribute them in software on multiple processor to avoid out of order recieve. That is pretty expensive and that is why RSS is there. Also in software version only one CPU will have to process all the interrupts but not in RSS enabled NIC. Do you get it now?
– UAC – Windows could run process under different user since windows 2000 as well. Did you hear of RunAs? UAC is about making sure that the applications run under normal user even if the application is badly coded and assumes that it is running under admin account. Windows does this to give a better user experience, what does Linux do? Oh well try 100s distro and see which one works…nice.
Edited 2006-12-06 10:56
Wow so my last post is modden down and read it and wonder why.
I don’t say bad about anyone, i am not offensive and i am not offtopic. Still the one that says TOE is bad gets modden up and the one that shows TOE is better gets modden down. Just goes to show the honesty in fanboys:)
“Still the one that says TOE is bad gets modden up and the one that shows TOE is better gets modden down”
I think it’s too early to say either way and even if it would be beneficial for Windows that doesnt mean it would be beneficial for other OS’s that are using different stacks and kernels.
Same goes for “RSS”.
You’re not really one to talk about “fanboys” though, you come across as a pretty big fanboy yourself in your posts.
You talk about how “advanced hardware support” isn’t available in BSD and Linux. Hmm..guess who had a generic framework for hardware accelerated cryptography long before Windows? Why has Windows still (with XP) such substandard IPSEC support (3DES? In 2006?) when both *BSD and Linux has much more mature and secure implementations?
Point being, different technologies develop at different speeds for different platforms.
I know the XP supports full IPSec in hardware. I don’t know when did BSD support IPSec in hardware, do you know? Which year?
According to http://en.wikipedia.org/wiki/IPSec Linux only supported IPSec in 2.6 so what’s your point?
Why no appreciation for Windows supporting TOE, RSS etc in hardware?
Let us see how mods me up or down;)
“I don’t know when did BSD support IPSec in hardware, do you know? Which year?”
OpenBSD has supported IPSec since 1997 and has supported hardware accelerated IPsec since 2000.
“According to http://en.wikipedia.org/wiki/IPSec Linux only supported IPSec in 2.6 so what’s your point?”
That the strongest cipher MS supports is 3DES and
and freeSWAN has been around for longer than kernel 2.6.
“I know the XP supports full IPSec in hardware.”
MS IPsec doesn’t support anything stronger than 3DES and considering how abysmally slow 3DES is that is a good thing. Meanwhile, *BSD, Linux and many others has moved on to the much, much faster and much more secure AES and Blowfish
“Why no appreciation for Windows supporting TOE, RSS etc in hardware?”
Since I don’t use Windows, especially not for networking tasks, i have no reason to “appreciate” it since it doesn’t concern me.
If it makes networking better and faster on Windows (something it desperately needs) that’s good for Windows users but it does not necessarily affect other OS’s.
Edited 2006-12-06 13:19
Again, this constantly flipping around between Linux as kernel and Linux as OS is starting to rear its ugly head. IPsec has been supported under Linux (the OS, and also the kernel if you count external modules) since the inception of FreeS/WAN somewhere in the nineties. This is also mentioned in that Wikipedia article. FreeS/WAN ended, and now continues development (yes, even with “native” support) under the name of Openswan. Even without counting FreeS/WAN and its successors, there have been IPsec implementations in Linux long before Microsoft became security aware all of a sudden.
Please compare apples to apples, and not switch back and forth between various interpretations of the term “Linux” as you see fit to support your arguments.
As said before, TOE *is* hardware. All you need is drivers to support it, and those are available to OSes which make sense to use in a datacenter where the use of TOE makes any sense in the firstplace. Yes, that includes Linux.
I know the XP supports full IPSec in hardware. I don’t know when did BSD support IPSec in hardware, do you know? Which year?
According to http://en.wikipedia.org/wiki/IPSec Linux only supported IPSec in 2.6 so what’s your point?
Windows *XP* supports full IPSec in *hardware*.
You’re really an amazing guy. Do you understand what a driver is ?
The rest is the same kind of rubbish. Really we can’t have a serious discussion.
Well, a few actual facts might help…true facts…that accurately reflect the real state of things…unfuddingly.
…and the one that shows TOE is better gets modden down.
Just where have you shown anything? We’ve had TOE NICs before, and they’ve all been failures because of the economics involved – duplication of effort and support. So will it be with the new TOE NICs people are dredging up today. Windows having support for them is no indication whatsoever of any new innovation in Windows – far from it – that you seem to sadly believe is the case.
Windows stealing a march, or Microsoft having the vaguest idea of what’s actually going to work in the world of networking will be the day that glaciers start appearing in hell.
End.
Sadly, I really do think you believe this bull.
Ok everyone in Broadcom, Intel is stupid and you are intelligent segedunum.
Linux keeps catching up thats why they got UDP fragmentation in hardware and TCP checksum in hardware.
Anyways who cares what 1% linux guys feel, TOE is here to stay and with Windows support behind it, it is already a success:) No matter how much you whine. And yes Linux will eventually get support for TOE just like other hardware offloads. Hopefully vendors don’t have to write hacked drivers on Linux anymore. Ahh it suck to develop on such a poor platform. But i am sure eventually linux will get these features because that’s what Linux does, play catchup to windows and solaris
Edited 2006-12-06 22:24
Well said.
Ok everyone in Broadcom, Intel is stupid and you are intelligent segedunum.
Linux keeps catching up thats why they got UDP fragmentation in hardware and TCP checksum in hardware.
Anyways who cares what 1% linux guys feel, TOE is here to stay and with Windows support behind it, it is already a success:) No matter how much you whine. And yes Linux will eventually get support for TOE just like other hardware offloads. Hopefully vendors don’t have to write hacked drivers on Linux anymore. Ahh it suck to develop on such a poor platform. But i am sure eventually linux will get these features because that’s what Linux does, play catchup to windows and solaris
Catchup to Solaris? I’m almost too busy pissing myself laughing to be able to respond coherently to your bull – er, post.
There’s an analogy to all this pontificating, which we’ve seen before with the supposed advantages of “WinPrinters” and “WinModems” and “WinGodKnowsWhatElse”. Proprietary products that are utterly useless without Windows.
In 1988 IBM was already losing – no, make that “had already lost” – the lion’s share of the PC market to the clone-makers. And DEC was still dithering over “whether a computer company could be successful without owning its own architecture”. And yes, if that sounds tremendously daft now, it is because that proprietary dithering is exactly why DEC is No More.
So excuse us if we carry on supporting open standards while you WinUsers continue experiencing massive lossage.
twenex – tell me what you are smoking, i wanna get half of that too
WinUser lossage lol…last i checked Windows is not losing share on desktop and gaining in server…again what are you smoking?
WinUser lossage lol…last i checked Windows is not losing share on desktop and gaining in server…again what are you smoking?
The lossage is in the vendor lock-in, you f–kwit. If over a billion Chinese live under communism, does that mean everyone else is a fool not to move to Shanghai?
So excuse us if we carry on supporting open standards while you WinUsers continue experiencing massive lossage.
massive lossage?
well if 1% is massive to you, its ok.
massive lossage?
well if 1% is massive to you, its ok.
If you weren’t blind, and either obtuse or just plain stupid (both?), you’d realise the lossage is in the vendor lock-in.
If you weren’t blind, and either obtuse or just plain stupid (both?), you’d realise the lossage is in the vendor lock-in.
oh, now Im blind and obtuse or stupid, adn you are being modded up after insulting, wow GPL advocates are so classy. .. not.
Hey where did da intelligent segdemunum go?
I think seeing the linux kernel supports many of the hardware offloads that were first available on windows, he finally got embarassed of his knowledge or lack of it:) and decided to run away with his tail between his legs.
Actually he is googling for useless stuff he can link in his next post.
Actually he is googling for useless stuff he can link in his next post.
LOL well said. Hey intelligent segdemunum we are waiting for your insights, where are you?
Sadly, I really do think you believe this bull.
Of course he does, he’s a MS supporter. Microsoft supporters are a perfect example of the Baldrick mentality:
Blackadder: PLEASE try to think, Baldrick; thinking is SO important! What do YOU think?
Baldrick: I think thinking is SO important, my Lord.
Oh yeah, and insulting makes you superior, go GPL advocates!!.
Oh yeah, and insulting makes you superior, go GPL advocates!!.
Actually, as you well know, the ones who do the most insulting around here are the increasingly desperate GPL-and-OSS-haters, who between the success of Linux, BSD, Solaris and FOSS software for Windows have less and less ground to stand on every day.
Keep on insulting and panicking and pissing in the wind. It’s entertaining to bystanders.
Actually no, I think we feel your frustration and desesparation of still being a minority w/o voice after years and years of trying to caching up the basics, let alone the complicated stuff.
Actually no, I think we feel your frustration and desesparation of still being a minority w/o voice after years and years of trying to caching up the basics, let alone the complicated stuff.
Gee, could that be because we don’t have a massive FUD-spreading marketign machine? People would rather believe a comforting lie – that the majority of people who use computers are using decent products, rather than being conned by a third-rate company into using obsolete and flaky crap – than the truth.
As for “trying to catch up to the basics and the complicated stuff”, er, exactly WHAT basics and complicated stuff can Linux NOT do?
Oh, you must mean BSODs.
Gee, could that be because we don’t have a massive FUD-spreading marketign machine?
Well, the Novell-MS deal show the contrary and wow, you really know how to spread FUD.
People would rather believe a comforting lie – that the majority of people who use computers are using decent products, rather than being conned by a third-rate company into using obsolete and flaky crap – than the truth.
yeah, the world is wrong and them GPL cult is right, mmmm, I’ve seen that kind of behavior before, oh yeah, the Jehova withnesess.
As for “trying to catch up to the basics and the complicated stuff”, er, exactly WHAT basics and complicated stuff can Linux NOT do?
Sorry I really have no time to list the 500+ issues, maybe next time.
Oh, you must mean BSODs
me no comprende, sorry.
Congratulations, you’re getting more and more desperate and less and less worth replying to with every post.
Congratulations, you’re getting more and more desperate and less and less worth replying to with every post.
Yeah you are so righ, really, whatever you say is the truth, yeah.
Mitrai, it would be better to talk to a buffalow than twenex. He is a GPL fanatic.
Check his history of comments, he is never constructively appreciated Microsoft or their technologies which goes to show that he is not honest and he can’t give credit where credit is due.
Mitrai, it would be better to talk to a buffalow than twenex. He is a GPL fanatic.
Twenex, it would be better to talk to a buffalo than Mitarai. He is a closed-source fanatic.
Check his history of comments, he is never constructively appreciated Microsoft or their technologies which goes to show that he is not honest and he can’t give credit where credit is due.
Check his history of comments, he is never constructively appreciated Linux or its technologies which goes to show that he is not honest and that he can’t give credit where credit is due.
Check his history of comments, he is never constructively appreciated Microsoft
That’s because there is nothing to appreciate
or their technologies
Perhaps you mean “their ripoffs”
which goes to show that he is not honest
Are we using Microsoft and their shills as a yardstick of honesty, here? Oh dear!
and he can’t give credit where credit is due.
I do give Microsoft credit. They have built a massive and very successful empire out of strong-arm tactics, FUD, and substandard products.
Well, the Novell-MS deal show the contrary and wow, you really know how to spread FUD.
MS and Novell: “We have signed a deal to promote an OS Microsoft hate, which presents a threat not to their existence but only to their utter and total domination of the software industry”
FOSS community: “Hmm, something must be fishy here”
“Pragmatists”: “Shut up, you stupid conspiracy theorists.”
MS: “Oh yeah, and we PWNZ L1NUX! yeah! All your code is belong to us!”
FOSS community: “Hmm. stupid conspiracy theorists, eh?”
Yeah, that sounds uber-Fuddy.
yeah, the world is wrong and them GPL cult is right, mmmm, I’ve seen that kind of behavior before, oh yeah, the Jehova withnesess.
Over a billion Chinese live under communism. Are you going to move to China to become part of the majority then?
Sorry I really have no time to list the 500+ issues, maybe next time.
Translation into Truthspeak: There aren’t any.
Edited 2006-12-06 23:27
Oh yeah you are so right, everything you say is the truth really. oh yeah.
Now you’re really running out of arguments.
Oh, I’m sorry. I mean, “running out of ‘arguments'”.
Blah, blah, blah.
Oh yeah you are so right, everything you say is the truth really. oh yeah.
Can’t take failure, eh?
Oh yeah you are so right, everything you say is the truth really. oh yeah.
I’m wondering why you’re bothering at this point…Enjoy embarrassing yourself?
Edited 2006-12-06 23:32
Oh yeah you are so right, everything you say is the truth really. oh yeah.
Competent argumentation at it’s finest.
Btw, my dad is stronger than your dad, nay nay nay.
Oh yeah you are so right, everything you say is the truth really. oh yeah.
Nice try but it falls flat because TOE is in demand.
TOE is not in demand, nor is it anything new. We’ve had TOE 10 Mbit cards, TOE 100Mbit cards and TOE 1 Gbit cards. Hands up who thinks that TOE on these cards here are cost effective, reliable and are worth the money? What, no hands?
Hands up also who thinks that each vendor replicating a network stack in hardware is a good thing? Wow. No hands either.
Hardware netowkr stack is not reliable, that is just your assumption, no proof, pure FUD.
So you believe different hardware based network stacks are more proven than the Linux, BSD, Unix or even Windows network stacks? Right……….
Oh, and do go and look up the definition of FUD. It doesn’t apply here.
TOE is bad for connection intensive protocols, ok which protocol is connection intensive?
Some little thing called HTTP.
Yes TOE is not for webserver type workloads
So why is this useful again?
So just because you can’t have all the connections fast, does it mean you don’t make the ones faster that you can?
Strawman argument which is knocked down by the fact that the costs of doing this outweight the benefits – they always have.
Why will it make DOS attack worse
Resource based algorithmic attacks on NICs themselves.
RFC compliant…point … MORE FUD.
Ooooh, yer. People really fear and are uncertain of the truth.
True I’m afraid. There is no guarantee whatsoever that any given hardware based network stack will be RFC compliant. Diversity is not a good thing here.
So what if they needs to be re-written.
Because it works and it works damn well. You can wander off to the Windows world of piss-poor networking if you want to. No change there.
TOE is not vendor specific lock-in because it is a specification and any hardware vendor can implement it.
History tells us that different implementations of supposedly the same standard work out badly. We’ve had TOE NICs before and they’ve been poor.
Jesus. We don’t want Microsoft and hardware vendors f***ing up the now stable world of ethernet like we used to have in the early years.
I am not talking about procesisng packets on multiple processors.
Yes you are.
Think about it, if the NIC doesn’t do it, then you would have to do that in software.
That would be so bad wouldn’t it? And you think that dumping this stuff over the hardware interconnects is good? Right……
That is pretty expensive and that is why RSS is there.
Maybe in your world.
Also in software version only one CPU will have to process all the interrupts but not in RSS enabled NIC. Do you get it now?
Maybe in Windows’ piss poor current implementation. Not in everyone elses’. It’s not even worthy of a reply to be honest. You don’t have the faintest idea what you’re talking about.
Did you hear of RunAs?
Hang on. That’s what I said wasn’t it? It’s piss poor, especially where you are logged in as yourself and need to run a particular application that needs resources like mapped drives that are assigned to you when you are logged in. It fails here. It’s not something Microsoft recommends either, nor do they build their admin tools with it in mind as you have to do in the Unix and Linux worlds.
UAC is about making sure that the applications run under normal user
Did you read tha part about UAC being crap on every version of Vista so far?
Windows does this to give a better user experience
Does it really? Everyone really seems to love those popups!
what does Linux do? Oh well try 100s distro and see which one works…nice.
Oh dear. Sadly, sudo and su works in exactly the same way on all distros. Bugger.
You’re full of it, quite clearly. Sadly, you even believe your own bull.
Edited 2006-12-06 22:06
so bsd has it but it is not on by default thus i assume not well tested. Linux doesn’t have it. So i am at least 50% correct
You are 120 % wrong !
You are so clueless I’m amazed you dare speak about this subject : do you believe everyone is stupid here ? That there are no knowledgeable people ?
Hear well : Linux had IPv6 stack since 2.2 (and before that in 2.1 dev versions). I can’t even tell you how many years ago this was.
Wow so you claim that BSD has a well designed TOE support in their TCP/IP stack? Something which is actually patented by others is in BSD? Or do you mean that it is upto individual drivers to hack TOE in? Vista has a well designed and well defined TOE support in the TCPIP driver. All big network card vendors like Broadcom, Intel will support that. Show me that on BSD or Linux please?
Wow, you should shut up right now. Really, you’re making a fool of yourself here !
TOE is not a standard, it has NOTHING to do in a TCP/IP stack implementation. It has to be in drivers.
It is not up to individuals to hack TOE in, but you’re right, TOE are a hack upon TCP/IP.
TOE are proprietary closed hacks to try to accelerate network traffic, made by the NIC vendors.
So TOE support can’t be “well designed” or “well defined”, it’s plain stupid to say that, as there’s no specification on this.
Each NIC vendor will do its own, and FYI, some NIC vendors provide closed drivers with TOE support for Linux.
You’re more than wrong again.
No it is receive side scaling. It is used so that a network card can direct interrupts to different processors based on TCP connection information
So don’t call that RSS. This is a Windows only marketing speak, that try to cope for Windows poor TCP/IP performance.
Of course, this also adds lots of latency.
“Tell me how you can run a process in Linux under a less privileged user automatically if you are running as admin?”
For one noone runs under admin/root in Linux by default and even if they did it would be simple to make the Firefox/Konqueror/Whatever shortcut do a “sudo -u username”.
According to you this is UAC?
Wow, so clueless !!
So when someone shames you with an answer, you then ask sth exactly like in Windows ?
You didn’t even realize that on Linux, it’s simpler than in your Windows example : you don’t need to have UAC at all, because you don’t have to run as root at all.
You could launch a session window to log into another user on your root GUI though (in both KDE and Gnome) if you really want to, or use su, or gksu, which won’t prompt you for any password, as you’re root already.
You just seem like a troll, Linux had IPv6 in the TCP/IP stack years ago. AIGLX and the X Window System model is far more superior than anything in windows, is not even comparable. They are both different things, Qt 4.3 is by far one of the best widgets and KDE 4 with Plasma, Phonon and D-BUS will rock our desktops. D-BUS, HAL, asynchronous booting are changing a lot in the desktop part, we already have a Composite/GL stack with AIGLX, we have nice toolkits like Qt 4.3 and GTK+/Cairo/Glitz, Aaron Seigo is a skilled programmer and artist and I’m sure he will do Plasma just great, it will even support Dashboard Widgets when it ship, just wait and see what I’m talking about and stop your troll circus.
MS is copying a lot from Linux and UNIX in general, since Windows NT come out.
Edited 2006-12-07 19:42
And we have root, not admin.
I figure Linux will rival Vista (in terms of kernel security features, performance improvements, etc.) within several months
Vista is playing catchup in the area you cited and hasn’t caught up at all, so what you say is wrong : Linux already outdo Vista in most area.
Didn’t you notice that for some years, Windows has no advantage over Linux except familiarity, monopoly and inertia ?
There’s years since MS stopped trying to use technical ways of saying Windows is better.
Desktop environments will also be on a level playing field when KDE 4.0+ and Gnome 2.20+ are shipped
I don’t think so, they are already better, KDE 4 and Gnome 2.20 will only make them way ahead.
For example, I hear now Vista can change the desktop language after reloading the session, a feature that Gnome (and KDE) had in 2001 !
I’m sorry, but your last sentence is wrong.
Windows Vista (which is a released product now) has far better technologies using the GPU for 2D rendering. For example, it can send glyphs as vertex data to the GPU and let it be rastered there. XGL can also accelerate text rendering on the GPU, but it relies on having the glyph available as pixmap first…
The open source community has the advantage of fast “release” cycles. So they can enhance their XLG/AIGLX code within some months and ship it. On the other side, there is lack of focus. So the X development was stuck for years.
So basically, Linux had compositing with GL first, but not stable and grown up. Now after Vista is out, XGL/AIGLX/compiz are clearly behind in terms of technology. But they have the opportunity to catch up very fast.
I don’t use any Windows version, but if you don’t have a look at it once in a while, you can indeed miss some (not many
) innovations…
“On the other side, there is lack of focus. So the X development was stuck for years.”
I would be tempted to argue that this statement is a little out of date. Lots of exciting things are happening with X right now, and its a lot to do with a fork to X-org and even more to do to moving down the modular path.
I’d be tempted to say that freedesktops standards are getting a rid of a lot of the annoyances that have plagued linux for years.
Although I am interested how is “XGL/AIGLX/compiz are clearly behind in terms of technology” is that even true?
@gonzo Linux is not a clone of Xenix
Edited 2006-12-06 14:28
Well, I said it _was_ stuck. You’re right, since the fork of Xorg, this is history
Well, Windows Vista has mastered some hurdles that aren’t solved in X yet. For example, having direct rendered GL applications inside XGL/AIGLX. Or at least, if rendered indirectly, have them access OGL 2.0 + Extensions and not only the GL 1.4 subset.
Also, video acceleration within the GL accel. X server doesn’t work stable / reliable yet.
What I also pointed out already, are further advanced shaders, for example for font rendering:
“Video cards that support Direct3D 10 will cache fonts in video memory and perform all ClearType text rendering in hardware.”
From: http://en.wikipedia.org/wiki/Features_new_to_Windows_Vista
One big problem of Xorg are still video drivers, esp. the closed source ones. Major changes in the drivers or their APIs aren’t easily done as ATI, but also nVIDIA take long to support them, if at all. Contrary, Microsoft just told video card vendors that a new API for allowing direct GL rendering is needed – and they implemented it ready for release.
“Well, Windows Vista has mastered some hurdles that aren’t solved in X yet. For example, having direct rendered GL applications inside XGL/AIGLX”
Ok just had a go in a AIGLX enviroment GL works great. Thats simply not true.
“Also, video acceleration within the GL accel. X server doesn’t work stable / reliable yet.”
So your arguing that its behind windows because of stability.
“Video cards that support Direct3D 10 will cache fonts in video memory and perform all ClearType text rendering in hardware.”
Thats a difference that Aero has to AIGLX. It doesn’t *seem* something thats required for a 3D desktop. It sounds like I need an expensive card, and it is only a small part of a large implementation.
What I don’t get is. I have a cheap card that will not support DX10 it runs better effects than Aero, makes better use of the desktop space. I’m starting to believe that that Linux’s implementation is superior. I actually see surprisingly little discussion either way.
There is GL and there is GL. If you want to have OpenGL 2.0(!) and extensions, for example shader programs, the answer is tricky or simply “No”. This is a fact and you can do research on it. If you want to have OpenGL 1.4 only, the answer is “Yes”, but even Quake3 (as released in ’99) will produce wrong graphics (only visible on certain surfaces)!
About video acceleration. Sure, it’s definitely behind because of – but not only because of – lack in stability. But what I explicitely ment with video accel. was that it only works (fast enough) with some video cards, not others, and so on. For example, on my system, I have to use mplayer -o gl2 and still can get performance problems.
You’re right in that freeing the CPU from doing font rasterization is not “required”. But it’s clearly superiour to only do blitting and scaling on the GPU. And it’s difficult and tricky. So it’s a definite technology advantage. It’s a big step forward, if the main goal is (and it should be) to get as most graphics processing as possible away from the CPU onto the mostly unused GPU.
You’re completely right about the advantage of XGL/AIGLX on “legacy” hardware. It works great even on poor powered, integrated or old graphics chips.
Partly this is due to the fact that it doesn’t go as far as Vistas techniques, and in Vista their seems to be no effort to make it scalable at all.
Partly this could be politics, or whatever. The whole point is, Vista gets more stuff done in this sector in a more stable and clean way. The reasons for that are easy, and I assume most Xorg developer know them very well. Still, if Xorg stays on it’s actual pace, it should catch up very fast.
If you want to have OpenGL 2.0(!) and extensions, for example shader programs, the answer is tricky or simply “No”. This is a fact and you can do research on it. If you want to have OpenGL 1.4 only, the answer is “Yes”, but even Quake3 (as released in ’99) will produce wrong graphics (only visible on certain surfaces)!
There is no fact there, only a tricky answer, as in closed drivers having OGL 2.0 since 1+ year, but FOSS drivers don’t have it yet AFAIK.
About video acceleration. Sure, it’s definitely behind because of – but not only because of – lack in stability. But what I explicitely ment with video accel. was that it only works (fast enough) with some video cards, not others, and so on. For example, on my system, I have to use mplayer -o gl2 and still can get performance problems
?!!! This doesn’t depend at all on you using GL or not. Video acceleration works fine, and its problem is not instability (are you talking of Linux there ?).
I mean, its main problem is that all implementations of video accel on Linux (which means, uh, from one vendor only) are closed, and don’t work on all cards, as it’s buggy (I think).
But if it works, it works well.
You’re right in that freeing the CPU from doing font rasterization is not “required”. But it’s clearly superiour to only do blitting and scaling on the GPU
It’s not so clear to me. the scaling is trickier than what you think, as it needs lots of context, which is the largest problem with font rasterization IIRC. OK for the blitting.
And it’s difficult and tricky. So it’s a definite technology advantage
No, it’s not. Anybody in the field can do it. The problem is getting these damned drivers. The definite technology advantage is having the drivers right.
It’s a big step forward, if the main goal is (and it should be) to get as most graphics processing as possible away from the CPU onto the mostly unused GPU
Which is true for expensive GPU. On Linux, the goal is more to allow use of this for most architectures.
Partly this is due to the fact that it doesn’t go as far as Vistas techniques, and in Vista their seems to be no effort to make it scalable at all
Ah OK, so we agree on that.
The whole point is, Vista gets more stuff done in this sector in a more stable and clean way
I don’t think so. Developing for only one set of API you defined, on only one architecture won’t make you do stuff in more stable and clean way that the same thing done in multi-architecture environment, using a true standard.
I also disagree, because MS doesn’t let anyone see the implementation, or any mean to use it efficiently on other systems.
Actually, all this is supposition on my part, as no one can say if their stuff is stable or clean.
So in the field, MS is no help at all actually.
If FOSS drivers don’t have decent OpenGL this is definitively a problem! Saying “we don’t have it anyway” doesn’t help.
Sorry, but I didn’t talk about OGL 2.0 only. For example, I talked about Quake3. You can get Quake3 operate 100%, with all the bang and bliss, with the free Intel driver. But _not_ with indirect rendering under XGL/AIGLX!
Sorry, I think you got lost out of focus. What I told about was that hardware video acceleration gets broken under XGL. Not generally. And no, there is an open X extension for hardware video accel. and it is used in many drivers, free or proprietary. Additionally, there is vidix, which is also used by several drivers. Both have problems under XGL.
Rasterizing vector graphics _is_ a very costy task. If it can be done by the GPU, it _is_ a huge win.
“Anybody in the field can do it” – so did you try to implement glyph drawing in a GPU program or even wrote a single GPU program? I bet not, you even don’t know what’s all there in between several GL versions.
“The whole point is, Vista gets more stuff done in this sector in a more stable and clean way”
Well, if you compare what is there from Xorg and what is there in Windows Vista, at least for the end user it is more stable and clean in Vista. And it has nothing to do with the fact that it won’t help anybody but Microsoft.
I don’t like Microsoft, I don’t like proprietary drivers, closed hardware specs and de-facto standards. I really appreciate free software, open specs, real standards and cross-platform solutions.
But nevertheless, I have to admit Vista is in advantage when it comes to the 2D graphic acceleration under the hood. Only after I did extensive research about this topic and followed the evolution in the last months / years on both sides.
For example, “long time ago” Microsoft announced that OpenGL would be crippled just like it actually is under XGL. Everybody cried and shouted (me too), and some people still believe it is that way. But instead, they found a solution to the problem and implemented it (I think due to public pressure), which the Xorg people didn’t yet. Sadly enough.
Edited 2006-12-06 17:39
>> Now with Vista, Linux is again falling behind in GPU functionality,
Woo, that’s worth quoting… have you tried XGL versus Vista. I mean Vista looks ok, but usage of GPU-wise, Vista is the one trying to catch up…
Also, games under Vista are expected to be 10% SLOWER than on XP. Not exactly the blazzing speed superiority you implied…
So exactly what is it that Linux is falling behind ?
Also, games under Vista are expected to be 10% SLOWER than on XP. Not exactly the blazzing speed superiority you implied…
At least the games will be playable, Linux by the other hand mmmm.
What’s the point of all this back and forth “My OS is better than yours”?
While I strongly prefer Unix-like OSes, I have no problem with the fact that the Windows family enjoys some undeniable advantages.
This may seem a bit sappy, but wouldn’t we *all* be better off acknowledging the strengths and weaknesses of OSes that happen not to be our favorites, as well as those of the one(s) that do(es)?
I dislike Windows. But that is totally orthogonal to my opinions of where it shines and where it sucks.
Just to take this post deeper into sappy-land, I’ll say that I grew up with the original Star Trek series, and the concept of IDIC (Infinite Diversity in Infinite Combinations) has a deeply rooted appeal to me.
That said, although the number of commercial games available for Linux is quite low indeed, all the ones that I care about run natively under Linux due to the fact (OK, my opinion) that the best gaming houses have a respect for portable code.
e.g. I noticed the post that I am replying to while taking a break from Quake4.
Of course, depending upon your gaming tastes, YMMV. The games you like may *not* run under Linux. But that’s not the end of the world.
At least the games will be playable, Linux by the other hand mmmm.
1) There are tons of open-source games that go far beyond solitaire, minesweeper, and tuxkart.
http://www.happypenguin.org/
http://www.linuxgames.com/
2) There are commercial games by id Software, Epic Games and more.
http://www.tuxgames.com/
3) You can run many hundreds Windows games with Win32 and DirectX API emulation. Some games will even run using Windows inside of VMware, Parallels, and Xen.
http://www.transgaming.com/
http://www.codeweavers.com/
http://www.winehq.com/
Edited 2006-12-06 19:33
Those games sucks, honestly, givme the games I want not limited options.
Thx.
Doom3 and Quake 4 suck?
Enemy Territory: Quake Wars, Coldwar, X2: The Threat, Half Life 2, Medal of Honor: Allied Assault, etc. all suck!
Oh and all the Windows game you can run suck too!
Isn’t that right Mitarai?
You mean, I’ll be able to play all the DX10 games in Linux the time they are released?
W/O wathing months or years to be ported?
Those games you mention are 2 years old, where is the innovation the “up to date factor” , I’ll give you a hint, in Linux it doesn’t exist.
Go that? or do I need to write it with crayons?
Woo, that’s worth quoting… have you tried XGL versus Vista. I mean Vista looks ok, but usage of GPU-wise, Vista is the one trying to catch up…
XGL and AIGLX aren’t even needed with the NVIDIA driver. ATI drivers shall be similar once official support for the OpenGL “GLX_EXT_texture_from_pixmap” extension is incorporated.
Compiz and Beryl window managers wont be necessary either once the next version of Gnome and KDE are released.
Also, games under Vista are expected to be 10% SLOWER than on XP. Not exactly the blazing speed superiority you implied…
That with Desktop Window Manager (DWM) and Aero disabled. The performance is worse when they are running.
Likewise, I expect we’ll see a way of disabling compositing and effects while running games on GNU/Linux.
someone shoot this man before he can breed
This article lacks the critism it has to have. Just a flamebait.
Just because I’m interested. I’m sure there is a good reason for this, but people hate copying.
I love copying, I love copying even more when the final product is better than the original. This is part of evolution of any OS.
The other thing that confuses me, but its OS X uses that play the copying card. I would just once like to see a post that said.
“I bought my Macintosh 3 years ago it was a lot of money, compared to the genetic XP machines around, and I had to make compromises when it came to my choice of software, little did I realize then that I was using features that Microsoft’s Users will not see until next year almost 4 years later and will have to pay again for new hardware for a machine of the same age. Its turned out to be excellent value. My next machine will be a Macintosh. I will probably have a dual boot environment, so I don’t have to compromise this time around”
Never see a post like that ever, and I’m sure what I said is true.
what separates linux from unixes of the past
linux is foss
how can anything compete with a free product that can never be shut down simply by bankrupting a company
long live FOSS, fire up your rifles fellow jihadies, Microsoft is hunting down the “terrorists”
If osnews came in paper edition, i’d take out the pages of this article, and the article leading up to it, and use it for toilete paper.
Sorry to say, but their both that bad.
Edited 2006-12-06 07:24
The original article I think suffered from not being written by someone whose first language was english. Editing should have fixed all of that before it was submitted. Giveaways were using loose instead of lose and the whole WYSIWYG stuff. Grammar errors were the mainstay as well.
However his points, where he thought Linux desktops were weak in the market, were presumably:
1> Always behind windows, OSX, (other?) in terms of GUI because they are just copying stuff.
2> Unix was killed by fragmentation. Linux already has a fragmentation problem with the oodles of distros that are all different enough. It’s a PITA for anyone making commercial closed-source software.
3> Competetion from other niche OS’s. Linux has a tiny market share. Statistically it’s not much ahead of a dozen insignificant OS’s.
All of these points have weaknesses:
Linux is not just a copy of other desktops, at heart its a copy of UNIX. While you might argue that UNIX is an old crap dead OS, well Linux says otherwise.
Fragmentation is not that big an issue for open source software. Generally if something is popular enough it will be ported to another distro or another UNIX. Commercial software is SOL tho.
As for other niche OSs, it’s really hard to say what affect on any market they will have. Linux is probably the most technically advanced of the minority kernels and most of the desktop software is written for linux. (BSD people will disagree of course)
While the market share of desktop linux is tiny, it is HUGE compared to other niche desktop OSs
All that aside, the original article had it’s merits. Unfortunately it was far too simplistic and clumsy. It came off like flamebait, which presumably it is. Why else would you attack linux on OSAlert?
The problem is not the grammatical errors, or the first language stuff. People are actually very forgiving on here.
The main problem is it tries to cover too much lightly. There are lots of common…catchphases to cover up lack of substance which shows serious lack of research.
I’ll run through your points. Always behind windows, thats a strong statement if the GUI is the issue. How? Comparing something like File Manager with something like Nautilus is serious work.
Unix fragmentation is different from Linux fragmentation. The world then and the world now are two totally different environments. And thats a serious article all on its own…their are even benefits to fragmentation.
Competition from other niche OS’s. To be fair I see more niche OS’s than competitors. The reality is I see the same applications come up time and time again on all the different *kernels*. If linux got replaced by ReactOS tomorrow I’d use most of the same programs. It also leaves out that even a tiny percentage of the computer market is still lots of computers.
To be honest I would have been happy to see any of these topic come up…with perhaps less dramatic headlines. None of these points screams *Death* or *War*
Take a step back from you computer and consider for a moment “the War on
Terrorism”, “the War on Drugs” and “the War on Cancer.” Notice anything
about those wars? Sometimes, when policy wonks run low on rhetorical
ammo, they get lost in their own rhetoric and fall back upon terms
that, in the dictionary sense, don’t really apply. In essence, these
people dumb down their strategic initiative titles for mass consumption.
“The War for Linux” is merely another such misappropriation of rhetoric.
Do GUIs matter? Yes, but they matter more to some folks than they do
to others. Do ease of use, ease of installation and upgrade, robustness
and boneheadproofness matter? Of course they do, at least to some people.
For if the latest laundry list of positive O/S properties didn’t matter,
we all might well be running open source copies of GNUMVS on our desktop
370 clones… Not that there’s anything wrong with that.
MVS, Windows, MacOS and other proprietary operating systems share a trait
with each other, but not with Linux: Ownership. Nobody really “owns”
Linux, not really. Sure, Linux and GNU software are copyrighted works,
but their actual ownership is a community of people formed to further the
idea of open software not a company or corporation that exists to squeeze
money from customers and trade intellectual property like baseball cards.
20 years ago, virtually every company that manufactured and sold computers
had their own proprietary operating system. And each of those proprietary
operating systems required a fairly massive programming staff to design,
develop, release, support and extend that company’s O/S. Where are those
people (at least those who haven’t retired, quit the business or died)
working today? In most cases, they’re now working with (but not _on_)
one of three operating systems: Windows, MacOS or Linux.
So today, while your average 50-year-old systems programmer can probably
amuse you at lunch with war stories from back when they worked at
YoYoDyne, they can’t fix your problem when your local file system is
randomly corrupted but only while playing Nexuiz over the net using GFS
data files cross-mounted from a Mac server in Tibet. Of course, back
in the day at YoYoDyne, you might have been fired for suggesting that
such a thing was even possible, let alone worth doing because it might
turn a profit.
In those same 20 years we’ve also seen shake-outs in hardware. BUNCH is
history. So are the 370 and the Alpha. These days, Intel and AMD are
pretty much “it” for CPU-centric hardware… and even they admit they
make commodity parts.
So are we at all surprised that we’ve also arrived at the commodity
operating system?
Not that there’s anything wrong with that.
So are we at all surprised that we’ve also arrived at the commodity
operating system?
What we should be surprised at is that (a) it’s taken so long; and (b) a commodity OS is considered “non-obvious” by some people.
(Hey, maybe that means we can patent it )
Of course, it’s not surprising if you factor in corporate greed.
In those same 20 years we’ve also seen shake-outs in hardware. BUNCH is history.
Two members of the BUNCH (Burroughs and UNIVAC) joined to form Unisys a while back, if you’ll recall, and both of those companies’ hardware and software architectures survive in each case (Unisys A-series and Clearpath NX and Libra servers still run Burroughs MCP with CANDE, WFL, COMS, and friends, while Unisys Clearpath IX and Dorado servers run OS1100 (renamed OS2200 except on the sign-on screens as you can see):
*UNISYS 1100 Operating System Level CP11-03D-DEV(RSI)*
************************************************************
* *
* DDDDDD EEEEEE VV VV SITA *
* D D EE VV VV “M” Series *
* D D EEEE VV VV Development System *
* D D EE VVV Running on CS7800-C *
* DDDDDD EEEEEE V Exec Level 47R5D (CP11.0) *
* *
************************************************************
Current session number: 700
Previous session was: THU 07 DEC 2006 16:52:29 GMT
DATE: 2006-12-07 TIME: 20:46:51 GMT
HVTIP, MAPPER, CTS, FURPUR, and friends all live there quite happily. No IBM architecture and very little IBM influence in sight (thank God).)
While each is a niche player when compared to IBM, each still has significant holdings at some companies and in some industries (the number of airlines which depend to some extent on the proper operation of Clearpath IX, Dorado, and 2200-series hardware is still fairly large percentagewise).
Edited 2006-12-07 20:53
The original author is not a native speaker of English, therefore criticising grammatical mistakes on his part is unnecessary and cheap.
Apart from that and a few minor other points, I agree completely.
I find it hilarious that there exists now several totally different definitions of Linux:
1. Linux the kernel
2. Linux the operating system (which consists of more than the kernel; call it distribution if you like)
3. Linux the community (of users and developers?)
4. Linux the philosophy of software development and marketing
Now, whenever people use the word Linux, I have to listen very hard to discern what Linux they are referring to exactly. Often there seems to be no distinction at all, which makes most of their talk ambiguous and pointless.
(No, GNU/Linux is not a perfect solution either.)
I’m sure your making 3 and 4 up.
I think its more confusing as to what an OS is especially as I would never describe one as having a chat, web browser etc etc.
The reality is the OS is pretty much GNU/Linux by my outdated understanding of what an OS is.
Linux here most of the time refers to a mythical meta-distribution of things that will run on the linux kernel
(and yes I just wanted to make up the term mythical meta-distribution)
Well, if you read the article, the author himself used definitions 3 and 4.
When you call the GNU/Linux distros by that name then there is no ambiguity. What else could you be referring to by using that name?
And what do you mean by “Linux user community”? I don’t know about you but I don’t do very much work interacting with the kernel directly. I only use it indirectly through the use of software which may or may not running on top of other software that runs on top of the kernel.
> Linux the philosophy of software development and marketing
There’s a Linux software development philosophy? There’s a Linux marketing philosophy? Could you please explain these philosophies? I have no idea what they are.
Short answer: Read the article. The author used Linux as the name of a community of users and developers. I’m just playing along.
Long answer: Why I added “Linux user community” and “Linux software development philosophy” there is not a flamebait. Maybe people aren’t conscious of the fact they sometimes use Linux as the name of “all things Linux”, but it sure looks to me like that.
Besides that, GNU/Linux is a correct term if we want to say “a system consisting of GNU utilities (or userland) and the Linux kernel”. However, take any distribution, and you’ll notice that it has components from a variety of sources. Should we say X11/MIT/Perl/GNU/…/Linux? Or */Linux? If we put GNU in the name, why aren’t we putting X11, too? It’s at least as important nowadays. And once we’re on that path, why don’t we put all the other names of the software packages?
Don’t take me wrong. GNU is and has been essential to what we call Linux (or GNU/Linux, if you like) today. The system has, however, long since passed the point where it would consist of solely GNU components and the Linux kernel.
The Subject says it all.
Toe
http://en.wikipedia.org/wiki/TCP_Offload_Engine
And why not Toe
http://linux-net.osdl.org/index.php/TOE
@silicon to be fair other that bad linux vs windows evaluation, and some dubious kernel discussion, and some even worse Linux vs BSD I’ve never heard of Toe.
@CrazyDude0 of all the things you said, you have some valid points. There seem pros and cons both sides, you don’t have to come across as arrogant, and stick to that one point, you made some good points about a potion of the the kernel and then talk about 1000’s of distro’s. Whats that about?
Edited 2006-12-06 11:46
Cyclops: Read the post by Ookaze above and see his tone and his comment score of 4 you will know the answer.
I for one am glad that i am not a part of Linux community anymore.
Edited 2006-12-06 11:52
@CrazyDude0. My view of OSAlert. Is computing is fun. I’ve done it for a long time, but I want to get better. Its a bit of fun and nothing to get worked up about. If you do you normally come off as a pratt.
You made good points…badly. The reality is Ookaze made some better points, and they weren’t his points they were well thought out arguments done by kernel developers who on mass, are pretty good, and to reject something that sounds as good as toe they need good reasons, and they have them.
I modded you down over the fanboys comment. Its a stupid thing to say, and not true. Its a bad insult, people here like often have preferences for different OS’s its true. The reality is that those who have the same bias that you have, will not back you with points if you come across as a pratt. I know from experience.
I learnt something today about toe, and I learnt it because of you, and I’m grateful for that. I will look out for your name in future to see how your hone your discussion skills.
cyclops: It is not really about better points. It is about being in favor of Linux vs Windows.
If you mod up post which uses language like:
Wow, you should shut up right now. Really, you’re making a fool of yourself here !
Then i treat the person who is modding up as dishonest.
If you mod me down just by referring to 100 distributions in Linux which is a fact, and mode up the pro linux but offensive comment, do you think it is really honest?
It is kind of fun for me to see how people behave. I was once a part of so called freedom fighters but i am not anymore because i chose the platform which served my needs better i.e. Windows and i am really happy using it. Be it general use, playing games or developing software with excellent MSDN documentation.
I was just pointing you to the fact that the linux fanboy community is quite intolerant and the modding above proves that.
Edit: btw Ookaze’s point on TOE was completely wrong. TOE is not vendor specific hack in Windows and TOE specifications are well documented in MSDN. Any NIC vendor can follow them to provide TOE support on their TOE enabled NIC.
Edited 2006-12-06 12:33
@CrazyDude0 Your missing the point, bias is part and parcel of this site simply because we spent so long working in these environments. Most people here are pretty reasonable and accept there a good reasons for liking Linux or Windows.
You simply came off badly, and his points were better.I wasn’t happy with silicon for trying to put a stop to what was interesting.
The only thing I have modded you down for is the fanboy thing. What surprised me about your comment is it was well thought out, and badly executed, you could have made the same comment and walked away with points.
Your clearly not seeing how people behave, look at some of the comments in the “Vista is pretty so better thread” you can see that Microsoft users get good points, on what is the most superficial comments, but there civil ;well thought out; and stick to one point.
cyclops: nah buddy, i don’t really care for the points that much to be political about it. People don’t like my views well they can mod me down but I don’t mod anyone cause i want everyone to see this discussion.
If people get pissed off, so can they…it is their problem that they treat computing as a religion.
@Crazydude It is political, but Linux users are in the *minority*, a very *vocal* minority, but they only have so many Votes. The problem with your post was *not* political, or even *technical*…you have weak discussion skills, and its a shame you make some good points.
You know what crazydude0, I’m doing all I can to stay civil when hearing so much nonsense coming from you.
You obviously don’t know what you’re talking about.
Ookaze’s point on TOE was completely wrong. TOE is not vendor specific hack in Windows and TOE specifications are well documented in MSDN. Any NIC vendor can follow them to provide TOE support on their TOE enabled NIC
What I said is completely true. Perhaps you’ll finally realize that the more you write, the more you look like a fool.
Just the few things you’ve written there are amazing. And FYI, I try to stay general, less we go in long off-topic technical discussions.
TOE is a Windows only marketing speak, which is true, as even you can cite only MS references on this.
You cite TOE in Windows, but these are just API to hook to the cards, so that NIC vendors use a common one.
But this has no advantage against Linux, as, again, vendors can and do provide Linux NIC cards with TOE drivers, because they have the source.
I won’t try to prove you how TOE is dangerous for network stability, or not efficient on most loads, do as you wish.
Just remember that having some TOE support in your kernel is no advantage at all.
And please, when I was talking standards, I was talking of at least a RFC, not MSDN. Or was this a joke of yours ?
Now, if MSDN is your reference, it’s pretty sad. I know the MCSE and what other MS certifications are basically brainwash material (yes, people even showed me these, IMHO they’re amazing piece of nonsense), but I urge you to try reading some true networking books at least.
Despite what the fanboys on this website may want you to believe, read Linux and IPv6 story here:
http://tldp.org/HOWTO/Linux+IPv6-HOWTO/basic-history-ipv6-linux.htm…
If i was a system administrator, this would make me very skeptical of IPv6 in Linux.
And who was it that claimed Linux has IPv6 since 2.2…liarrrrrrrr
Edit: So it was again Ookaze. Ok he is clueless for sure.
Edited 2006-12-06 12:46
Despite what the fanboys on this website may want you to believe, read Linux and IPv6 story here:
http://tldp.org/HOWTO/Linux+IPv6-HOWTO/basic-history-ipv6-linux.htm…..
If i was a system administrator, this would make me very skeptical of IPv6 in Linux.
And who was it that claimed Linux has IPv6 since 2.2…liarrrrrrrr
Edit: So it was again Ookaze. Ok he is clueless for sure
You really are an amazing guy, you have no shame at all.
Did you even understand what you read ?
I specifically said that Linux had IPv6 support since 2.2, and I think I said that it was also in the 2.1 dev version.
So where is the lie ? I think that’s a misunderstanding on your part, because you don’t even understand what you read there.
So let me explain it to you.
After the implementation of IPv6 in Linux 2.2 (10 years ago it seems, so you were really 120 % wrong), the IPv6 spec continued to be updated, and, guess what, nearly nobody used IPv6. But it was used during kernel 2.4 time, to connect to an experimental high speed edu network, encapsulated in IPv4 of course.
The IPv6 of Linux was completely functional and worked at least with this one and only IPv6 internetwork.
Now, the updates on IPv6 weren’t implemented, because, you see, IPv6 worked with the project I just talked about, and nothing else could use it.
USAGI helped put the already working Linux IPv6 stack up to date. But this stack was already there and working since even before 2.2.
The new improved and well maintained IPv6 stack was integrated in Linux 2.6 (that’s also many years ago, mind you, 2003 IIRC), as IPv6 is more used today.
So yes, IPv6 was in Linux 2.2, and you are indeed really clueless beyond all recognition.
And if I want to concede to you, in a laaaarge kind move, that you were talking about the latest extensions of IPv6, you would still be completely wrong, as you said Linux doesn’t even have an IPv6 stack. And you’re wrong 120 % anyway, because Linux had even that years ago, and it’s well tested everyday, which is not the case with the Windows Vista stack.
At the chance of wandering off topic, TOE does have its uses. Especially in an iSCSI world. TOE is also not dependant on the OS, but the drivers for the TOE enabled NIC in your box. Yes, for the OS, TOE is just a bit of hardware to support.
For the rest of the discussions in here…everything else revolves, again, around the presumption that Windows and Linux based OSes serve the same purpose, are interchangeable and should work the same for all tasks. Sorry, that’s simply not the case. Underprivileged users in Windows are, and always have been, an afterthought. All the solutions which are built into Windows are a “solution” to still allow Administrator access when needed, preferably as user friendly as possible. Daily use shows that for a *lot* of things, UAC is needed to get the job done. Compare that to the design of an OS which is built from the ground up to be multiuser. Everything is based oin the position that everything you need to do in your day to day work is doable under a standard unprivileged user account, and that only by high exception you need to execute something as different user. Its tools and their use reflects that kind of use. They’re simply not comparable because the underlying OS design is totally different.
BTW, ipv6 has been around on unix like OSes (including Linux) even before it became fashionable to mention it. Same with IPsec, etc. Sorry, especially in the field of networking, Microsoft is the follower, not the leader. Nor do they need to be, but that’s a different subject.
Fred: One correction – Windows NT was designed from scratch to be mutli-user. It was due to the legacy of DOS and Windows 3.x and 9x that Microsoft made the poor choice of promoting people to use Admin account all the time.
On my laptop in XP, i always use limited user just like Linux and i do runas to get an admin command shell and do my admin work.
So it is wrong to say that Windows as it is today wasn’t designed for multi-user, it is just that Microsoft gave preference to user friendliness over security. So yes if you blame Microsoft for bad decision i would agree, but i wouldn’t agree if you say Windows is inferior in technical terms.
…but i wouldn’t agree if you say Windows is inferior in technical terms.
I didn’t. Just different.
I still think multiuser in anything Windows has been an afterthought…or badly designed, depending on how you look at it
But alas, that’s not really the point in this thread, not the point I’m trying to make. Any comparison between Windows and non-Windows OSes ultimately fail because there’s really so little common ground to compare on.
I agree that windows NT is built multi-user, and I respect any effort to run it limited user.
I, for one, admin a few XP boxes. Needless to say, I have them adequately locked down and thus suffer from little malware. However, I simply cannot achieve the same amount of ease running in limited user as admin in windows-land as in linux.
Moreover, with the implementation so focused on legacy compatibility, the multi-user OS is designed mono-user and becomes a chore to run in enterprise settings. Hence the view that the multi-user OS is “multi-user”.
Furthermore, CrazyDude0, the rest of the people who replied to you had great points. And they have evidences to support why TOE is not exactly sought-after, why windows isn’t exactly the networking king (and IPv6 too) and so forth. If you are going to dismiss their viewpoint as being politically inclined, it will show more about you than about them.
Finally, this is really off-course. No. This is total derailment. Please do not continue with it.
…better than the “other” article.
But neither of them were very meaty. Linux’s fortunes may rise and fall, but it is basically here to stay. The user and developer community/fan base is just too great to allow it to simply die off. I mean, I’d love to see Haiku steal all linux users but it will never happen because linux is solidly established and rightfully so because of how the options it presents to different types of users.
I think it’s shortsighted and wrong to not recognise the advancements Vista *did* bring. Though sometimes overstated in importance, they are interesting enough. In the dicklenth competition of who does what better/faster/bigger (like how heavy the (mis)use of the GPU has penetrated into each OS) people tend to lose track of what is really important: The question wether the feature in question is actually a useful addition to the toolbox at hand. If it is, I see nothing wrong with adopting a similar feature in another OS. It makes sense to do so. The discussion of who copied what from whom, is, IMO, secondary. OTOH, IP lawyers probably disagree.
Would be nice if a new OS wouldn't make a fool out of existing hardware.I mean you hardly buy every 3 years or so a new TV why should a PC be different?
Linux gives you more for the buck.You can choose to work longer with existing hardware.
“Let’s say you have an error. How would you handle it?”
you write a log message on /var/log/messages so there’s a record of WHEN happened WHAT… then, dump the error message on the console for the user with an return code for any automatic agents, and, in the case of a fatal error, write a core dump on the filesystem.
Traceability, baby !
Dialog Boxes ?!?! Who need them ?…
twenex – if you spend some time honestly appreciating Linux than badmouthing windows, may be you will do community a favor.
I was checking the comments and it seems twenex is in every thread about windows and is badmouthing it. Whats up with that? Why are you so infatuated?
Usually people who badmouth others are the ones who don’t have confidence in their own stuff so they want to bring the other down by badmouthing. Linux camp is badmouthing Windows for years but instead of Windows falling, it made windows more stable and now making it more secure. Ahh too bad for them.
This is how Microsoft takes criticism and act on that. What does Linux community do, they say RTFM, RTF configuration files etc and the classic do it yourself as if every user can compile kernel LMFAO:)
Edited 2006-12-06 23:39
twenex – if you spend some time honestly appreciating Linux than badmouthing windows, may be you will do community a favor.
Oh, I appreciate Linux every day, believe me.
I was checking the comments and it seems twenex is in every thread about windows and is badmouthing it. Whats up with that?
Because it’s crap?
EDIT: That’s not the whole story though. I can’t think of a single other industry where a single range of products holds such a dominant place in the industry, let alone one that fails so abysmally in comparison to the competition.
Actually, if you actually read my comments you’ll see I don’t so much “badmouth Windows” as “shoot down dangerous anti-Linux FUD”.
Why are you so infatuated?
“Infatuation” implies positive, if needlessly exaggerated, feelings.
Usually people who badmouth others are the ones who don’t have confidence in their own stuff so they want to bring the other down by badmouthing.
Erm, no. People don’t badmouth dictators and criminals because they secretly feel that torture, theft and rape are good ideas.
Linux camp is badmouthing Windows for years but instead of Windows falling, it made windows more stable and now making it more secure. Ahh too bad for them.
Yeah, Windows now is a BIT more stable, a BIT more secure and a LOT more bloated. And maintains its illegally-gained 90% desktop market monopoly.
Film at 11.
Meanwhile, Linux, the “free, hobby OS that won’t be big and professional like GNU and is a cancer programmed by emotionally-challenged communist h4ck0rz in their parents’ basements” now has over 25% of the server market, and growing (at a faster rater than Windows).
Edited 2006-12-07 00:59
now has over 25% of the server market, and growing (at a faster rater than Windows).
11.8% in Q3. It was 12% in Q2.
Windows has 37%. Unix has about 30%. The rest is held by mainframes.
Linux 3 years ago was growing at 60% year over year.
Linux 2 years ago was growing at 40% year over year.
Linux 1 year ago was growing at over 20% year over year.
In Q2 of this year it was growing at 6.1%.
In Q3 it was growing at 5.4%.
Rapidly approaching zero.
11.8% in Q3. It was 12% in Q2.
I believe that’s revenue share, not market share. Market share is indeed around 25%. Don’t confuse the two (either unintentionally or on purpose).
Rapidly approaching zero.
No. Rapidly leveling off at 5%, more than Windows.
You’ve made your unscientific prediction many times. How about giving it a rest until after Q4 to see who was right, hmm? In the meantime, I’ll keep providing my own estimate everytime you mention yours.
I wouldn’t waste your breath. Er typing. I doubt you could change NotParker’s mind if you gave him a brain transplant.
Or do I mean “implant”?
I believe that’s revenue share, not market share. Market share is indeed around 25%.
Linux has a .4% installed base as I’ve proven.
Servers only represent a small portion of overall computers in use.
In the meantime, I’ll keep providing my own estimate everytime you mention yours.
The ritual stalking of the heretic!
Edited 2006-12-07 15:58
Linux has a .4% installed base as I’ve proven.
You haven’t proven anything, as I’ve demonstrated many times. Moreover, you’re not even talking about the correct topic! This was about *servers*, NotParker! Play pay attention.
Gartner and IDC, which you yourself claimed were respectable sources, establish Linux market share at around 20-25% for servers and 2-2.5% for desktops. Of course, you only accept their estimates when they serve your agenda, and then claim they are biased towards Linux when they don’t…
The ritual stalking of the heretic!
No, simply countering your disinformation, that’s all.
Gartner and IDC, which you yourself claimed were respectable sources, establish Linux market share at around 20-25% for servers and 2-2.5% for desktops.
Since both Gartner IDC put server sales for Linux at less than 12%, and smaller than that in previous years, that is mathematically impossible.
Post your source Mr. Ritual “Math Challenged” Stalker.
As for desktop, Gartner also points out that 40-80% of Linux installs are replaced by a pirated copy of Windows.
But go ahead, post some numbers Mr. Stalker.
You are so funny. You cult members are sooo threatened by reality.
Since both Gartner IDC put server sales for Linux at less than 12%, and smaller than that in previous years, that is mathematically impossible.
Gartner and IDC talk about revenue, not installed base. Since many Linux servers installs are not bought copies, but downloaded (and installed on legacy hardware), that is not only possible but very logical.
Post your source Mr. Ritual “Math Challenged” Stalker.
“In servers, IDC predicts Linux’ market share based on unit sales will rise from 24% today to 33% in 2007, compared with 59% for Windows. In a survey of business users by Forrester Research, 52% said they are now replacing Windows servers with Linux. On the desktop side, IDC sees Linux’ share more than doubling, from 3% today to 6% in 2007, while Windows loses a bit of ground.”
http://blogs.zdnet.com/ITFacts/?p=6932
As for desktop, Gartner also points out that 40-80% of Linux installs are replaced by a pirated copy of Windows.
That was only for desktop computers sold pre-installed with Linux in Asia, and you know it. On the other hand, it means that in 20 to 60% of these cases, Linux is kept on the PC…that’s not a bad percentage, actually.
But go ahead, post some numbers Mr. Stalker.
I did, but you’ll conveniently ignore it, as you do *every time* you are proven wrong.
Oh, and “stalker”? Is that your new insult for me? Poor NotParker, still incapable of having a normal, rational discussion, always resorting to ad hominem attacks…
You are so funny. You cult members are sooo threatened by reality.
I think you’re the one feeling threatened here. That would explain why you lash out at those that provide counter-arguments to your FUD.
As long as you keep misrepresenting the same statistics to further your agenda, I’ll be there to make sure the truth comes out.
“In servers, IDC predicts Linux’ market share based on unit sales will rise from 24% today to 33% in 2007, compared with 59% for Windows. In a survey of business users by Forrester Research, 52% said they are now replacing Windows servers with Linux. On the desktop side, IDC sees Linux’ share more than doubling, from 3% today to 6% in 2007, while Windows loses a bit of ground.”
That prediction story was from January 2005. Do you have any indication that the prediction came true?
Do you have numbers from 2006?
Are you sure those numbers were for all servers? 33 + 59 = 92%. I really doubt Windows and Linux had the whole market to themselves. What about Unix?
Nice try. But I’ll stick with numbers from Q3 of 2006, not wild predictions from January of 2005.
In Q1 of 2006 Linux had 12.2% of server revenue.
In Q2 of 2006 Linux had 12.0% of server revenue.
In Q3 of 2006 Linux had 11.8% of server revenue.
Anyway, I prefer server revenue as it correctly indicates the value of a 64-way or 8-way server vs a 1 cpu server.
I mean, I don’t really mind claiming Windows has 59% of the market vs the 37% I usually claim based on server revenue. But the Unix and mainframe people may object to that!
Edited 2006-12-07 18:18
That prediction story was from January 2005.
The story notes that *current* market shares are 24% and 3% for servers and desktops, respectively.
Do you have any indication that the prediction came true?
Do you have any indication that it didn’t? In any case, I’ve already proven my point with the fact that in 2005 the *current* server market share was 24%.
Are you sure those numbers were for all servers? 33 + 59 = 92%.
I believe the 59% figure for Windows was the current market share for 2005 as well, so that would be 24% + 59% = 83%, which makes sense.
Nice try. But I’ll stick with numbers from Q3 of 2006, not wild predictions from January of 2005.
As I predicted: when confronted with numbers that go against your agenda, you simply ignore them, even when they come from a source which you’ve already identified as reliable in previous post.
Again, the 24% figure is *not* a prediction. It is the measured market share per unit, which also serves to demonstrate that MS offerings are overpriced.
Your dishonesty has been uncovered yet again. Be a man, for once in your life, and admit that you were wrong!
Do you have any indication that it didn’t? In any case, I’ve already proven my point with the fact that in 2005 the *current* server market share was 24%.
Since the prediction was made in January 2005, it could hardly have accurate numbers for any part of 2005 since those numbers lag by a month or so.
That number could have been for 2004.
Anyway, I’ll stick with revenue as it properly differentiates between 1 cpu servers and 64 cpu server.
It is also much fairer to Unix and Mainframes since they comprise over 50% of the revenue generated by servers.
Since the prediction was made in January 2005, it could hardly have accurate numbers for any part of 2005 since those numbers lag by a month or so.
Now you’re just being silly. All that pussyfooting just to avoid admitting you’ve made a mistake…it’s sad, really.
The number was given by IDC, which you’ve already recognized as a reliable source. Face it: the market share per unit for 2005 was 24%, and it’s unlikely that i has decreased since then.
Anyway, I’ll stick with revenue as it properly differentiates between 1 cpu servers and 64 cpu server.
No, you’ll keep using these figures because the percentages appear smaller and that suits your anti-Linux agenda. You’re not fooling *anyone*, NotParker.
BTW, Linux can scale up to 256-cpu servers, as demonstrated by SGI, IIRC.
It is also much fairer to Unix and Mainframes since they comprise over 50% of the revenue generated by servers.
No, it only means that Unix servers (like Windows) are overpriced, which is one of the main reasons why their market share is dwindling.
As for Mainframes, they can run Linux as well, didn’t you know?
Face it: the market share per unit for 2005 was 24%, and it’s unlikely that i has decreased since then.
Yes it could. IDC shows revenue market share has dropped in 2006 from Q1 to Q3.
BTW, Linux can scale up to 256-cpu servers, as demonstrated by SGI, IIRC.
And thats my point. You wish to use unit sales which counts a 256CPU server (Unix or Linux) as 1 unit and a 1 cpu server as 1 unit.
Revenue gives a better idea of the reality that the 256cpu server workload and the number of clients connected is going to be close to 256x the workload of a 1cpu server.
However, if you insist I will congratulate Windows on having 59% of unit sales in 2004.
No, it only means that Unix servers (like Windows) are overpriced
I don’t understand? Are you saying 256cpu Unix servers should cost the same as 1 cpu Linux servers? Thats weird.
In Q1 of 2006 Linux had 12.2% of server revenue.
In Q2 of 2006 Linux had 12.0% of server revenue.
In Q3 of 2006 Linux had 11.8% of server revenue.
References:
http://www.idc.com/getdoc.jsp?containerId=prUS20180706
“Although Linux servers now represent 12.2% of all server revenue, revenue growth for the quarter was approximately half the growth rate observed in 1Q05”
http://blogs.zdnet.com/ITFacts/?p=11637
“Linux servers now represent 12.0% of all server revenue, up slightly from Q2 2005.”
http://www.itjungle.com/breaking/bn112206-story01.html
“Linux machines accounted for 11.8 percent of overall server sales, and its growth has moderated considerably.”
Edited 2006-12-07 18:56
es it could. IDC shows revenue market share has dropped in 2006 from Q1 to Q3.
Not by a significant number, and growth is still positive (and higher than MS), so there’s no reason to believe this…unless one is so completely blinded by his hatred of Linux that he no longer thinks rationally. Sounds familiar?
And thats my point. You wish to use unit sales which counts a 256CPU server (Unix or Linux) as 1 unit and a 1 cpu server as 1 unit.
That’s completely besides the point. The majority of servers fall into the 2 to 8 CPU range, and there’s no reason to believe that these aren’t equally distributed between Windows, Linux and Unix.
I don’t understand? Are you saying 256cpu Unix servers should cost the same as 1 cpu Linux servers? Thats weird.
That’s not what I’m saying at all. You’re grasping at straws, now.
According to IDC, a source you’ve already quoted as reliable:
Linux server market share: between 24% and 33%
Desktop market share: between 3% and 6%
Suck it.
The majority of servers fall into the 2 to 8 CPU range, and there’s no reason to believe that these aren’t equally distributed between Windows, Linux and Unix.
Believe? No numbers? I post numbers. You post cultish beliefs!
Linux server market share: between 24% and 33%
Predicted for 2005 – 2007. We don’t know what the real numbers are.
Desktop market share: between 3% and 6%
Predicted for 2005 – 2007. We do know the real numbers are .4%.
Suck it.
Pervert.
In Q1 of 2006 Linux had 12.2% of server revenue.
In Q2 of 2006 Linux had 12.0% of server revenue.
In Q3 of 2006 Linux had 11.8% of server revenue.
References:
http://www.idc.com/getdoc.jsp?containerId=prUS20180706
“Although Linux servers now represent 12.2% of all server revenue, revenue growth for the quarter was approximately half the growth rate observed in 1Q05”
http://blogs.zdnet.com/ITFacts/?p=11637
“Linux servers now represent 12.0% of all server revenue, up slightly from Q2 2005.”
http://www.itjungle.com/breaking/bn112206-story01.html
“Linux machines accounted for 11.8 percent of overall server sales, and its growth has moderated considerably.”
Edited 2006-12-07 19:46
Believe? No numbers? I post numbers.
I have yet to see you post numbers about the proportion of 1-cpu to 2-cpu, 4-cpu, etc. up to 256-cpu machines. Without those numbers, there’s no reason to believe that the number of cpu is really relevant for this.
If anything, more CPUs would skew the revenue share towards Unix, not Windows or Linux, since servers with lots of CPUs tend to run Unix.
Predicted for 2005 – 2007. We don’t know what the real numbers are.
The first number is not predicted, it’s the current server market share for 2005 by IDC.
Predicted for 2005 – 2007. We do know the real numbers are .4%.
The first number is not predicted, it’s the current desktop market share for 2005 by IDC – and we now know that the real number is certainly not as low as your unscientific, heavily-biased estimate is.
Pervert.
I take you haven’t watched much WWE…
Keep spamming those number, NotParker, maybe if you do it enough that you’ll get banned.
I have yet to see you post numbers about the proportion of 1-cpu to 2-cpu, 4-cpu, etc. up to 256-cpu machines. Without those numbers, there’s no reason to believe that the number of cpu is really relevant for this.
Or irrelevant. I’m waiting for numbers.
The first number is not predicted, it’s the current server market share for 2005 by IDC.
It can’t be since the article is from January 2005 and sales for Q1 2005 would not be reported until the end of April 2005. Therefore it is either a prediction for 2005 or numbers from 2004.
The first number is not predicted, it’s the current desktop market share for 2005 by IDC
It can’t be since the article is from January 2005 and sales for Q1 2005 would not be reported until the end of April 2005. Therefore it is either a prediction for 2005 or numbers from 2004.
And we know from Google that in late 2004 Linux’s share was 1% or lower. And we know it is now .4%.
Keep spamming those number, NotParker, maybe if you do it enough that you’ll get banned.
Is that your goal? You plan to keep lying about the numbers I post hoping that someone will not see through your perverted call for me to “suck it”?
In Q1 of 2006 Linux had 12.2% of server revenue.
In Q2 of 2006 Linux had 12.0% of server revenue.
In Q3 of 2006 Linux had 11.8% of server revenue.
References:
http://www.idc.com/getdoc.jsp?containerId=prUS20180706
“Although Linux servers now represent 12.2% of all server revenue, revenue growth for the quarter was approximately half the growth rate observed in 1Q05”
http://blogs.zdnet.com/ITFacts/?p=11637
“Linux servers now represent 12.0% of all server revenue, up slightly from Q2 2005.”
http://www.itjungle.com/breaking/bn112206-story01.html
“Linux machines accounted for 11.8 percent of overall server sales, and its growth has moderated considerably.”
I’m waiting for numbers.
No, *I’m* waiting for numbers. The burden of proof is on you to show that the number if cpus per server has any impact on our discussion. It’s *your* stupid theory, not mine.
It can’t be since the article is from January 2005 and sales for Q1 2005 would not be reported until the end of April 2005. Therefore it is either a prediction for 2005 or numbers from 2004.
It doesn’t matter. They’re the *current* market share number of Jan. 2005. Show me numbers that *prove* that market share per unit sold has decreased since then and we’ll talk.
The same goes for desktop numbers, except that in that case you can’t make some desperate attempt to spin it using number of cpus.
Linux server market share per unit sold as of Jan. 2005: 24% (projection for 2006: 33%)
Linux desktop market share per unit sold as of Jan. 2005: 3% (projection for 2006: 6%)
And we know from Google that in late 2004 Linux’s share was 1% or lower. And we know it is now .4%.
I’ve already demonstrated that webstats cannot be used to determined market share. That’s why Google removed it from their Zeitgeist, too.
ICD – whose record you have defended – places it at 3% at the beginning of 2005. It’s probably at 5 or 6% now.
Is that your goal? You plan to keep lying about the numbers I post hoping that someone will not see through your perverted call for me to “suck it”?
No, my goal is to expose *your* lies about numbers. As for “sucking it”, please enlighten yourself:
http://en.wikipedia.org/wiki/D-Generation_X#Suck_it.21
Suck it.
Oh, and please keep spamming – the more you spam, the more childish you look like, and the more you expose yourself as an immature troll.
Suck it.
Pervert.
Linux’s market share by server revenue is shrinking for 3 straight quarters.
Its growth rate is approching zero.
Bye bye Linux.
Linux’s server growth is stabilizing at a higher level than Windows’ growth rate.
Suck it.
NotParker (0.00)
Hey, congratulations: your Trust score is now equal to your credibility.
Heh.
As I predicted: when confronted with numbers that go against your agenda, you simply ignore them, even when they come from a source which you’ve already identified as reliable in previous post.
Oh, bravo! Nice catch
In Q1 of 2006 Linux had 12.2% of server revenue.
In Q2 of 2006 Linux had 12.0% of server revenue.
In Q3 of 2006 Linux had 11.8% of server revenue.
References:
http://www.idc.com/getdoc.jsp?containerId=prUS20180706
“Although Linux servers now represent 12.2% of all server revenue, revenue growth for the quarter was approximately half the growth rate observed in 1Q05”
http://blogs.zdnet.com/ITFacts/?p=11637
“Linux servers now represent 12.0% of all server revenue, up slightly from Q2 2005.”
http://www.itjungle.com/breaking/bn112206-story01.html
“Linux machines accounted for 11.8 percent of overall server sales, and its growth has moderated considerably.”
You’re sounding like a broken record, NotParker. You asked me to provide numbers to indicate Linux’ server market share, and I did. Now you’re just frantically reposting the same data which I indicated was share of *revenue*, not *units*.
You’re not convincing anyone of anything, except of your own incredibly obvious pro-MS bias.
Windows is a fine product. It doesn’t need you lying about Linux to be successful. Just give it up already, you’ve already given enough of a bad name to pro-Windows advocates.
Meanwhile, Linux … now has over 25% of the server market …
I read that number here a few times. Where does it come from?
I know a few large enterprises with hundreds of servers. The only place you see Linux is in special-purpose appliance boxes, like specialized firewall machines (IDS and such). I.e., where the vendor supplies and maintains the box.
Moreover, it is not uncommon to meet a windows sysadmin who has inherited a unix box to maintain, and who complains about unix being a PITA to work with.
Now, I’m talking about command prompt unix, not Linux with GNOME/KDE, but the point is, many non-enthusiastic IT personal simply have an aversion to anything unix-ish.
Personally I feel the 25% share number is a _gross_ over estimation.
Moreover, it is not uncommon to meet a windows sysadmin who has inherited a unix box to maintain, and who complains about unix being a PITA to work with.
Well, could that be because he’s a *Windows* sysadmins?
Many studies have shown that Linux servers require a lot less maintenance than Windows one. Speaking from experience I can tell that this is true.
many non-enthusiastic IT personal simply have an aversion to anything unix-ish.
That’s because they don’t know it as well. If they did, they would most likely *prefer* it, as it is a lot easier to maintain once you’re familiar with it.
Personally I feel the 25% share number is a _gross_ over estimation.
The overwhelming majority of Web hosting companies use Linux. Most of the Internet runs on Linux.
IDC set server market share at 28% in Jan. 2005. Back then they estimated that it would be at 33% now.
This is how Microsoft takes criticism and act on that. What does Linux community do, they say RTFM, RTF configuration files etc and the classic do it yourself as if every user can compile kernel LMFAO:)
Actually, Microsoft’s modus operandi is to make the minimum amount of improvements they can get away with, whilst spreading FUD about competitors’ products and increasing the size of their “operating system” by multiples of 100%. Linux, on the other hand, has gone (in the time between Windows ME and Vista) from needing to be edited via configuration files – a perfectly reasonable requirement, and one which makes a lot more sense than the bloated Registry – to being almost totally configurable via GUIs.
With vastly improved hardware support.
And improved user interfaces.
Oh, and the last bit of news is that whilst going through the options can be tedious, compiling the kernel is not hard. The computer does it for you.
Linux vs. other systems as a server:
There are many good server OS’s today:
– Solaris and other SYSV UNIXes
– *BSD’s
– Win2003 R2
Win2003 R2, according to the data sheet, is non-typical Microsoft product. It has low hardware requirements
(133Mhz, 128MB), built in UNIX command line interface with scripting, and ability to compile some UNIX software.
Linux without support can be btained free, but with support option, it is not less expensive than the others.
Conclusion: Linux server share is not going to change much in the near future.
Linux vs. other systems as a desktop:
Linux GUI, KDE in the first place is good enough. The problems are: multimedia, gaming and productivity
Multimedia is a problem. Most distributions can’t play
encrypted DVD’s and MP3’s out of the box. Fixing that
requires expertise beyond reach for the typical consumers. DRM for Linux uncertain, and multimedia will depend on DRM more in the future. There are problems with patents on one side and rigid FSF stance on the other side. The things will get worse in the future.
Gaming is almost non-existing. High performance 3D drivers are not the part of Linux, and game vendors
will not bother to make Linux ports, Average consumer is not able to get drivers working. Again, FSF rigid stance is a problem. The things will get worse in the future.
Productivity is based, mostly, on OpenOffice/StarOffice. OpenOffice is good enough in many situations, but MS Office is more advanced. MS Office is very expensive, too.
All desktop factors together put Linux on desktop right where it is 1-3% share.
Overall conclusion. Linux share on both, server and desktop will not change much in the near future.
DG
Fixing that
requires expertise beyond reach for the typical consumers.
So installing software is “beyond the reach of typical consumers”?
DRM for Linux uncertain, and multimedia will depend on DRM more in the future.
Linux does not want DRM – like the majority of people with more sense than money.
So installing software is “beyond the reach of typical consumers”?
In this case, yes. Consumers usually haven’t heard of dvdcss or something that is needed for playing DVD movies. They don’t know what to look for, they can’t find it. Installing some particular pieces of software is beyond the reach of typical consumers. Correct.
Linux does not want DRM
Couldn’t put it better myself. That’s why Linux is 1% -3% of desktops. Linux does not want consumers. Linux wants hackers. DRM is very likely to be requirement for
multimedia in near future. Do you expect people to
listen to Stallmans speeches in OGG format, instead of
music ? Get real. No DRM = no multimedia, except pirated.
Do you expect people to
listen to Stallmans speeches in OGG format
Good to see some nice sense of humour.
BTW did you guy ever read stallman’s resume, the fatso is looking for a gf LOL
“They don’t know what to look for, they can’t find it. Installing some particular pieces of software is beyond the reach of typical consumers. Correct.”
I haven’t looked in a while but for entry-level distributions you normally find them, well placed. I’ve seen Ubuntu screenshots of a simple click dialog box, for codecs, mp3, and dvd all in one place. I’m actually quite surprised this hasn’t been resolved elsewhere.
Its easier that installing a codec pack
“Couldn’t put it better myself. That’s why Linux is 1% -3% of desktops. Linux does not want consumers. Linux wants hackers. DRM is very likely to be requirement for
multimedia in near future. Do you expect people to
listen to Stallmans speeches in OGG format, instead of
music ? Get real. No DRM = no multimedia, except pirated.”
I’m actually quite shocked at this. Clearly you are not familiar with the web today…people want their DVD players to come with DivX.
DRM is not a Linux issue its not even an OS issue. Its a people like to feel they own stuff. Its a battle not yet fought.
It used to be the likes of Stallman who talked about DRM…its becoming everyone.
In this case, yes. Consumers usually haven’t heard of dvdcss or something that is needed for playing DVD movies.
So put an icon on the desktop saying “Get DVD movie software here”?
Linux does not want DRM
Couldn’t put it better myself. That’s why Linux is 1% -3% of desktops. Linux does not want consumers. Linux wants hackers.
Well, if only “hackers” are not stupid enough to put up with DRM, then thank God I’m a “hacker”.
So put an icon on the desktop saying “Get DVD movie software here”?
That is not a bad idea. Most people do not know what they actually need. When I tried to play encrypted DVD on Linux for the first time, couple of years ago, there was just a message about copyrights, and my first impression was that I am not going to be able play DVD’s at all.
It was sheer luck that I stumbld on dvdcss.
Put yourself in the position of a person who finds cheap PC with Linux preinstalled and buys it. That is consumer.
DG
“DRM is very likely to be requirement for multimedia in near future.”
Only if it wont make it harder (or impossible) for consumers to listen to and view their multimedia the way they want to.
DRM is very likely to be requirement for
multimedia in near future. Do you expect people to
listen to Stallmans speeches in OGG format, instead of
music ? Get real. No DRM = no multimedia, except pirated.
Actually, it is YOU who should get real. Since there are already laws in place to prosecute illegal copying of DVDs and music, etc., no-one needs DRM except control freaks. When I buy music or a DVD, as long as I do nothing illegal with it then whether I play it on my HDDVD player, my normal DVD player, my computer or my washing machine is no-one’s business but mine. If putting DRM into products forces people who don’t have DRM-capable players to pirate the DRM’d content – thereby increasing, not reducing piracy, which is the technology’s supposed intent – then the people who are pushing DRM will have no-one to blame but themselves. No-one.
Actually, it is YOU who should get real. Since there are already laws in place to prosecute illegal copying of DVDs and music, etc., no-one needs DRM except control freaks.
LMFAO! The law is the last refuge of a scoundrel. How realistic do you consider enforcement of these laws to be? Copyright violation occurs in peoples’ homes, which is slightly different than a kid holding up a gas station or a liquor store. The FBI isn’t going to break down your door when you rip a DVD, CDs, etc. Consequently, intellectual property and content owners have taken the pragmatic initiative of trying to thwart illegal copying through DRM. DRM isn’t intended to be foolproof. It’s designed to be inconvenient enough for the average person that they won’t consider it worth their time to rip a DVD, copy a game, etc. The original poster got it right. Content owners who are interested in evolving media formats are going to use DRM *more*, not less, because they see it as a means of stemming losses due to piracy.
LMFAO! The law is the last refuge of a scoundrel.
Funny, I thought it was scoundrels who broke laws. And now I hear it isn’t? Right, I’m off to raid the nearest liquor score. (After all, I don’t want to be known as a law-abiding scoundrel).
How realistic do you consider enforcement of these laws to be? Copyright violation occurs in peoples’ homes, which is slightly different than a kid holding up a gas station or a liquor store. The FBI isn’t going to break down your door when you rip a DVD, CDs, etc. Consequently, intellectual property and content owners have taken the pragmatic initiative of trying to thwart illegal copying through DRM. DRM isn’t intended to be foolproof. It’s designed to be inconvenient enough for the average person that they won’t consider it worth their time to rip a DVD, copy a game, etc. The original poster got it right. Content owners who are interested in evolving media formats are going to use DRM *more*, not less, because they see it as a means of stemming losses due to piracy.
Blah, blah.
It is a well-established principle of common law that it is better for a guilty man to go free than for an innocent man to be wrongfully imprisoned. It is for that reason that, whilst with a secret police force it would be possible to crack down on a vast amount of crimes which it is not possible to crack down on without one, democracies don’t have them. Furthermore, private police forces are illegal. Therefore, there are two very good reasons why film distributors and so on should not be allowed to establish their own Stasi, even if that Stasi is technological, rather than human, in nature.
Lastly, if as you say DRM is intended “only to be a deterrent to piracy” then it will not, in fact, stem piracy, but it WILL cause inconvenience to customers.
Between inconveniencing and suing your own customers, the way the technology market is being led by Microsoft and others, it’s all so very “stupid! stupid! stupid!” it’s like a plotline from Plan 9 from Outer Space.
Edited 2006-12-07 18:18
Funny, I thought it was scoundrels who broke laws. And now I hear it isn’t? Right, I’m off to raid the nearest liquor score. (After all, I don’t want to be known as a law-abiding scoundrel).
Who says you haven’t broken the law? I’m reasonably certain that you’ve broken at least a couple laws today alone.
It is a well-established principle of common law that it is better for a guilty man to go free than for an innocent man to be wrongfully imprisoned. It is for that reason that, whilst with a secret police force it would be possible to crack down on a vast amount of crimes which it is not possible to crack down on without one, democracies don’t have them. Furthermore, private police forces are illegal. Therefore, there are two very good reasons why film distributors and so on should not be allowed to establish their own Stasi,
Agree. Nice straw man. I never advocated the use of secret police to break down peoples’ doors. I merely used that analogy to point out that enforcement of copyright infringement law (something that you were advocating as an alternative to DRM) simply isn’t working.
even if that Stasi is technological, rather than human, in nature.
Totally disagree. If I design a product, then *I* get to decide what it does and how it does it (provided my product doesn’t break any laws), not you. If you don’t like my product, then get over it, don’t buy it, or modify the product to do what you want (with the caveat that you don’t break the law in the process — copyright violation is a clear violation of the law).
Lastly, if as you say DRM is intended “only to be a deterrent to piracy” then it will not, in fact, stem piracy, but it WILL cause inconvenience to customers.
BS, for several reasons. First, from an anecdotal standpoint, virtually nobody that I know has any idea how to rip a DVD. Second, sales of DVDs are BOOMING, which has increased competition for those dollars and, thus, caused prices to decline sharply. Whereas DVDs used to cost $25 (US), many can now be found between $5 to $10. It simply isn’t worth peoples’ time to pirate a $5 DVD.
Between inconveniencing and suing your own customers, the way the technology market is being led by Microsoft and others, it’s all so very “stupid! stupid! stupid!” it’s like a plotline from Plan 9 from Outer Space.
Between inconveniencing and suing your own customers, the way the technology market is being led by Microsoft and others, it’s all so very “stupid! stupid! stupid!” it’s like a plotline from Plan 9 from Outer Space.
Nonsense. For example, customers aren’t inconvenienced by DVD encryption at all — and they’ve gone out and spent tens of BILLIONS of dollars on DVD players and DVD movies. The only ones who are inconvenienced are those who are trying to illegaly copy the media (ie. freaks and geeks).
I’m reasonably certain that you’ve broken at least a couple laws today alone.
That’s a completely gratuitious accusation, BTW.
First, from an anecdotal standpoint, virtually nobody that I know has any idea how to rip a DVD.
There are tons of programs that do this on Windows. Note: making a backup copy of a DVD is perfectly legal. In fact, the act of copying copyrighted media is *not* illegal – it’s not even illegal to download a copy. The only illegal act is unauthorized *redistribution* of copyrighted material.
Whereas DVDs used to cost $25 (US), many can now be found between $5 to $10. It simply isn’t worth peoples’ time to pirate a $5 DVD.
I don’t know where you live, but here new releases are always 20$+. 5$ DVDs are bargain bin items.
Nonsense. For example, customers aren’t inconvenienced by DVD encryption at all — and they’ve gone out and spent tens of BILLIONS of dollars on DVD players and DVD movies. The only ones who are inconvenienced are those who are trying to illegaly copy the media (ie. freaks and geeks).
False. I am a customer who buys DVDs. I have a collection of about 150, counting TV series and documentaries. I have never copied a DVD in my life. Yet I am penalized for making a legitimate consumer choice, i.e. running Linux instead of Windows or Mac OS X. Granted, the inconvenience is minor, and fixed in less than a minute, but I’m still unfairly singled out.
Meanwhile, the *vast majority* of illegal copying is done by people using Windows…
I’m reasonably certain that you’ve broken at least a couple laws today alone.
That’s a completely gratuitious accusation, BTW.
So it is, (and thankyou for bringing that to my attention) unless of course calling Microsoft and tomcat on their bullshit have suddenly become illegal. In which case it’s another case for the ACLU.
Totally disagree. If I design a product, then *I* get to decide what it does and how it does it (provided my product doesn’t break any laws), not you. If you don’t like my product, then get over it, don’t buy it, or modify the product to do what you want (with the caveat that you don’t break the law in the process — copyright violation is a clear violation of the law).
I’d be perfectly happy not buying Windows or any other DRM’ed crap. The problem is that you often can’t avoid it. And if I buy a product, then *I* get to decide how I use it.
BS, for several reasons. First, from an anecdotal standpoint, virtually nobody that I know has any idea how to rip a DVD.
Speaks volumes about the company you keep. Which in turn speaks volumes about you.
Second, sales of DVDs are BOOMING, which has increased competition for those dollars and, thus, caused prices to decline sharply. Whereas DVDs used to cost $25 (US), many can now be found between $5 to $10. It simply isn’t worth peoples’ time to pirate a $5 DVD.
I’m not advocating being able to pirate DVD’s. But despite your hysterical fanaticism, simply playing DVD’s on Linux or other open-source software is as unlike media piracy as your posts are unlike cogent arguments.
Nonsense. For example, customers aren’t inconvenienced by DVD encryption at all — and they’ve gone out and spent tens of BILLIONS of dollars on DVD players and DVD movies. The only ones who are inconvenienced are those who are trying to illegaly copy the media (ie. freaks and geeks).
What an utter load of stinking bullcrap. Why should I have to sit through adverts that are as long as the movie before I get to the main feature? You can do that if you want to (without DRM), but as for me, no thanks.
Yes, there are laws copying media, but DRM covers downloaded content, too. You are right, if it was only about media copying, DRM would be useless.
DG
Consumers usually haven’t heard of dvdcss or something that is needed for playing DVD movies. They don’t know what to look for, they can’t find it.
Automatix2 is your friend. Ubuntu newbies are all urged to install it, and it may be installed by default in the next version.
DRM is very likely to be requirement for
multimedia in near future.
DRM can already be handled by the Linux kernel. However, DRM faces different legal challenges (as it limits what someone can do with the media, and therefore may infring on fair use rights).
Do you expect people to
listen to Stallmans speeches in OGG format, instead of
music ?
Well, if they’re interested in RMS speeches, chances are they’d be open to using an OGG-capable media player. Fortunately, media players are not tied-in to one set of Codecs, and can play a variety of protected and unprotected media.
Get real. No DRM = no multimedia, except pirated.
That’s not true. There will always be unprotected media around, and the fact that a growing portion of consumers grows weary of DRM (remember Sony’s spyware fiasco) will IMO tend to limit it.
Personally, I think media companies are going the wrong way by trying to lock down information. They should stop treating their customers like thieves (DRM is a nuisance to paying customers, but has proven completely ineffective in slowing down piracy) and instead explore other business models. Fortunately, this idea is starting to gain some ground among media execs…
When I saw this article I thought “Poor man, saying one negative word about linux to a bunch of nerds is like mocking the prophet in front of orthodox muslims”
What you have written may be correct, but I would rather have shut up, than provocating all these people. They will, and probable already have torn you into piecses, hacked your computer and posted negative remarks about you all over the internet.
While this comment is likely to be drowned out in the din of normal commenting here I’ll give it a shot anyway.
This article was a very good response to the first post, if a bit to flamish. It is never a good thing to attack the spelling and grammar of someone for whom English is obviously not their native language. Other than that is was ok but for some minor nits I want to pick.
Windows has a much easier to use graphical interface than Linux for novices. Even if you’re an ardent Linux aficionado, you’ve got to concede on this point.
This is a myth. The UI for MS Windows is really quite unintuitive and cumbersome. The only reason everyone thinks it’s easy is that it is known and used by so many people that it is the only thing they know. My son started playing with computers when he was five, mostly for playing games. I have available for him a Linux box, a Win2K box and an iMac with OS X. He had as much trouble with the Win2K box as the Linux one but picked up the Mac so fast he started showing me how to do things. Bottom line; the WinXX GUI is not easier than most X-windows options.
In many ways you can think of this like V for Vendetta. Linux is an idea, and ideas never die. Hell, people still program in Fortran. Why? Because no matter how many companies out there release programming tools to compete with Fortran, it’s an idea that’ll never die.
Here is another myth. Fortran is used because it is the best programming language available to perform certain specific computing jobs. The right tool for the job. Just as COBOL has yet to be replaced by anything for business computing. Just as Assembler
GUI’s are not Microsoft-invented concepts as the previous author implies; indeed many attribute the first GUI to Apple.
Two word: Xerox PARC.
Someone will probably have something to say about these points (if this comment even gets any eyes) but it’s enough to know that I did my part in trying to break the myths holding the computer world in their grips.
Or something like that.
I’m afraid I can’t mod you up. Afraid, because that post deserves it.
Here is another myth. Fortran is used because it is the best programming language available to perform certain specific computing jobs. The right tool for the job. Just as COBOL has yet to be replaced by anything for business computing. Just as Assembler
Absolutely. We’re slowly moving from Fortran on HVTIP to other platforms and languages, but for airline mainframe transaction stuff the Fortran language seems to be pretty decent. Of course, we’ve created our own libraries of external functions over the years so it isn’t just a vanilla Fortran we’re using, but the core language is still Fortran 77. @FTN in Sperry parlance.