Would it be possible though to implement, say, an SVG interface on top of X without breaking anything?
Yes. It's already been done. Next version of GNOME will do all icon rendering through SVG. They could extend that to all window decorations if they wanted to.
In fact, XFree86 offers Display Postscript (basically the same thing as MacOSX display rendering), OpenGL (I don't know of any desktop that renders entirely in OpenGL, but it's possible), and Xlib (which is what GNOME and KDE currently use). Keith has been writing various rendering extensions on top of Xlib to implement alpha blending and so on, but there's no technical impediment preventing GNOME and KDE from dropping Xlib and moving to OpenGL or DPS right now.
But X's latency is TERRIBLE and it takes WAY too long to do things like open windows, resize windows, move windows around, etc.
The latency is higher than some other designs but it's not "TERRIBLE". It's in the same ballpark as most windowing systems. You'd be surprised how little the latency actually matters; humans aren't that quick and computers are. Of course, the application should be designed so as to not exaggerate the latency.
And don't say "they've done tests." Everyone who uses X can tell that it's extremely slow.
I would respectfully suggest that most people don't use X. They use GNOME on top of X or KDE on top of X. This is why you have to run tests; you must test X in isolation to determine how fast X is. Subjective impressions of the entire desktop are useless for pinpointing the precise cause of slow response.
Apparently X is slow because it requires a context switch to kernel space every time you move a window one pixel to the left.
That's one of the reasons, but it's not so bad that a context switch occurs for every "pixel" operation. Commands are collected and flushed at regular intervals.
Additionally, X is unbuffered, so every window beneath that window must redraw itself even if its contents are unchanged.
This is wrong. X11 includes backing store as a standard feature. On XFree86 is it disabled by default because the performance gains were not considered worthwhile, especially considering the large amount of memory that would be wasted.
As has been said many many times before, the network transparency does not affect the local transport. The X team amongst others have done tests (you know, where you measure things), and the implementation (using unix sockets, which are massively efficient data-transports, and shared-memory (no transport at all)) is as fast as you can get. It's within the theoretical margin of error of the peak performance of the system. Nothing goes faster.
Almost. Yes, the XFree86 UNIX socket is a very efficient transport. It doesn't add any measurable overhead compared to shared-memory transports (not the same thing as MITSHM, if anybody wonders about that). Solaris/X has a shared-memory transport IIRC, but the UNIX sockets on Linux are so damn quick it doesn't matter.
But that's not the whole story. There is still the marshalling/unmarshalling of the X11 protocol stream. That adds the same overhead no matter what transport you use. Direct rendering gets rid of that marshalling/unmarshalling wastage. That's why the OpenGL implementation on XFree86 uses DRI if at all possible.
There are also the context switches from the X client to the X server. Once again, direct rendering avoids those context switches because the X client fiddles the hardware directly.
My point is that XFree86 isn't as fast as is possible. We can get a lost faster. Direct rendering is one way to get massive improvements out of Xlib. So your statement "Nothing goes faster" is simply wrong.
X is massively efficient on the local channel. Direct-X on the PC is a different name for the same thing - an API into the low-level drivers.
No. No. No. Direct-X is a direct rendering infrastructure. On XFree86, only OpenGL currently enjoys direct rendering. Xlib is still sent over the socket so there is 1 (possibly 2) redundant copies, marshalling, unmarshalling, and context switches.
X itself is pretty bloody good at getting the maximum performance out of any hardware you throw at it - try running the 2-D blit in X11perf, then multiply the area * bitdepth * fps, divide by your AGP bandwidth and read the number you get.... You'll be surprised if you're running nvidia or ATI cards. Even venerable matrox cards push the bandwidth limit...
And here's the perverse bit: you're right. I've been harping on about the Xlib in XFree86 not being as efficient as possible. I'm about to point out that it doesn't matter! Modern CPUs are so much faster than 2D GPUs or 2D framebuffers that XFree86 easily keeps them fully occupied. Even with all the XFree86 inefficiencies (conservatively estimated between 5-15% for most operations) the video card is still the bottleneck! So optimising Xlib in XFree86 is a waste of time... for now.
If the GPU/CPU ratios ever change then XFree86 will have to change as well, of course. For 3D this has already happened. The traditional method was client -> GLX -> transport -> server -> hardware. The GPUs improved so quickly that XFree86 needed a direct rendering infrastructure for 3D. Now the path is client -> OpenGL/DRI -> hardware. The Xlib path is still client -> Xlib -> transport -> server -> hardware. Maybe in the future we will see client -> Xlib/DRI -> hardware but for now, it doesn't matter, the hardware is already going flat-chat.
WTF? It would be better than the current state of affairs? I'll give you a clue: there is nothing wrong with the current state of affairs. People are paid to work full time on Open-Source software by companies selling support for it. Other people do it on a volonteer basis for free. What is the problem here exactly?
In fact, not only is there no problem, it's a wonderful improvement over the previous models.
Cast your mind back to the late 80s. MS-DOS was dominant but there was no source code. If you found a bug (and there were lots of bugs to be found) then you had three options.
Grin and bear it.
Submit a bug report and pray that Microsoft fixed it in the next release, which you paid for, of course.
Pay heaps of money to Microsoft for a special bug fix; only the very richest companies could afford to do this.
People back then wished fondly that they could get the source code so they could fix the bugs themselves, or hire a contractor at whatever rate to fix the bug.
So if the person before you doesn't feel like "working for free" then they don't have to. They have to pray that Red Hat decides to fix the bug for them. Or if they don't mind "working for free" then they can fix the bug themselves. How is this bad? It's not! It's choice. It's something we didn't have before. You don't have to fix it for free, but you can if you so choose. This is the FREEDOM aspect that RMS keeps harping on about. It's not about cost. It's about freedom of choice.
Users, developers, owners, volunteers, all with the same level of access. All with the same rights to modify the software. It's... utopian.
If she is sharing 1100 songs that SHE DID NOT PAY FOR, then she is stealing from them.
Copyright infringing, not stealing, but at any rate she's being punished well in excess of her "crimes". This is why people are upset. It's like if you're caught speeding, you get fined. People grumble but they accept it. But imagine if the cop pulled you out of the car, called over 3 of his buddies, and they beat you with nightsticks until you couldn't walk. People would say that despite your guilt you didn't deserve to be punished that severely.
There's no way this girl deserved to be fined $3500. She wasn't selling the music. She wasn't defrauding anybody. She was illegally copying the songs but is that so severe a crime? People can cruelly beat a dog to death and get a $1000 fine. You can exceed the speed limit by 20kph in a school zone and pay a $200 fine. But make a copy of Britney Spears for your friend and pay $3500? And that was a settlement figure! The potential fine was 1000s of times higher. That's completely out of proportion.
Jesus christ. She is BREAKING THE LAW. Why shouldn't she be punished?
Because Jesus Christ would have turned the other cheek. The RIAA acts within the law and to maximise their profit, but they are immoral.
I'd honestly like to know what this "real new stuff" in MacOSX is.
The kernel is BSD. That's the 1979 technology that Bill Joy was so quick to dismiss.
The display system is Display PDF. That's not exactly a gigantic leap from Display Postscript in the late 1980s.
The desktop interface is traditional Mac (the menubar at the top) from 1984 with the addition of a panel at the bottom. The panel concept is mid 1980s.
MacOSX is fundamentally minor tweaks on proven technology and proven interface design, using a proven operating system that's older than the Mac itself. I don't see why anybody thinks MacOSX is cutting edge.
That said, I think MacOSX is a sexy interface, the PowerBooks are great value for money, and the entire package is extremely slick. But I'm always baffled when people say shiny buttons demonstrate technological leadership. It's just shiny buttons! The technology in MacOSX is really ancient.
Most people are looking for an experience. If you dig compiling, recompiling and using x systems on top of each other, not allowing for drag and drop between them, can't set your monitor resolution without turning your machine inside out, don't have the luxury of font management, printing, etc etc. then Linux is for you!
I don't know what system you're using, but your experience is not the same as mine.
I haven't compiled anything in years. I did all that back in 1993. These days I just click on the update tool. Why would I want to recompile things? I think you're confusing Linux with BSD.
I drag and drop all the time. My file manager (Nautilus) is completely drag and drop. I drag and drop into my music player (RhythmBox). I drag and drop into the CD burning software. I drag and drop into the word processor (OpenOffice). I drag and drop from my picture editor (also OpenOffice) into the word processor. No hassle.
Changing monitor resolution? I click on the icon that looks like a monitor in the panel and choose the new resolution from the list there. Works great. It's called the GNOME RandR Applet.
Font management? I click on Desktop Preferences then click on Fonts. That's it. Adding new fonts is as simple as dragging and dropping the font into my fonts folder. This baffles you?
Printing? I'm using GNOME CUPS and it's as easy as could be. Click New Printer and choose my model of printer from the list. That's it. I can administer the printer and manage the queues entirely in a graphical interface. Yes, drag and drop works.
Sounds like your experience with Linux is the same as Bill Joy's, but is based on Linux from 1993. Maybe you should update your knowledge before attempting another rant.
If high-performance Linux systems are in production, develop plans that would enable a quick changeover in case SCO wins a favorable judgment and requires the Linux kernel code to be substantially changed. Unix systems are the best alternatives.
No, BSD is the best alternative.
No. BSD had its chance and lost. Linux has a license that forces greedy unscrupulous companies to play fair. The BSD license was too permissive and it almost led to the destruction of UNIX. BSD is not going to get a second chance.
And yes, ls is availible to me on every OSX Mac in the world. However, almost no mac users use it on a daily basis. There's a reason for this.
Uhh, yes... so what's your point?
Are you trying to say that because Linux has ls, just like BSD in 1979, and just like MacOSX today, that Linux is inferior, but MacOSX isn't inferior?
Or are you saying that because Linux users use ls, and Mac users don't, that Linux is inferior to MacOSX? Or inferior to BSD? Does MacOSX become inferior if I choose to use ls?
Or are you saying that because the modern ls manpage is 15 pages long, it is inferior to the 1979 version of ls? Or inferior to the MacOSX ls?
Re-implementing what I designed in 1979 is not interesting to me personally.
And if Linux was entirely about re-implementing what Bill Joy designed in 1979, then he might have a point.
But the things Bill Joy designed and partially wrote back in the 1970s are functionally inferior to features found in modern Linux.
Sure, Linux and BSD share similar APIs, but it is more than a little deceptive to claim that BSD and Linux are the same design. Internally they're completely different.
This is like a 100 year old Mr Ford looking at a modern V8 EFI car with independent suspension and AWD and ABS and saying "pfft, it's not very interesting, I designed all this back in the early 1900s". It shows a complete lack of comprehension regarding the modern state-of-the-art.
Has it occurred to you that, under US law, a CEO must act in the best interests of his shareholders? Bear in mind that the company would have folded within a year without the lawsuit. With knowledge of possible IP infringement by IBM and others, it would have been illegal for these gentlemen not to follow up on it as agressively as possible.
Sure, but he doesn't have to destroy Linux in the process. If SCO has been wronged then SCO will be compensated. But Darl is doing his level best to ruin everything good about Linux in his attempt to make money.
Oh yeah, he says he doesn't want to "destroy" Linux. He's lying. In the CRN interview he states without ambiguity that he wants SCO to be paid for every installation of Linux. He publicly defames the GPL. He attacks the development process. He attacks the developers! He makes critical but not constructive statements about Free Software.
But the worst act of all. The most despicable and heinous act. Darl refuses to help the Linux developers remove the alleged copied code. Darl seems content to allow the alleged infringements to continue in perpetuity so as to line his own pockets. Linux will be destroyed by that act alone. Darl doesn't give a fuck about that. It's his money, right? Linux is his bitch and he can charge whatever he damn well likes. Bad luck to all the millions of developers who poured their heart and soul into creating a community owned project. It's his! He gets to charge $699 for it. Nobody else. Just him!
These are immoral acts. He has every right to defend SCO's property. He has every right to get compensation for use of SCO's property. He has every right to sue IBM so a court can decide the truth of that matter. But he's a fucking prick in the way he's going about it.
I'm sorry if your feelings are hurt, but the bottom line is that he's a brilliant businessman, and he wouldn't have achieved his current wealth were it not for the ability to market a product effectively.
Some people would argue that he achieved his current wealth by screwing the little guy (Seattle Software), illegally abusing Microsoft's monopoly power (Netscape), and being fortunate enough to be in the right place at the right time (PC-DOS).
If IBM had realised the potential for PCs, do you really think they would have handed such a gift to Microsoft? Bill Gates got lucky. Gates is no genius: Paul Allen was the genius behind early Microsoft. Gates doesn't have the brains of a Wozniak. He doesn't even have the entertainment value of an Ellison or Jobs. Bill Gates is rich. That's his sole claim to fame. In computing, Bill is a nobody. There are no famous papers written or even co-written by Gates. His only book is a lame self-serving (and in the first version, dismissive of the Internet) treatise on What Bill Gates Thinks Is Cool. He has written nothing worth remembering. He has coded nothing worth using. He has designed nothing worth learning from. He doesn't teach. He doesn't inspire. He doesn't do anything except accumulate money.
Yes, he will be remembered for being rich. But he won't ever have the respect commanded by the true computing legends like Sutherland, Thompson, Joy, Knuth, Engelbart, Turing, etc. Bill Gates is more famous, has more money, but gains the least respect of any prominent figure in computing that I'm aware of. Do you honestly think this is the sign of a marketing genius? I think it's the sign of a lucky dolt.
This diversity is an important factor in the community you belong to; some groups are more interested in tinkering, in cost, in the functionality and extensibility of Linux, than the political aspects.
The original person belonged to a group that was more interested in Macintoshes, not Linux, so your points have no relevance.
You seem to hold the freedom of Linux in high regard but people actually using that same freedom in contempt.
No. I have no contempt for anybody. I'm a very relaxed and easy-going person.
Realise that there are Linux (and *BSD, and Mac, etc) OSS communities out there that are just as legitimate as the rather narrow one you espouse which is all about idealism.
Of course I recognise the legitimacy of other communities. Nothing I said would imply otherwise.
I think I speak for more people when I say,"Who gives a shit?"
You don't speak for me. I think this is great. Why? Because it means Linux has finally entered the collective consciousness.
What the fuck am I talking about? Parody only works when the audience knows exactly what you're talking about with just the subtlest hints. The spoof didn't need to spell out what Linux is. It was assumed knowledge. Microsoft assumed that the audience knew that Larry Ewing's Tux logo is the Linux mascot, that the audience knew what Linux is and what Linux does, that the audience knew that Linux is competing with Microsoft, and so on.
It is great news that Microsoft acknowledges Linux in this way. It shows that Linux has become big enough to not just attract attention from Microsoft (the Halloween documents demonstrated that) but that everybody in the computing industry is expected to know about Linux by now.
Woohoo. I still remember when this site was called Chips and Dips. I remember the cheering and stomping of feet when the first paper magazine ran an article on Linux. Now Linux inspires spoofs from one of the world's biggest software companies! This is a great day for Linux.
I was a member of a LUG here in the area while in College. Last year it folded, people lost interest. I will tell you why too: OS X. After 10.2, about 80% of the LUG purchased a mac as their next computer including myself. For me, I had the stablity and usablity of a native Unix enviroment and support from hardware and software vendors for products like Photoshop and Quark.
Sure. Stability. Usability. Support. But it's not Free Software. It looks like you never really grokked why GNU/Linux was different. You thought it was just "something better than Windows". Not that there's anything wrong with that, nor is there anything wrong with your new Mac, but the LUG you were in simply wasn't a LUG. It was an "anything-but-Windows" group. Possibly it was a "play-with-the-latest-cool-thing" group. Maybe it was a UNIX group. But it was never a Linux group.
Plus the true user base of Linux in the United States comes from corperate IT staffers in datacenters. To them, its about cost, not community.
It was never about cost. It was about freedom. RMS was right: people never heard the message so they never understood it. UNIX was about community. It died. There have been cost-free operating systems before. They died.
GNU/Linux was about freedom. You had the same rights as the authors. The same rights as the guy next to you, the guy down the street, the big corporate fat cat... you were all equals. You had the same opportunities with Linux as they did.
The cost was never the issue.
Keep in mind that UNIX had the same sense of community as Linux. But UNIX was never owned by the users. It was "stolen" from them by the MegaCorps and many of the UNIX gray-beards felt betrayed by that. Linux guarantees that won't happen again. It's community and freedom that makes Linux something special.
There is a lot of politics of ideals in the OSS world and in particular Linux. In fact, I would call it baggage to many in the business world.
I'm glad your LUG folded. If that was the message you and your friends were propagating then you were doing more harm than good. You sound smart enough but you seem more concerned with the technical gadgetry than the freedom that the software offers. Perhaps you think "it's just software" and concepts like freedom don't apply? I believe both the technical and the political are important but ultimately it will be politics that revolutionises the software industry, not the shiny new features released by MegaCorp #183134.
Now I'll sit back and wait for the accusations of "zealot". It seems Americans all too easily confuse idealism with religion. If you disagree, at least consider the possibility that your own opinion might be wrong before dismissing what I have said. I've been following GNU for 14 years and Linux for 11 years. It has taken me a long time to arrive at my current opinion and I have not made my decision lightly.
We have heard a lot of stories about people, states, and countries moving away from Microsoft. Is this a trend? If you are a manager of a fund heavily invested in MS, or an individual investor, when does this news begin to worry you. In the long run does MS really have a chance when competing against free, well written, well understood software?
Yes. What will happen is, next week Bill Gates will meet with George W Bush. Two weeks from now, The Chimp (as we affectionately call him) will declare Brazil as the new country providing assistance to terrorist training cells. Monkey Man will initiate war against the Brazilians, looking for elusive Weapons of Mass Destruction. The entire country will be destroyed in the Liberation Process. In about 6 months the USA will start a Reconstruction Project which incidentally involves lucrative contracts to USA companies such as Microsoft.
In the USA, free software will simply be ruled illegal as a violation of the DMCA, or the Patriot Act, or something equally non-sensical.
I disagree with people saying that if the computer wins over the human it means that "That's it, here we are, computers are more intelligent than man".
Beyond that, Caldera distributed the code under the GPL.
And the BSD license. Let's not forget that. SCO/Caldera distributed the "new" code (SMP, RCU, NUMA, JFS) under the GPL with their SCOLinux 4.0 product. They also advertised those features in their SCOLinux product brochures so they can't claim ignorance that the code was there. However in the SCOForum slides they were not showing "new" code but extremely old code; in fact, code that's nearly 30 years old! This code was released by SCO/Caldera as "Ancient UNIX" under a BSD license.
Also, SCO/Caldera recently listed the files that supposedly infringe their "UNIX methods" (notice the non-specific SCO/Caldera claim). One file listed contains exactly 6 lines, and the only relevant line is the comment which states that there's no support for SMP yet (in the M68000 tree). This proves beyond any reasonable doubt that SCO hasn't taken due diligence in preparing their list. They've grepped the kernel tree for keywords; they haven't made any attempt to distinguish between SCO/Caldera SMP code and independently developed SMP code. SCO/Caldera definitely does not own all implementations of SMP!
SCO is fishing. It's obvious they're fishing. IBM has even accused them of fishing, in writing, in a court document!
Or are you one of those who thinks all religious people are crazy/lunatic?
Are you saying they aren't? If I was to start speaking to mysterious invisible beings, despite never actually hearing from them or seeing any signs of their presence, and then I start donating significant amounts of my time and money to people who claim they can converse with these invisible beings, wouldn't you think that's a little crazy?
Or does it stop being crazy when millions of people share the same delusion?
Yes, I'm trolling, but only because there's a hint of truth in what I say.
I really wish we could abolish "Left", "Right", "Liberal", and "Conservative" from political language. They've become no more than insults. The "Left" is in charge? Oh, then the "Right" is a bunch of evil zealots come to crush us under their heels! The "Right" is in charge? Now the "Left" is a bunch of evil terrorist sympathizers who want to bring about the downfall of America!
Well, your sort would say that, you damn communist!
You could buy an official port of CDE for Linux in 1995. [...] If you're only going to include the free WMs and DEs for Linux then you're slanting the facts to win the argument.
Yep, got me there. I didn't consider a product maybe 0.1% of people used.
Do you think that's a rebuttal? You say "blah blah didn't have this". I say "yes it did". You say "but not many people used it!". That's maybe the dumbest counter-argument I've ever heard.
KDE was 1996 and GNOME was 1997.
Some people disagree with you. [links to 1.0 announcements]
And if you only consider 1.0 releases, then Linux didn't start in 1991, but only in 1994. Clearly that's false reasoning in the world of open source where pre-1.0 releases are often considered useful.
Well, it must have been some custom-made release of CDE, because CDE doesn't have a Windows 95-like taskbar - now or in 1995.
I never said it was exactly the same, but the CDE Taskbar was common parlance and that exact phrase was used in numerous CDE books (also CDE Panel and CDE Wharf are acceptable variants). Many of the ideas from the CDE Taskbar were copied into the OS/2 Taskbar, the Windows 3.0 Dashboard (third party add-on) and subsequently into the Windows 95 Start Bar. It's fairly pathetic of you to say "but it doesn't have this obscure little feature that the Windows version offers" and thus claim that CDE didn't have the concept at all.
Ah, no. My version of Linux history is relevant because I saw most of it but only commented on the part of it 99.9% of people would have experienced. If you consider CDE in 1995 to be a major part of Linux history, you obviously kept to different circles than me.
The conversation was about inspiration - where the ideas came from and which desktops offered the feature first - not about marketshare. You are conveniently changing the argument into "what was really popular" so you can not admit your mistake. Fairly juvenille, I think.
Even so, Windows 95 was (and remains) a better GUI than CDE.
And now you're twisting the conversation into some subjective claim about which one is "better". Lame. Lame. Lame. If you can't win the original argument then walk away from it, admit defeat, or change your opinion... but DON'T change the argument.
I don't need to look at a screenshot. I used CDE for many years before switching to KDE 1.0 (and I'd used twm or fvwm for many years before that). It's called the CDE Taskbar (or the CDE Panel, or the CDE Wharf) and that terminology predates Windows 95.
The state of Linux "GUIs" (and I use the term loosely) in 1995 wasn't even competitive with Windows *3.0*, let alone Windows 95, OS/2 or MacOS.
I disagree. You could buy an official port of CDE for Linux in 1995. CDE was on-par with Windows 3.0. Not surprising, seeing as Microsoft was a member of the CDE consortium and Windows 3.0 used all the standard CDE conventions (eg, alt-F4, ctrl-C, ctrl-V, the window decorations, etc).
If you're only going to include the free WMs and DEs for Linux then you're slanting the facts to win the argument.
The first WM+other stuff for Linux that could rightfully be called "GUIs" were probably KDE and GNOME, and early versions were pretty light-on in terms of functionality (in its defense, many of the deficincies were due to a lack of "KDE compliant" applications). That was ca. 1998.
KDE was 1996 and GNOME was 1997.
You remember wrong. IIRC FVWM95 was the first WM to try and clone the taskbar. Have a think about why it was called FVWM*95*.
I first used a taskbar with CDE, well before 1995. CDE wasn't just a WM (the WM in CDE was MWM). It defined many of the things that you claim didn't exist:
Yes, yes, yes, yes, yes, yes. CDE had them all. KDE was originally an attempt to create a free CDE (thankfully the KDE people overshot their goal and created something much better).
Your vision of Linux history is tainted because you only saw a small part of it.
Recall that Australia [wikipedia.org] was Great Britain's prison state, during the heydey of the Empire.
And let's not forget that America is now infested by the descendants of terrorists who attacked the British King's legally appointed rulers of the North American colonies.
Slashdot Mirror
Yes. It's already been done. Next version of GNOME will do all icon rendering through SVG. They could extend that to all window decorations if they wanted to.
In fact, XFree86 offers Display Postscript (basically the same thing as MacOSX display rendering), OpenGL (I don't know of any desktop that renders entirely in OpenGL, but it's possible), and Xlib (which is what GNOME and KDE currently use). Keith has been writing various rendering extensions on top of Xlib to implement alpha blending and so on, but there's no technical impediment preventing GNOME and KDE from dropping Xlib and moving to OpenGL or DPS right now.
The latency is higher than some other designs but it's not "TERRIBLE". It's in the same ballpark as most windowing systems. You'd be surprised how little the latency actually matters; humans aren't that quick and computers are. Of course, the application should be designed so as to not exaggerate the latency.
I would respectfully suggest that most people don't use X. They use GNOME on top of X or KDE on top of X. This is why you have to run tests; you must test X in isolation to determine how fast X is. Subjective impressions of the entire desktop are useless for pinpointing the precise cause of slow response.
That's one of the reasons, but it's not so bad that a context switch occurs for every "pixel" operation. Commands are collected and flushed at regular intervals.
This is wrong. X11 includes backing store as a standard feature. On XFree86 is it disabled by default because the performance gains were not considered worthwhile, especially considering the large amount of memory that would be wasted.
Almost. Yes, the XFree86 UNIX socket is a very efficient transport. It doesn't add any measurable overhead compared to shared-memory transports (not the same thing as MITSHM, if anybody wonders about that). Solaris/X has a shared-memory transport IIRC, but the UNIX sockets on Linux are so damn quick it doesn't matter.
But that's not the whole story. There is still the marshalling/unmarshalling of the X11 protocol stream. That adds the same overhead no matter what transport you use. Direct rendering gets rid of that marshalling/unmarshalling wastage. That's why the OpenGL implementation on XFree86 uses DRI if at all possible.
There are also the context switches from the X client to the X server. Once again, direct rendering avoids those context switches because the X client fiddles the hardware directly.
My point is that XFree86 isn't as fast as is possible. We can get a lost faster. Direct rendering is one way to get massive improvements out of Xlib. So your statement "Nothing goes faster" is simply wrong.
No. No. No. Direct-X is a direct rendering infrastructure. On XFree86, only OpenGL currently enjoys direct rendering. Xlib is still sent over the socket so there is 1 (possibly 2) redundant copies, marshalling, unmarshalling, and context switches.
And here's the perverse bit: you're right. I've been harping on about the Xlib in XFree86 not being as efficient as possible. I'm about to point out that it doesn't matter! Modern CPUs are so much faster than 2D GPUs or 2D framebuffers that XFree86 easily keeps them fully occupied. Even with all the XFree86 inefficiencies (conservatively estimated between 5-15% for most operations) the video card is still the bottleneck! So optimising Xlib in XFree86 is a waste of time... for now.
If the GPU/CPU ratios ever change then XFree86 will have to change as well, of course. For 3D this has already happened. The traditional method was client -> GLX -> transport -> server -> hardware. The GPUs improved so quickly that XFree86 needed a direct rendering infrastructure for 3D. Now the path is client -> OpenGL/DRI -> hardware. The Xlib path is still client -> Xlib -> transport -> server -> hardware. Maybe in the future we will see client -> Xlib/DRI -> hardware but for now, it doesn't matter, the hardware is already going flat-chat.
In fact, not only is there no problem, it's a wonderful improvement over the previous models.
Cast your mind back to the late 80s. MS-DOS was dominant but there was no source code. If you found a bug (and there were lots of bugs to be found) then you had three options.
People back then wished fondly that they could get the source code so they could fix the bugs themselves, or hire a contractor at whatever rate to fix the bug.
So if the person before you doesn't feel like "working for free" then they don't have to. They have to pray that Red Hat decides to fix the bug for them. Or if they don't mind "working for free" then they can fix the bug themselves. How is this bad? It's not! It's choice. It's something we didn't have before. You don't have to fix it for free, but you can if you so choose. This is the FREEDOM aspect that RMS keeps harping on about. It's not about cost. It's about freedom of choice.
Users, developers, owners, volunteers, all with the same level of access. All with the same rights to modify the software. It's ... utopian.
Copyright infringing, not stealing, but at any rate she's being punished well in excess of her "crimes". This is why people are upset. It's like if you're caught speeding, you get fined. People grumble but they accept it. But imagine if the cop pulled you out of the car, called over 3 of his buddies, and they beat you with nightsticks until you couldn't walk. People would say that despite your guilt you didn't deserve to be punished that severely.
There's no way this girl deserved to be fined $3500. She wasn't selling the music. She wasn't defrauding anybody. She was illegally copying the songs but is that so severe a crime? People can cruelly beat a dog to death and get a $1000 fine. You can exceed the speed limit by 20kph in a school zone and pay a $200 fine. But make a copy of Britney Spears for your friend and pay $3500? And that was a settlement figure! The potential fine was 1000s of times higher. That's completely out of proportion.
Because Jesus Christ would have turned the other cheek. The RIAA acts within the law and to maximise their profit, but they are immoral.
I'd honestly like to know what this "real new stuff" in MacOSX is.
The kernel is BSD. That's the 1979 technology that Bill Joy was so quick to dismiss.
The display system is Display PDF. That's not exactly a gigantic leap from Display Postscript in the late 1980s.
The desktop interface is traditional Mac (the menubar at the top) from 1984 with the addition of a panel at the bottom. The panel concept is mid 1980s.
MacOSX is fundamentally minor tweaks on proven technology and proven interface design, using a proven operating system that's older than the Mac itself. I don't see why anybody thinks MacOSX is cutting edge.
That said, I think MacOSX is a sexy interface, the PowerBooks are great value for money, and the entire package is extremely slick. But I'm always baffled when people say shiny buttons demonstrate technological leadership. It's just shiny buttons! The technology in MacOSX is really ancient.
I don't know what system you're using, but your experience is not the same as mine.
I haven't compiled anything in years. I did all that back in 1993. These days I just click on the update tool. Why would I want to recompile things? I think you're confusing Linux with BSD.
I drag and drop all the time. My file manager (Nautilus) is completely drag and drop. I drag and drop into my music player (RhythmBox). I drag and drop into the CD burning software. I drag and drop into the word processor (OpenOffice). I drag and drop from my picture editor (also OpenOffice) into the word processor. No hassle.
Changing monitor resolution? I click on the icon that looks like a monitor in the panel and choose the new resolution from the list there. Works great. It's called the GNOME RandR Applet.
Font management? I click on Desktop Preferences then click on Fonts. That's it. Adding new fonts is as simple as dragging and dropping the font into my fonts folder. This baffles you?
Printing? I'm using GNOME CUPS and it's as easy as could be. Click New Printer and choose my model of printer from the list. That's it. I can administer the printer and manage the queues entirely in a graphical interface. Yes, drag and drop works.
Sounds like your experience with Linux is the same as Bill Joy's, but is based on Linux from 1993. Maybe you should update your knowledge before attempting another rant.
No. BSD had its chance and lost. Linux has a license that forces greedy unscrupulous companies to play fair. The BSD license was too permissive and it almost led to the destruction of UNIX. BSD is not going to get a second chance.
Linux 2.0 is the best alternative.
Uhh, yes... so what's your point?
Are you trying to say that because Linux has ls, just like BSD in 1979, and just like MacOSX today, that Linux is inferior, but MacOSX isn't inferior?
Or are you saying that because Linux users use ls, and Mac users don't, that Linux is inferior to MacOSX? Or inferior to BSD? Does MacOSX become inferior if I choose to use ls?
Or are you saying that because the modern ls manpage is 15 pages long, it is inferior to the 1979 version of ls? Or inferior to the MacOSX ls?
Please enlighten us with your wisdom.
And if Linux was entirely about re-implementing what Bill Joy designed in 1979, then he might have a point.
But the things Bill Joy designed and partially wrote back in the 1970s are functionally inferior to features found in modern Linux.
Sure, Linux and BSD share similar APIs, but it is more than a little deceptive to claim that BSD and Linux are the same design. Internally they're completely different.
This is like a 100 year old Mr Ford looking at a modern V8 EFI car with independent suspension and AWD and ABS and saying "pfft, it's not very interesting, I designed all this back in the early 1900s". It shows a complete lack of comprehension regarding the modern state-of-the-art.
Well, first I'd buy a really big aquarium to hold the 120 sharks...
Sure, but he doesn't have to destroy Linux in the process. If SCO has been wronged then SCO will be compensated. But Darl is doing his level best to ruin everything good about Linux in his attempt to make money.
Oh yeah, he says he doesn't want to "destroy" Linux. He's lying. In the CRN interview he states without ambiguity that he wants SCO to be paid for every installation of Linux. He publicly defames the GPL. He attacks the development process. He attacks the developers! He makes critical but not constructive statements about Free Software.
But the worst act of all. The most despicable and heinous act. Darl refuses to help the Linux developers remove the alleged copied code. Darl seems content to allow the alleged infringements to continue in perpetuity so as to line his own pockets. Linux will be destroyed by that act alone. Darl doesn't give a fuck about that. It's his money, right? Linux is his bitch and he can charge whatever he damn well likes. Bad luck to all the millions of developers who poured their heart and soul into creating a community owned project. It's his! He gets to charge $699 for it. Nobody else. Just him!
These are immoral acts. He has every right to defend SCO's property. He has every right to get compensation for use of SCO's property. He has every right to sue IBM so a court can decide the truth of that matter. But he's a fucking prick in the way he's going about it.
I hope he goes to prison.
Some people would argue that he achieved his current wealth by screwing the little guy (Seattle Software), illegally abusing Microsoft's monopoly power (Netscape), and being fortunate enough to be in the right place at the right time (PC-DOS).
If IBM had realised the potential for PCs, do you really think they would have handed such a gift to Microsoft? Bill Gates got lucky. Gates is no genius: Paul Allen was the genius behind early Microsoft. Gates doesn't have the brains of a Wozniak. He doesn't even have the entertainment value of an Ellison or Jobs. Bill Gates is rich. That's his sole claim to fame. In computing, Bill is a nobody. There are no famous papers written or even co-written by Gates. His only book is a lame self-serving (and in the first version, dismissive of the Internet) treatise on What Bill Gates Thinks Is Cool. He has written nothing worth remembering. He has coded nothing worth using. He has designed nothing worth learning from. He doesn't teach. He doesn't inspire. He doesn't do anything except accumulate money.
Yes, he will be remembered for being rich. But he won't ever have the respect commanded by the true computing legends like Sutherland, Thompson, Joy, Knuth, Engelbart, Turing, etc. Bill Gates is more famous, has more money, but gains the least respect of any prominent figure in computing that I'm aware of. Do you honestly think this is the sign of a marketing genius? I think it's the sign of a lucky dolt.
The original person belonged to a group that was more interested in Macintoshes, not Linux, so your points have no relevance.
No. I have no contempt for anybody. I'm a very relaxed and easy-going person.
Of course I recognise the legitimacy of other communities. Nothing I said would imply otherwise.
You don't speak for me. I think this is great. Why? Because it means Linux has finally entered the collective consciousness.
What the fuck am I talking about? Parody only works when the audience knows exactly what you're talking about with just the subtlest hints. The spoof didn't need to spell out what Linux is. It was assumed knowledge. Microsoft assumed that the audience knew that Larry Ewing's Tux logo is the Linux mascot, that the audience knew what Linux is and what Linux does, that the audience knew that Linux is competing with Microsoft, and so on.
It is great news that Microsoft acknowledges Linux in this way. It shows that Linux has become big enough to not just attract attention from Microsoft (the Halloween documents demonstrated that) but that everybody in the computing industry is expected to know about Linux by now.
Woohoo. I still remember when this site was called Chips and Dips. I remember the cheering and stomping of feet when the first paper magazine ran an article on Linux. Now Linux inspires spoofs from one of the world's biggest software companies! This is a great day for Linux.
Sure. Stability. Usability. Support. But it's not Free Software. It looks like you never really grokked why GNU/Linux was different. You thought it was just "something better than Windows". Not that there's anything wrong with that, nor is there anything wrong with your new Mac, but the LUG you were in simply wasn't a LUG. It was an "anything-but-Windows" group. Possibly it was a "play-with-the-latest-cool-thing" group. Maybe it was a UNIX group. But it was never a Linux group.
It was never about cost. It was about freedom. RMS was right: people never heard the message so they never understood it. UNIX was about community. It died. There have been cost-free operating systems before. They died.
GNU/Linux was about freedom. You had the same rights as the authors. The same rights as the guy next to you, the guy down the street, the big corporate fat cat... you were all equals. You had the same opportunities with Linux as they did.
The cost was never the issue.
Keep in mind that UNIX had the same sense of community as Linux. But UNIX was never owned by the users. It was "stolen" from them by the MegaCorps and many of the UNIX gray-beards felt betrayed by that. Linux guarantees that won't happen again. It's community and freedom that makes Linux something special.
I'm glad your LUG folded. If that was the message you and your friends were propagating then you were doing more harm than good. You sound smart enough but you seem more concerned with the technical gadgetry than the freedom that the software offers. Perhaps you think "it's just software" and concepts like freedom don't apply? I believe both the technical and the political are important but ultimately it will be politics that revolutionises the software industry, not the shiny new features released by MegaCorp #183134.
Now I'll sit back and wait for the accusations of "zealot". It seems Americans all too easily confuse idealism with religion. If you disagree, at least consider the possibility that your own opinion might be wrong before dismissing what I have said. I've been following GNU for 14 years and Linux for 11 years. It has taken me a long time to arrive at my current opinion and I have not made my decision lightly.
Yes. What will happen is, next week Bill Gates will meet with George W Bush. Two weeks from now, The Chimp (as we affectionately call him) will declare Brazil as the new country providing assistance to terrorist training cells. Monkey Man will initiate war against the Brazilians, looking for elusive Weapons of Mass Destruction. The entire country will be destroyed in the Liberation Process. In about 6 months the USA will start a Reconstruction Project which incidentally involves lucrative contracts to USA companies such as Microsoft.
In the USA, free software will simply be ruled illegal as a violation of the DMCA, or the Patriot Act, or something equally non-sensical.
Cynical? I don't know the meaning of the word.
And I've yet to find anybody saying such a thing.
And the BSD license. Let's not forget that. SCO/Caldera distributed the "new" code (SMP, RCU, NUMA, JFS) under the GPL with their SCOLinux 4.0 product. They also advertised those features in their SCOLinux product brochures so they can't claim ignorance that the code was there. However in the SCOForum slides they were not showing "new" code but extremely old code; in fact, code that's nearly 30 years old! This code was released by SCO/Caldera as "Ancient UNIX" under a BSD license.
Also, SCO/Caldera recently listed the files that supposedly infringe their "UNIX methods" (notice the non-specific SCO/Caldera claim). One file listed contains exactly 6 lines, and the only relevant line is the comment which states that there's no support for SMP yet (in the M68000 tree). This proves beyond any reasonable doubt that SCO hasn't taken due diligence in preparing their list. They've grepped the kernel tree for keywords; they haven't made any attempt to distinguish between SCO/Caldera SMP code and independently developed SMP code. SCO/Caldera definitely does not own all implementations of SMP!
SCO is fishing. It's obvious they're fishing. IBM has even accused them of fishing, in writing, in a court document!
Are you saying they aren't? If I was to start speaking to mysterious invisible beings, despite never actually hearing from them or seeing any signs of their presence, and then I start donating significant amounts of my time and money to people who claim they can converse with these invisible beings, wouldn't you think that's a little crazy?
Or does it stop being crazy when millions of people share the same delusion?
Yes, I'm trolling, but only because there's a hint of truth in what I say.
Well, your sort would say that, you damn communist!
Do you think that's a rebuttal? You say "blah blah didn't have this". I say "yes it did". You say "but not many people used it!". That's maybe the dumbest counter-argument I've ever heard.
And if you only consider 1.0 releases, then Linux didn't start in 1991, but only in 1994. Clearly that's false reasoning in the world of open source where pre-1.0 releases are often considered useful.
I never said it was exactly the same, but the CDE Taskbar was common parlance and that exact phrase was used in numerous CDE books (also CDE Panel and CDE Wharf are acceptable variants). Many of the ideas from the CDE Taskbar were copied into the OS/2 Taskbar, the Windows 3.0 Dashboard (third party add-on) and subsequently into the Windows 95 Start Bar. It's fairly pathetic of you to say "but it doesn't have this obscure little feature that the Windows version offers" and thus claim that CDE didn't have the concept at all.
The conversation was about inspiration - where the ideas came from and which desktops offered the feature first - not about marketshare. You are conveniently changing the argument into "what was really popular" so you can not admit your mistake. Fairly juvenille, I think.
And now you're twisting the conversation into some subjective claim about which one is "better". Lame. Lame. Lame. If you can't win the original argument then walk away from it, admit defeat, or change your opinion... but DON'T change the argument.
I don't need to look at a screenshot. I used CDE for many years before switching to KDE 1.0 (and I'd used twm or fvwm for many years before that). It's called the CDE Taskbar (or the CDE Panel, or the CDE Wharf) and that terminology predates Windows 95.
I disagree. You could buy an official port of CDE for Linux in 1995. CDE was on-par with Windows 3.0. Not surprising, seeing as Microsoft was a member of the CDE consortium and Windows 3.0 used all the standard CDE conventions (eg, alt-F4, ctrl-C, ctrl-V, the window decorations, etc).
If you're only going to include the free WMs and DEs for Linux then you're slanting the facts to win the argument.
KDE was 1996 and GNOME was 1997.
I first used a taskbar with CDE, well before 1995. CDE wasn't just a WM (the WM in CDE was MWM). It defined many of the things that you claim didn't exist:
Yes, yes, yes, yes, yes, yes. CDE had them all. KDE was originally an attempt to create a free CDE (thankfully the KDE people overshot their goal and created something much better).
Your vision of Linux history is tainted because you only saw a small part of it.
And let's not forget that America is now infested by the descendants of terrorists who attacked the British King's legally appointed rulers of the North American colonies.