Thats only for MS's XAML script language thats in Silverlight/Moonlight, there may be patents MS has on Silverlight that aren't related to XAML. And for Mono, the potential patent issues with Winforms, ADO.NET and ASP.NET still remain (is Moonlight using Winforms, or doing its graphics stuff using GTK via Cairo directly?) In any event, Mono/Winforms is still a minefield.
No, it uses WPF/E. It's fully covered. Moonlight is safe territory. Winforms is a legacy technology, so you wouldn't find it in something like silverlight.
Wake me up when the Mono guys can clear up the issues discussed in the 'Mono and Microsoft's patents' section of Mono's wikipedia page [wikipedia.org], and the Moonlight guys can do the same for the 'Controversy' section on their wikipedia page [wikipedia.org], *and* MS changes the language of their 'Covenant' to include non-Novell users. Until then, Mono/Winforms & Moonlight are still non-starters for any non-Novell Linux distro or user.
Mono is still a brilliant development environment with its open source stack. I can't think of any major Mono applications that use Winforms, ADO.NET, or ASP.NET. This is just more FSF FUD. Core mono and its GPL'd bits like GTK# are well written, designed, and maintained. They're under irrevocable patent protection, making them much safer than technologies like GNUStep or Theora from a patent perspective.
As was pointed out in the argument over HTML5 we had a few days ago, using theora is no more dangerous than using H264, which in turn is no more, or less!, dangerous than using any other software that may have patents on it. We have no way of knowing if MPEGLA is the only entity that has patents on H264. This problem is simply inherent in our screwed-up software patent system.
Where did I ever mention H.264? Although this argument you've had with yourself is amusing, I would feel much safer licensing the codec and letting the licensor deal with any patent situations implicit in the codec, as most real software companies have to do.
A little bit of money to limit liability can be a powerful thing. That's why Alcatel's patent claims on MP3 attacked Microsoft, not all its users.
Protections like the GPL are extremely weak. Code is merely protected art, any amount of reimplementation and re-engineering can destroy any copyright protections. The only truly protected technology is patented. The only safe free technologies patent-wise are those that are patented and community licensed under the OSP, CP, or IBM or the OIN's patent covenants or anything similar.
Software patents exist because software copyrights are extremely flimsy protection of companies' IP. If a software company had no IP, they'd have little to no assets, making the entire software market extremely weak and unstable. It would be a terrible market to invest in. It would be as if the only market was the open source software/UNIX market, weak and outdated technologies, poor implementations, unsupported 1970's technology on every desktop. This would be a market of incremental improvements but utter stagnation in innovation, just like F/OSS or Unix. It would be a bastion of "good enough" solutions and "worse is better" thinking, as the governing trait of software design would be laziness. Companies' ability to own technology provides a market that is safe for investment, so it is well funded and creates an environment where a company like Microsoft or Apple can pour billions into R&D, knowing that it won't simply float into another company.
So, I certainly hope you Freetards never get your way. It will spell the death of powerful innovation and set us back to an era when software was only a stupid toy for manchild academics, like RMS, who would rather software never left the ivory tower became a tool for the masses in industry.
So Moonlight supports all of Silverlight 3's features?
Moonlight 2.0 is very close. Most of the work involved in implementing silverlight was between version 1.0 and 2.0 where the.NET CLR was added. Moonlight 2.0 has preliminary support for many Silverlight 3.0 features already, but the time between Moonlight 2.0 and 3.0 will be much much shorter.
You can develop Flash apps in vim as well. It's been available for some time actually. Flex3 SDK can compile Actionscript files into a SWF file. [wordpress.com]
Right, but comparing Moonlight to Silverlight is sort of unfair, in this case. When will this stuff work in gnash?
There's a fully supported open source implementation of Silverlight. Once Moonlight is at 2.0, it should be quite multi-platform.
..I still think that Microsoft did not understand what the Internet is about: interoperability. You can create whatever nice framework you want - as long as it is not supported by many systems the adoption rate will be slim. If they would make the API a public standard (that is not restricted) then people might adapt, if it is any good.
You mean like Moonlight? The free implementation of Silverlight? Silverlight runs in IE, Firefox on Windows, Safari, Firefox on Mac, and Firefox in Linux (x86 and x64) through Moonlight. It's coming to mobile soon, too.
Now I know, someone will surely insist that the Windows platform still has the majority of the market share and most users don't care, but you see, most users also don't write applications, and as long as you try to feed BS to the later group of people, you are going nowhere.
What? Why does this matter? It's a cross platform and browser dynamic content plugin, not unlike Flash.
Another thing is I see is that the Silverlight frameworks seems to have some severe design issues as it is necessary to bring out a new version seemingly every half year. A well designed platform would try to get the basics right in the first few iterations and then add libraries to it that provide more functionality without having to do a 180 on the whole basic coding.
What the Hell are you talking about? This is a new feature release, not a bugfix. Silverlight 1.1 code will run fine in Silverlight 3, etc. This is simply a newer expanded version of the last framework, as Flash 10 was to Flash 9, or 8. Your "well designed platform" model is exactly what Silverlight is doing. It's specifically a "well designed platform."
Guess this will even turn down Microsoft sympathising developers as they can't keep up with the change that's happening continuously. I mean many people are fed up that everything Microsoft does is obsolete in three years time and you can start anew with learning and development (see VB, classic ASP and so forth).
You can write silverlight apps in vim and run them in a fully opensourced plugin in firefox on linux... it's fully documented publicly unlike Flash. I have linux hacker friends who do just this because silverlight/moonlight can be developed without expensive tools and flash can't. You can even play theora videos natively in IE, Safari/Mac, and Firefox/Linux with it TODAY. I would say the technology is quite freeing.
Another thing is, that though the feature list sounds impressive, there are a lot of undressed issues like security that is a very important one with this kind of networked technology.
Unless you have numbers to back it up, that is an assertion, and not "the fact of the matter". Many benchmarks have shown that OpenGL on Linux is faster than Windows in some configurations. I can't say "most", since I don't have numbers either, but objective benchmarks of both native OpenGL applications and even applications running under Wine do show that X11 is not only acceptable, but faster than Windows. Do you have anything to back up your assertions, or are you just spouting off about your subjective experience that the GNU desktop software feels slow? The latter is something a lot of people complain about, but it's not actually X11's fault.
Slow down before you choke on the irony of your situation, here. Which of these systems have you never used before? Windows or Linux?
Let me make it far simpler:
Firefox runs faster on Windows, Flash runs faster on Windows, hardware video acceleration is available and stable on Windows... this is no fault of the Linux kernel, it's just awkward to step around X. These three issues pretty much represent the crux of what is important on a netbook. Thus, we should assume that X is not fit for the job.
The inability of OSX or Windows to this sort of thing actually matters in an enterprise setting. Whatever issues X has supporting games matters somewhat less. And seeing as Linux is hardly going to beat XBox et al in the games area, maybe the enterprise would be a good place to focus?
Have you never used VNC or RDP? It doesn't matter, anyway. This is off-topic. We're talking about a casual home user web-centric operating system. You fit precisely into the category of people who are using X for what it was built for. It fails miserably at what Google is trying to do with their cute little home OS.
X fails when you are using Linux as an alternative home desktop operating system. It succeeds when you're doing network-transparent applications the "X way."
Oh my god, dude. Do not read the documentation on the modern WDDM... it will break your heart. You're like a child whose puppy died at the vet and no one has broken the news to you... you're still.. just thinking that he lives on a farm somewhere.
I can't believe you think these features are advanced enough to brag about.
Do yourself a favor and boot up Windows on your box and watch a video... enjoy the effortless sound support, and smooth video acceleration. Or activate DWM and move your windows around, watch them not tear... watch the compositing layer not crash. Start up a game and be entranced by what modern graphics hardware is capable of!
Why, you've just listed off a bunch of really basic implementations of hardware acceleration, really life support for X to make it not seem ancient, and yet they're just words. When it all comes down to it, it underperforms in almost any metric of display performance... you can't port games to it, you can't easily accelerate flash on it. It's the reason JavaFX came out on Windows and Mac first, despite the fact that Sun is a major UNIX vendor.
What else is there to say that hasn't been said? It's still constrained through the filesystem socket layer... so you'll always be making more syscalls when performing basic drawing commands. DRI is not broad or extensible enough to take advantage of advanced features on modern graphics hardware, and DRM is fundamentally flawed. It will have to be redesigned if you are ever to get the entirety of OpenGL working.
If you think X is "impressive" because you are able to fire cryptic commands into the CLI and get windows to pop up on different machines, then you need to stay away from discussions on linux for the destkop and restrain yourself. Your corner case is irrelevant on modern hardware, it's difficult to use, and it's people like you who are keeping UNIX from ever having modern display capabilities with your antiquarian usage habits and loud activism thereof. Just keep using X and let regular users have an at least comparable or competitive display system on the linux platform. Let go!
The OS X GUI layer is essentially NeXTStep on a revised Display Postscript. It's slower and more resource intensive than X11, its graphics is targeted primarily at desktop usage. Where is the innovation?
You're right, it's much slower until you try to display video or graphics with it, then it's much faster... not to mention a lot less awkward to program for.
X11 has been innovative from its inception, and it continues to be amazingly innovative today. For example, the kinds of visual effects Compiz delivers effortlessly and cleanly are much harder to achieve in OS X.
You mean 16 year olds can't add plugins to OS X's visual effects framework? The reason you don't see the same plugins is because Compiz has the fashion sense of a 92 Toyota Tercel with 24 inch spinning rims. I've gotten smooth desktop acceleration with my windows moving smoothly on a Powerbook G3 400 with an ATI Rage 128 w/ 8mb of VRAM on Mac OS X Tiger. And you know what? The window management was smoother, lighter and more stable than it was on Ubuntu on a modern Nvidia, ATI, or Intel graphics solution. Compiz is awful, DWM on Vista or 7 makes it look pathetic on the same hardware, with its superior graphics capability and performance, and reduced memory usage. Compiz is not just bad... it's ridiculously awful compared to what Microsoft and Apple provide. So I hope you're kidding, I really do.
What exactly do you think will be the "wake-up call"? Both Gnome and KDE have non-X11 backends, but people don't use them because there really is no benefit associated with getting rid of X11.
How exactly do you "get rid of X"? Does an equivalent even exist on Unix? It sounds like no one does it because no one wants to rewrite the display paradigm on the platform.
A non-X11 backend may make sense for Chrome OS because Chrome OS probably needs less functionality than X11 provides and it makes writing drivers easier. But in terms of innovation and functionality, X11 is second to none.
You're right, it's second to none, it's like fifth. Functionality? What the hell are you talking about? Apple and Microsoft have full network desktop solutions that outperform X (Microsoft's uses less bandwidth, even) and they even offer you direct video hardware access locally. Not only is your display not constrained through a slow and awkward filesystem IO layer, but you also get fast network display access. They beat X both at what it now does (display graphics) and what it was written to do (display graphics over the network). Where do you get this? What can I do with X that I can't do more efficiently in Windows or Mac?
Please, someone present me with some awkward corner case that no home desktop user would ever care about and tell me why writing graphical applications on UNIX is still like listening to an oldies' station.
You say "single socket" as if all of the X clients were contending for a resource. Every connection accepted by a server (any server) creates a new file descriptor in that process. There's no more a problem with a "single socket" in X11 than there is in any other server.
Moreover, unless all of your applications are running over the network, they're almost certainly using shared memory rather than file IO (through the socket) for display. Your entire characterization of X11 shows how little you know about how it works.
This is just details, the fact of the matter is that Microsoft, Apple, and even Be had much faster methods of accessing video hardware and displaying things on the screen. Whether or not it's constrained to a single socket, it's constrained to a socket model and thus the filesystem IO interface. The network transparency would be wonderful if we still had our graphics hardware in separate boxes from our servers, but Google is making a desktop system. DRI/DRM are not really broad enough for modern graphics hardware, anyway. If Google is clever, they'll use their muscle to start from scratch, providing a sane opengl accelerated driver model, like Apple. X's architecture is probably at the peak performance-wise of what the open source community can make it do.
What, you mean framebuffer? Yes, it exists, but it's extremely slow. If there are games that use it, I still wouldn't characterize it as being "for games". OpenGL under X11 is definitely the preferred setup for accelerated video.
It's better, but it's not WDDM... this is just the best of what's currently available. Any graphical application from video playback to 3d will always perform better on Windows than Linux on the same hardware. It doesn't matter whether you're using OpenGL or Direct X, they just have a proper display model for the desktop. The last thing the linux community needs is people trying to pull X into another decade, making it now almost 30 years out of date.
Calling Google Apps "Beta" was likely a pragmatic move on the front of both marketing to bleeding edge internet enthusiasts who are addicted to novelty and engineering in limiting the expectations and liability of google products. They could maintain beta quality products and code and levels of support as long as they kept the beta moniker.
However, I feel that the web's incestuous advertising scheme is beginning to dry up in these times of economic peril, so google needs to go for harder sources of money, like enterprises. Now they're no longer circumventing Microsoft in the market but facing them head to head for a position in the enterprise. Microsoft has as strong position in this market, so they have a certain legacy and stability, which enterprises appreciate.
The first step for Google in combating this will be the simple rebranding of their products to give the semblance of maturity. In reality, any recent changes to the code are minimal to superficial, so this is merely a marketing maneuver and says nothing about the practical roles of beta and gold software in software engineering. It's a welcome change, but it is yet to be seen whether google has the attention span to maintain stable enterprise products. Offering a consistent platform will also open them up to the sort of demonization that Microsoft has faced up until now, as expectations may rise above what they can deliver.
Do any knowledgeable Windows users ever actually get viruses?
I haven't seen one for years. I've only seen one security threat try to attack me since I started using Vista... and it asked UAC for permission to access my system. I denied it and deleted it.
I am seriously asking. All these linux chumps act like the Virus situation in Windows is dire... but I've not really had this experience since the 90's. It was the same with Mac back then, too, though.
Am I just excessively lucky? Am I bizarrely wise for being careful about what downloaded files I allow to privilege escalate?
You're right, I hadn't really thought of it as being in conjunction with the high-availability and high-throughput systems you'd want handling the bulk of the work... the systems handling the trades themselves could be a massive pool of virtual nodes running instances of a high end RTOS... then you could have you high throughput where you need and absolute availability where it's necessary.
I take it you've never seen professional system code, then? If you're curious about that, I think QNX's code is now browseable. Even opensolaris might be a better example.
The code in the Linux kernel does not meet the quality standards of any commercial system code I've seen. In short, it's a total mess. It's archaic, hackish, and just plain ugly.
The debates on the monokernel could go on forever.
Actually, the debate is over. Only amateur developers and CS undergrads argue in favor of monolithic kernels... a modern kernel is a microkernel. The NT kernel is more of a microkernel than even Microsoft wants to admit it is... I've seen the code.
Meanwhile, MS broke the NT security model by moving the GUI into ring 0. Their arguments for doing so and why it wouldn't be a problem are very similar to the arguments in favor of Linux being a monokernel.
NT 6 has userspace video drivers... in fact, the entire WDDM is in userspace, so I am not sure that I would consider their GUI to still be in the kernel.
I also wouldn't characterize the security fixes in the kernel as spackle. Things like that tend to be fixed right. The various subsystems also tend to get considerable review and periodic cleanups.
Have you ever seen the linux kernel code? It's a giant childish mess with no semblance of fore-thought or architecture. It's basically organically grown. The majority of the code comes in from random Chinese outsourcing firms and is briefly glanced at by a large poorly organized team of mediocre developers. It shows in the code. Have you ever been on a linux kernel security mailing list? It's practically a comedy piece because of all the hilarious obvious security exploits that pop up on a regular basis. I guarantee you the people who will break your system are much much more knowledgeable than anyone wasting their time developing it.
As for source availability, it also means that Linux receives thousands of independent security audits on a regular basis.
Yes, thousands of eyes on thousands of little pieces of the system. And yet, it lacks any sort of organization or architecture, so everything is just going in a thousand different directions. In reality, there are probably about 20-30 real security experts in the entire linux community who have to pay attention to the unmanaged work of a thousand developers working on snippets of code. With the inconsistency of the driver API's and the number of eyes looking different directions, I would say sneaking an insecure driver into the kernel would be like sneaking a stick of dynamite into a supermarket. Who's paying enough attention to notice?
It really draws to mind the 1,000,000 monkeys on typewriters writing Shakespeare. A million amateur developers will not create a cohesive system, the best kernels are written and maintained by a small number of people with a strong adherence to a firm design ideal. It's a fantastic example of Quality vs. Quantity.
Besides, any UNIX kernel, no matter how clean, can best be summed up as a "giant collection of hacks." Nothing more.
As for the licensing, I find that the various Linux distros not caring about licenses, keys, serial numbers, and "authenticity" saves a great deal of time and occasionally saves the day. It's all much easier when your OS doesn't figuratively look at you sideways always sustpecting that you're a thief who just hasn't been caught yet.
Yes, IT deals with the licenses during the installation and deployment. It's not the users' problem after that. I would hardly call this an issue vs. unexpected and undocumented behavior.
Actually who planned the entire system that way should get a headsmack, if you need dedicated response times you gotta go with a realtime system, no matter what. You cannot push an os which clearly is not realtime (by Microsofts definition it is soft realtime which means sometimes it reaches it) and on top a vm based system which is not even closely realtime in its garbage collection strategies and then expect to get something useful out of it!
Would you like to recommend a clustering RTOS for the LSE? Somehow, I don't think this is really the sort of place you'd want a full blown RTOS. It seems to me that this could really use a massive high-throughput commercial clustering system like say OpenSolaris with a Java based platform... but that would not be dissimilar from their current WinServer/.NET sysem. If they're afraid of downtime, this would certainly be a terrible place for Linux HPC, which would be down an awful lot more than the occasional day.
I can't imagine what you're hinting at, actually. If you're interested in something like INTEGRITY as the trading platform, I'd have to say I commend you, but it just doesn't seem like the kind of thing that you'd put on big server iron...
I'm not sure what you mean here?!? DMA is a hardware function. non-root processes cannot even touch (or find) the dma control registers. Processes most certainly do NOT share a common address space though that may map a common space into their own (possibly not at the same address) It's not as if one process can just decide to share address space with another. It has to have appropriate permissions.
I am more talking about linux than I am about POSIX, now that I think about it. It's how something as big as linux slides into that model. Most assumptions on the hardness of the UNIX security model are based on classical BSD or UNIX, which are tiny systems running little more than network, tty, and tape drivers. The reality is that there are massive sound, multimeda, and usb subsystems with interfaces in userspace ten times larger than every internal interface of a classical unix kernel. Linux has hundreds of system calls, hell some of its subsystems do. It's a real victim of the monolithic kernel, with putting such a large block of code in one address space. Essentially, it's supermassive and much of it is ridiculously low quality. A skilled attacker with the source code (they have it;)) can simply exploit a poorly written driver and take over a DMA engine.
Besides, you certainly don't need admin privileges to create processes, listening sockets, or just plunder browser caches for personal information and passwords.
So, UNIX is not inherently insecure in its pure form. But linux, as an implementation, is too much ground to cover.
Don't even think to mention Windows in that respect, either, because it doesn't really have a monolithic architecture. It's more architecturally secure as a massive system because it more resembles a microkernel architecture, which is easier to manage and protect.
So, you can have a secure UNIX system and a secure monolith, but linux really outgrew the model where that was secure from remote attack and exploit. It's just spackle upon spackle avoiding known exploits at this point, but in reality it's hanging over an endless abyss of unknown exploits.
And I am quite familiar with the open source "alternatives," but they really don't compare. They just don't. You can site license commercial software for a few grand and save every one of your employees minutes of work each day, hours at times... and all the while end up with better products. It adds up and makes the final cost really moot. If the free product offered suboptimal functionality, it will hurt you enterprise in a fashion that's difficult to conceive.
I'm with you. But I'd actually like to limit the HTML specs even further. Let it be a document format and not a fscking media center.
I have a feeling that people will just start ignoring HTML once it reaches a certain level of inconsistency and bloat. I wonder if XHTML will see a new renaissance as people reach out for a more level-headed document standard.
I really believe plugins were just fine for this sort of thing, given the level of complexity required for rich media applications. They can have a much faster development cycle than HTML specs. Otherwise, you end up with Firefox implementing internal features, then submitting them to some sort of mailing list and claiming they're de facto standards.
It seems like people would just benefit from usable and well documented technologies that put media outside the browser and into the space of consistency and maintainability... but what can I say? I am a pariah.
Moving from object to video has several advantages which are not nullified by using external codecs. The video tag allows standardized integration with the web page and scripting. Just take a look at the video tag demos for Firefox 3.5.
Ah, I see. I hope the standard is strict enough where browser developers won't mess this up.
Actually I think it is good that a list of codecs won't make it into the spec. There are going to be more and better codecs in the future. Do you want to change the spec to keep up? Format negotiation can be handled scriptless on the HTTP level. Just like web browsers indicate the mime-types of images which they can display, they can indicate the mime-types of videos that can be played with the system codecs and the server can choose the file accordingly.
I was discussing this with someone earlier, how it seems like with the lack of a standard codec, H.264/AAC will just continue to be the de facto media standard as Safari and Chrome support it and videos are already encoded that way for Flash. My theory is that when IE implements the video tag, it will just use DirectShow and pull for its pool of codecs. If this is the case, then Windows 7 and later will have these codecs already supported by default.
Otherwise, Microsoft might be well advised to implement HTML 5 compatibility in silverlight, as odd as that sounds.
IMHO this is a grasp for power by the Ogg supporters. The reasons for requiring Ogg formats to be implemented by all browsers are purely political.
I agree completely. Many firms are worried about the technical implications of tying their browser to a certain video format, especially one as immature as theora. Codecs are a sensitive issue for major firms and they're likely to embrace technologies that allow them to provide better content while not being trapped in the design by committee breadline. I feel like the industry has really accepted H.264/AAC, anyway, as much as I dislike Apple.
By not offering anything sudo like until UAC, they pushed customers to prefer the simplicity of always running as admin. People didn't like having to log out and log back in every time something needed Admin privileges.
Run as Administrator was around in XP, actually. Probably before, also. UAC is an extra layer of security for people running as adiminstrators.
In what way do you find the Posix security model a joke? Particularly with ACLs and the other extras added?
The shared memory architecture is inherently insecure. A skilled attacker can DMA all over the system and use the generic and non-obfuscated architecture to do whatever he pleases.
As far as names, you claimed that Linux had nothing to prevent exploits, and I named several things included in Linux to prevent exploits.
Are most linux users using SELinux? I found it rather restricting. It's not unlike using a locked down windows machine. Optional security doesn't assure all users of the platform benefit from it, same problem as Windows.
I was saying that the anti-exploit code is inferior, like the DEP and ASLR... and I stand by tthat. It is. The notion of Linux security is merely de facto based on the small userbase. If it ever became even remotely popular, the facade would collapse.
Personally, I found CDE to be an abomination. KDE or Gnome are much better.
Right, but imagine your workplace needed consistent documents and the ability to create graphics and such. Practical groupware... etc. For any task that requires attractive and functional documents and media, I can't think of a single UNIX application that is even competitive with the commercial Mac/Windows alternatives. Inferior tools means less productive employees and ugly documents, this erodes the professional quality of documents and opresentations within a workplace, making your business appear shoddy. It's not worth it.
You're right on a technical level, you really are. But wouldn't that make playing video on the web more like it was in the web 1.0 era? People would have to stay on top of codecs and go surf for these sorts of things. I believe flash won out originally because it was a seamless solution for the end users-- one plugin to rule them all.
Honestly, the solution you're suggesting is not unlike the way Silverlight/Moonlight handles media-- except that it does have a default/preferred codec.
Why, you could circumvent the lack of a video tag on IE (or anything else) by using the pluggable codec support in Silverlight 3 to provide a Theora codec.;) And that won't require any proprietary tools and very little code- just (if the browser is IE, load the following silverlight control, point it to the codec and your theora video)
We might as well just keep using the object tag to embed media files and let the system figure out what's supposed to run it, if we're going to use system codecs. On Windows, WMP will do it, on Linux, mplayer (or gstreamer if the user is a sadomasochist), and on mac it will be Quicktime. I mean, it's progressive, in an absolutely regressive sort of way.
I second (or fourth?) this advice. Game studios aren't going to care about what obscure technologies you're working on, they're going to care that you found the right tools and started pumping out products. They want to see you get through projects from start to finish. Essentially, they want to know that you'll be adaptable to the technologies they use internally and have the attention span to focus on what building games is really about.
The fun part of game creation is the first moment you think of a new idea, the rest is just painstaking and insane hard work, top to bottom. The game industry is not for the faint of heart, but I can tell you that there's no feeling in the world better than shipping a AAA game.:)
That is a direct result of MS training people (however unintentionally) that they should run as Administrator and (intentionally) that they need no technical knowledge to use and maintain Windows.
Microsoft has been advising users not to run as Administrator for as long as they've been doing multi-user systems, actually... but users prefer the single user system model, since it's simpler. UAC came from the realization that people were just going to run as Adminstrator anyway. It had to be locked down. In Windows 7, it's pretty smooth at this point.
I do find it humorous that you see MAC as inadequate when applied to Linux but somehow virtuous when applied to Windows (several years later). It's just as much a retrofitted old idea from Multics in NT. In both cases, I'd say it was retrofitted because it was a useful idea. I am glad to hear that MS has finally addressed the shatter attack.
I don't care when the features showed up. It isn't relevant. They are competing now and they are competing with NT 6. The point is that you can't just razzle-dazzle me with names of Linux features and expect me to be impressed. You're not talking about SELinux, you're talking about MAC. It's not NX bit, it's DEP, it's not anti-exploit code, it's ASLR. These are generic security features supported in many systems.
The fact of the matter is that Microsoft is doing a great job implementing some of these features. They're doing a better job than the unix people are in many cases. And just because they didn't exist before didn't mean you couldn't run a rock solid and secure NT system. The mighty infallible UNIX security model is a joke and it always will be as long as you have POSIX support... at least Microsoft has control over their architecture, so their system doesn't have to be built on 1970's technology and ideas-- I mean, just look at PowerShell, a fully object oriented CLI/Shell that uses objects instead of strings. See? Progress.
If they're able to get equivalent security model out of their system and offer sane driver API's, proper and documented use of the PCI/PC Specification, correct ACPI, modern graphics technology, usable sound, superior development tools, and a usable office solution, then the license costs should be simply moot. So I need a properly maintained network with an admin or two? All my employees will be more productive. I think the TCO argument is well in the bag here.
UNIX systems are and always have been a nightmare, especially for the regular workstation user. Maybe it'll be a different case when the UNIX world figures out how to write a proper GUI. The last complete and usable desktop solution offered by that community was CDE, and it's really aging now.
So you can argue better theoretical security several years ago... why would that convince me to inflict the terrible terrible productivity applications from the open source world on my workplace? Where's the rest of the TCO picture, here? Every single one of the employees are more expensive than almost all the software licenses combined.
Yes, I wish Mozilla would instead concentrate on advertising the "user doesn't need a mouthful of cocks at all times" feature. This is an area in which Chrome and especially Safari will simply never be able to compete.
Yeah!! You show The Man! He'll never be able to use chemtrails to read your mind!
Do you mean to tell me that Firefox 3.5 succeeded at failing to defeat Safari 4 at the one thing it does well, Javascript performance? Wow!
It also succeeded at failing to defeat Opera and Safari at the only other thing Firefox does well, standards support! They still fail the Acid 3!
Does it also succeed in failing to defeat Opera and IE 8 at page-load speed?
INCREDIBLE!
I think the only thing Firefox does better than other browsers at this point is attracting frothy-mouthed morons to shout their message from the tree tops and aggressively attack users of other browsers. So technically speaking, its only major strength is its wacky collection of extensions... just like IE 6! Welcome to mediocrity, Mozilla!
Slashdot Mirror
Thats only for MS's XAML script language thats in Silverlight/Moonlight, there may be patents MS has on Silverlight that aren't related to XAML. And for Mono, the potential patent issues with Winforms, ADO.NET and ASP.NET still remain (is Moonlight using Winforms, or doing its graphics stuff using GTK via Cairo directly?) In any event, Mono/Winforms is still a minefield.
No, it uses WPF/E. It's fully covered. Moonlight is safe territory. Winforms is a legacy technology, so you wouldn't find it in something like silverlight.
Wake me up when the Mono guys can clear up the issues discussed in the 'Mono and Microsoft's patents' section of Mono's wikipedia page [wikipedia.org], and the Moonlight guys can do the same for the 'Controversy' section on their wikipedia page [wikipedia.org], *and* MS changes the language of their 'Covenant' to include non-Novell users. Until then, Mono/Winforms & Moonlight are still non-starters for any non-Novell Linux distro or user.
Mono is still a brilliant development environment with its open source stack. I can't think of any major Mono applications that use Winforms, ADO.NET, or ASP.NET. This is just more FSF FUD. Core mono and its GPL'd bits like GTK# are well written, designed, and maintained. They're under irrevocable patent protection, making them much safer than technologies like GNUStep or Theora from a patent perspective.
As was pointed out in the argument over HTML5 we had a few days ago, using theora is no more dangerous than using H264, which in turn is no more, or less!, dangerous than using any other software that may have patents on it. We have no way of knowing if MPEGLA is the only entity that has patents on H264. This problem is simply inherent in our screwed-up software patent system.
Where did I ever mention H.264? Although this argument you've had with yourself is amusing, I would feel much safer licensing the codec and letting the licensor deal with any patent situations implicit in the codec, as most real software companies have to do.
A little bit of money to limit liability can be a powerful thing. That's why Alcatel's patent claims on MP3 attacked Microsoft, not all its users.
Protections like the GPL are extremely weak. Code is merely protected art, any amount of reimplementation and re-engineering can destroy any copyright protections. The only truly protected technology is patented. The only safe free technologies patent-wise are those that are patented and community licensed under the OSP, CP, or IBM or the OIN's patent covenants or anything similar.
Software patents exist because software copyrights are extremely flimsy protection of companies' IP. If a software company had no IP, they'd have little to no assets, making the entire software market extremely weak and unstable. It would be a terrible market to invest in. It would be as if the only market was the open source software/UNIX market, weak and outdated technologies, poor implementations, unsupported 1970's technology on every desktop. This would be a market of incremental improvements but utter stagnation in innovation, just like F/OSS or Unix. It would be a bastion of "good enough" solutions and "worse is better" thinking, as the governing trait of software design would be laziness. Companies' ability to own technology provides a market that is safe for investment, so it is well funded and creates an environment where a company like Microsoft or Apple can pour billions into R&D, knowing that it won't simply float into another company.
So, I certainly hope you Freetards never get your way. It will spell the death of powerful innovation and set us back to an era when software was only a stupid toy for manchild academics, like RMS, who would rather software never left the ivory tower became a tool for the masses in industry.
Fully supported... as in "We might not sue you for reimplementing this!"
Hooray for mono!
No, it's patent secured. It's patented but under an irrevocable Open Specification Promise, so they can't sue anyone for implementing the technology.
http://en.wikipedia.org/wiki/Microsoft_Open_Specification_Promise
So, it's actually safer to use than theora, which is under shaky and precarious ground patent-wise.
So Moonlight supports all of Silverlight 3's features?
Moonlight 2.0 is very close. Most of the work involved in implementing silverlight was between version 1.0 and 2.0 where the .NET CLR was added. Moonlight 2.0 has preliminary support for many Silverlight 3.0 features already, but the time between Moonlight 2.0 and 3.0 will be much much shorter.
You can develop Flash apps in vim as well. It's been available for some time actually. Flex3 SDK can compile Actionscript files into a SWF file. [wordpress.com]
Right, but comparing Moonlight to Silverlight is sort of unfair, in this case. When will this stuff work in gnash?
There's a fully supported open source implementation of Silverlight. Once Moonlight is at 2.0, it should be quite multi-platform.
http://go-mono.com/moonlight-preview/
..I still think that Microsoft did not understand what the Internet is about: interoperability. You can create whatever nice framework you want - as long as it is not supported by many systems the adoption rate will be slim. If they would make the API a public standard (that is not restricted) then people might adapt, if it is any good.
You mean like Moonlight? The free implementation of Silverlight? Silverlight runs in IE, Firefox on Windows, Safari, Firefox on Mac, and Firefox in Linux (x86 and x64) through Moonlight. It's coming to mobile soon, too.
Now I know, someone will surely insist that the Windows platform still has the majority of the market share and most users don't care, but you see, most users also don't write applications, and as long as you try to feed BS to the later group of people, you are going nowhere.
What? Why does this matter? It's a cross platform and browser dynamic content plugin, not unlike Flash.
Another thing is I see is that the Silverlight frameworks seems to have some severe design issues as it is necessary to bring out a new version seemingly every half year. A well designed platform would try to get the basics right in the first few iterations and then add libraries to it that provide more functionality without having to do a 180 on the whole basic coding.
What the Hell are you talking about? This is a new feature release, not a bugfix. Silverlight 1.1 code will run fine in Silverlight 3, etc. This is simply a newer expanded version of the last framework, as Flash 10 was to Flash 9, or 8. Your "well designed platform" model is exactly what Silverlight is doing. It's specifically a "well designed platform."
Guess this will even turn down Microsoft sympathising developers as they can't keep up with the change that's happening continuously. I mean many people are fed up that everything Microsoft does is obsolete in three years time and you can start anew with learning and development (see VB, classic ASP and so forth).
You can write silverlight apps in vim and run them in a fully opensourced plugin in firefox on linux... it's fully documented publicly unlike Flash. I have linux hacker friends who do just this because silverlight/moonlight can be developed without expensive tools and flash can't. You can even play theora videos natively in IE, Safari/Mac, and Firefox/Linux with it TODAY. I would say the technology is quite freeing.
Another thing is, that though the feature list sounds impressive, there are a lot of undressed issues like security that is a very important one with this kind of networked technology.
http://blogs.msdn.com/shawnfa/archive/2007/05/09/the-silverlight-security-model.aspx
Unless you have numbers to back it up, that is an assertion, and not "the fact of the matter". Many benchmarks have shown that OpenGL on Linux is faster than Windows in some configurations. I can't say "most", since I don't have numbers either, but objective benchmarks of both native OpenGL applications and even applications running under Wine do show that X11 is not only acceptable, but faster than Windows. Do you have anything to back up your assertions, or are you just spouting off about your subjective experience that the GNU desktop software feels slow? The latter is something a lot of people complain about, but it's not actually X11's fault.
Slow down before you choke on the irony of your situation, here. Which of these systems have you never used before? Windows or Linux?
Let me make it far simpler:
Firefox runs faster on Windows, Flash runs faster on Windows, hardware video acceleration is available and stable on Windows... this is no fault of the Linux kernel, it's just awkward to step around X. These three issues pretty much represent the crux of what is important on a netbook. Thus, we should assume that X is not fit for the job.
The inability of OSX or Windows to this sort of thing actually matters in an enterprise setting. Whatever issues X has supporting games matters somewhat less. And seeing as Linux is hardly going to beat XBox et al in the games area, maybe the enterprise would be a good place to focus?
Have you never used VNC or RDP? It doesn't matter, anyway. This is off-topic. We're talking about a casual home user web-centric operating system. You fit precisely into the category of people who are using X for what it was built for. It fails miserably at what Google is trying to do with their cute little home OS.
X fails when you are using Linux as an alternative home desktop operating system. It succeeds when you're doing network-transparent applications the "X way."
Oh my god, dude. Do not read the documentation on the modern WDDM... it will break your heart. You're like a child whose puppy died at the vet and no one has broken the news to you... you're still.. just thinking that he lives on a farm somewhere.
I can't believe you think these features are advanced enough to brag about.
Do yourself a favor and boot up Windows on your box and watch a video... enjoy the effortless sound support, and smooth video acceleration. Or activate DWM and move your windows around, watch them not tear... watch the compositing layer not crash. Start up a game and be entranced by what modern graphics hardware is capable of!
Why, you've just listed off a bunch of really basic implementations of hardware acceleration, really life support for X to make it not seem ancient, and yet they're just words. When it all comes down to it, it underperforms in almost any metric of display performance... you can't port games to it, you can't easily accelerate flash on it. It's the reason JavaFX came out on Windows and Mac first, despite the fact that Sun is a major UNIX vendor.
What else is there to say that hasn't been said? It's still constrained through the filesystem socket layer... so you'll always be making more syscalls when performing basic drawing commands. DRI is not broad or extensible enough to take advantage of advanced features on modern graphics hardware, and DRM is fundamentally flawed. It will have to be redesigned if you are ever to get the entirety of OpenGL working.
If you think X is "impressive" because you are able to fire cryptic commands into the CLI and get windows to pop up on different machines, then you need to stay away from discussions on linux for the destkop and restrain yourself. Your corner case is irrelevant on modern hardware, it's difficult to use, and it's people like you who are keeping UNIX from ever having modern display capabilities with your antiquarian usage habits and loud activism thereof. Just keep using X and let regular users have an at least comparable or competitive display system on the linux platform. Let go!
The OS X GUI layer is essentially NeXTStep on a revised Display Postscript. It's slower and more resource intensive than X11, its graphics is targeted primarily at desktop usage. Where is the innovation?
You're right, it's much slower until you try to display video or graphics with it, then it's much faster... not to mention a lot less awkward to program for.
X11 has been innovative from its inception, and it continues to be amazingly innovative today. For example, the kinds of visual effects Compiz delivers effortlessly and cleanly are much harder to achieve in OS X.
You mean 16 year olds can't add plugins to OS X's visual effects framework? The reason you don't see the same plugins is because Compiz has the fashion sense of a 92 Toyota Tercel with 24 inch spinning rims. I've gotten smooth desktop acceleration with my windows moving smoothly on a Powerbook G3 400 with an ATI Rage 128 w/ 8mb of VRAM on Mac OS X Tiger. And you know what? The window management was smoother, lighter and more stable than it was on Ubuntu on a modern Nvidia, ATI, or Intel graphics solution. Compiz is awful, DWM on Vista or 7 makes it look pathetic on the same hardware, with its superior graphics capability and performance, and reduced memory usage. Compiz is not just bad... it's ridiculously awful compared to what Microsoft and Apple provide. So I hope you're kidding, I really do.
What exactly do you think will be the "wake-up call"? Both Gnome and KDE have non-X11 backends, but people don't use them because there really is no benefit associated with getting rid of X11.
How exactly do you "get rid of X"? Does an equivalent even exist on Unix? It sounds like no one does it because no one wants to rewrite the display paradigm on the platform.
A non-X11 backend may make sense for Chrome OS because Chrome OS probably needs less functionality than X11 provides and it makes writing drivers easier. But in terms of innovation and functionality, X11 is second to none.
You're right, it's second to none, it's like fifth. Functionality? What the hell are you talking about? Apple and Microsoft have full network desktop solutions that outperform X (Microsoft's uses less bandwidth, even) and they even offer you direct video hardware access locally. Not only is your display not constrained through a slow and awkward filesystem IO layer, but you also get fast network display access. They beat X both at what it now does (display graphics) and what it was written to do (display graphics over the network). Where do you get this? What can I do with X that I can't do more efficiently in Windows or Mac?
Please, someone present me with some awkward corner case that no home desktop user would ever care about and tell me why writing graphical applications on UNIX is still like listening to an oldies' station.
I wouldn't say we "have" Wayland.
You say "single socket" as if all of the X clients were contending for a resource. Every connection accepted by a server (any server) creates a new file descriptor in that process. There's no more a problem with a "single socket" in X11 than there is in any other server.
Moreover, unless all of your applications are running over the network, they're almost certainly using shared memory rather than file IO (through the socket) for display. Your entire characterization of X11 shows how little you know about how it works.
This is just details, the fact of the matter is that Microsoft, Apple, and even Be had much faster methods of accessing video hardware and displaying things on the screen. Whether or not it's constrained to a single socket, it's constrained to a socket model and thus the filesystem IO interface. The network transparency would be wonderful if we still had our graphics hardware in separate boxes from our servers, but Google is making a desktop system. DRI/DRM are not really broad enough for modern graphics hardware, anyway. If Google is clever, they'll use their muscle to start from scratch, providing a sane opengl accelerated driver model, like Apple. X's architecture is probably at the peak performance-wise of what the open source community can make it do.
What, you mean framebuffer? Yes, it exists, but it's extremely slow. If there are games that use it, I still wouldn't characterize it as being "for games". OpenGL under X11 is definitely the preferred setup for accelerated video.
It's better, but it's not WDDM... this is just the best of what's currently available. Any graphical application from video playback to 3d will always perform better on Windows than Linux on the same hardware. It doesn't matter whether you're using OpenGL or Direct X, they just have a proper display model for the desktop. The last thing the linux community needs is people trying to pull X into another decade, making it now almost 30 years out of date.
Calling Google Apps "Beta" was likely a pragmatic move on the front of both marketing to bleeding edge internet enthusiasts who are addicted to novelty and engineering in limiting the expectations and liability of google products. They could maintain beta quality products and code and levels of support as long as they kept the beta moniker.
However, I feel that the web's incestuous advertising scheme is beginning to dry up in these times of economic peril, so google needs to go for harder sources of money, like enterprises. Now they're no longer circumventing Microsoft in the market but facing them head to head for a position in the enterprise. Microsoft has as strong position in this market, so they have a certain legacy and stability, which enterprises appreciate.
The first step for Google in combating this will be the simple rebranding of their products to give the semblance of maturity. In reality, any recent changes to the code are minimal to superficial, so this is merely a marketing maneuver and says nothing about the practical roles of beta and gold software in software engineering. It's a welcome change, but it is yet to be seen whether google has the attention span to maintain stable enterprise products. Offering a consistent platform will also open them up to the sort of demonization that Microsoft has faced up until now, as expectations may rise above what they can deliver.
In short, Google is growing up.
Do any knowledgeable Windows users ever actually get viruses?
I haven't seen one for years. I've only seen one security threat try to attack me since I started using Vista... and it asked UAC for permission to access my system. I denied it and deleted it.
I am seriously asking. All these linux chumps act like the Virus situation in Windows is dire... but I've not really had this experience since the 90's. It was the same with Mac back then, too, though.
Am I just excessively lucky? Am I bizarrely wise for being careful about what downloaded files I allow to privilege escalate?
You're right, I hadn't really thought of it as being in conjunction with the high-availability and high-throughput systems you'd want handling the bulk of the work... the systems handling the trades themselves could be a massive pool of virtual nodes running instances of a high end RTOS... then you could have you high throughput where you need and absolute availability where it's necessary.
I take it you've never seen professional system code, then? If you're curious about that, I think QNX's code is now browseable. Even opensolaris might be a better example.
The code in the Linux kernel does not meet the quality standards of any commercial system code I've seen. In short, it's a total mess. It's archaic, hackish, and just plain ugly.
The debates on the monokernel could go on forever.
Actually, the debate is over. Only amateur developers and CS undergrads argue in favor of monolithic kernels... a modern kernel is a microkernel. The NT kernel is more of a microkernel than even Microsoft wants to admit it is... I've seen the code.
Meanwhile, MS broke the NT security model by moving the GUI into ring 0. Their arguments for doing so and why it wouldn't be a problem are very similar to the arguments in favor of Linux being a monokernel.
NT 6 has userspace video drivers... in fact, the entire WDDM is in userspace, so I am not sure that I would consider their GUI to still be in the kernel.
I also wouldn't characterize the security fixes in the kernel as spackle. Things like that tend to be fixed right. The various subsystems also tend to get considerable review and periodic cleanups.
Have you ever seen the linux kernel code? It's a giant childish mess with no semblance of fore-thought or architecture. It's basically organically grown. The majority of the code comes in from random Chinese outsourcing firms and is briefly glanced at by a large poorly organized team of mediocre developers. It shows in the code. Have you ever been on a linux kernel security mailing list? It's practically a comedy piece because of all the hilarious obvious security exploits that pop up on a regular basis. I guarantee you the people who will break your system are much much more knowledgeable than anyone wasting their time developing it.
As for source availability, it also means that Linux receives thousands of independent security audits on a regular basis.
Yes, thousands of eyes on thousands of little pieces of the system. And yet, it lacks any sort of organization or architecture, so everything is just going in a thousand different directions. In reality, there are probably about 20-30 real security experts in the entire linux community who have to pay attention to the unmanaged work of a thousand developers working on snippets of code. With the inconsistency of the driver API's and the number of eyes looking different directions, I would say sneaking an insecure driver into the kernel would be like sneaking a stick of dynamite into a supermarket. Who's paying enough attention to notice?
It really draws to mind the 1,000,000 monkeys on typewriters writing Shakespeare. A million amateur developers will not create a cohesive system, the best kernels are written and maintained by a small number of people with a strong adherence to a firm design ideal. It's a fantastic example of Quality vs. Quantity.
Besides, any UNIX kernel, no matter how clean, can best be summed up as a "giant collection of hacks." Nothing more.
As for the licensing, I find that the various Linux distros not caring about licenses, keys, serial numbers, and "authenticity" saves a great deal of time and occasionally saves the day. It's all much easier when your OS doesn't figuratively look at you sideways always sustpecting that you're a thief who just hasn't been caught yet.
Yes, IT deals with the licenses during the installation and deployment. It's not the users' problem after that. I would hardly call this an issue vs. unexpected and undocumented behavior.
Actually who planned the entire system that way should get a headsmack, if you need dedicated response times you gotta go with a realtime system, no matter what. You cannot push an os which clearly is not realtime (by Microsofts definition it is soft realtime which means sometimes it reaches it) and on top a vm based system which is not even closely realtime in its garbage collection strategies and then expect to get something useful out of it!
Would you like to recommend a clustering RTOS for the LSE? Somehow, I don't think this is really the sort of place you'd want a full blown RTOS. It seems to me that this could really use a massive high-throughput commercial clustering system like say OpenSolaris with a Java based platform... but that would not be dissimilar from their current WinServer/.NET sysem. If they're afraid of downtime, this would certainly be a terrible place for Linux HPC, which would be down an awful lot more than the occasional day.
I can't imagine what you're hinting at, actually. If you're interested in something like INTEGRITY as the trading platform, I'd have to say I commend you, but it just doesn't seem like the kind of thing that you'd put on big server iron...
I'm not sure what you mean here?!? DMA is a hardware function. non-root processes cannot even touch (or find) the dma control registers. Processes most certainly do NOT share a common address space though that may map a common space into their own (possibly not at the same address) It's not as if one process can just decide to share address space with another. It has to have appropriate permissions.
I am more talking about linux than I am about POSIX, now that I think about it. It's how something as big as linux slides into that model. Most assumptions on the hardness of the UNIX security model are based on classical BSD or UNIX, which are tiny systems running little more than network, tty, and tape drivers. The reality is that there are massive sound, multimeda, and usb subsystems with interfaces in userspace ten times larger than every internal interface of a classical unix kernel. Linux has hundreds of system calls, hell some of its subsystems do. It's a real victim of the monolithic kernel, with putting such a large block of code in one address space. Essentially, it's supermassive and much of it is ridiculously low quality. A skilled attacker with the source code (they have it ;)) can simply exploit a poorly written driver and take over a DMA engine.
Besides, you certainly don't need admin privileges to create processes, listening sockets, or just plunder browser caches for personal information and passwords.
So, UNIX is not inherently insecure in its pure form. But linux, as an implementation, is too much ground to cover.
Don't even think to mention Windows in that respect, either, because it doesn't really have a monolithic architecture. It's more architecturally secure as a massive system because it more resembles a microkernel architecture, which is easier to manage and protect.
So, you can have a secure UNIX system and a secure monolith, but linux really outgrew the model where that was secure from remote attack and exploit. It's just spackle upon spackle avoiding known exploits at this point, but in reality it's hanging over an endless abyss of unknown exploits.
And I am quite familiar with the open source "alternatives," but they really don't compare. They just don't. You can site license commercial software for a few grand and save every one of your employees minutes of work each day, hours at times... and all the while end up with better products. It adds up and makes the final cost really moot. If the free product offered suboptimal functionality, it will hurt you enterprise in a fashion that's difficult to conceive.
I'm with you. But I'd actually like to limit the HTML specs even further. Let it be a document format and not a fscking media center.
I have a feeling that people will just start ignoring HTML once it reaches a certain level of inconsistency and bloat. I wonder if XHTML will see a new renaissance as people reach out for a more level-headed document standard.
I really believe plugins were just fine for this sort of thing, given the level of complexity required for rich media applications. They can have a much faster development cycle than HTML specs. Otherwise, you end up with Firefox implementing internal features, then submitting them to some sort of mailing list and claiming they're de facto standards.
It seems like people would just benefit from usable and well documented technologies that put media outside the browser and into the space of consistency and maintainability... but what can I say? I am a pariah.
Moving from object to video has several advantages which are not nullified by using external codecs. The video tag allows standardized integration with the web page and scripting. Just take a look at the video tag demos for Firefox 3.5.
Ah, I see. I hope the standard is strict enough where browser developers won't mess this up.
Actually I think it is good that a list of codecs won't make it into the spec. There are going to be more and better codecs in the future. Do you want to change the spec to keep up? Format negotiation can be handled scriptless on the HTTP level. Just like web browsers indicate the mime-types of images which they can display, they can indicate the mime-types of videos that can be played with the system codecs and the server can choose the file accordingly.
I was discussing this with someone earlier, how it seems like with the lack of a standard codec, H.264/AAC will just continue to be the de facto media standard as Safari and Chrome support it and videos are already encoded that way for Flash. My theory is that when IE implements the video tag, it will just use DirectShow and pull for its pool of codecs. If this is the case, then Windows 7 and later will have these codecs already supported by default.
Otherwise, Microsoft might be well advised to implement HTML 5 compatibility in silverlight, as odd as that sounds.
IMHO this is a grasp for power by the Ogg supporters. The reasons for requiring Ogg formats to be implemented by all browsers are purely political.
I agree completely. Many firms are worried about the technical implications of tying their browser to a certain video format, especially one as immature as theora. Codecs are a sensitive issue for major firms and they're likely to embrace technologies that allow them to provide better content while not being trapped in the design by committee breadline. I feel like the industry has really accepted H.264/AAC, anyway, as much as I dislike Apple.
By not offering anything sudo like until UAC, they pushed customers to prefer the simplicity of always running as admin. People didn't like having to log out and log back in every time something needed Admin privileges.
Run as Administrator was around in XP, actually. Probably before, also. UAC is an extra layer of security for people running as adiminstrators.
In what way do you find the Posix security model a joke? Particularly with ACLs and the other extras added?
The shared memory architecture is inherently insecure. A skilled attacker can DMA all over the system and use the generic and non-obfuscated architecture to do whatever he pleases.
As far as names, you claimed that Linux had nothing to prevent exploits, and I named several things included in Linux to prevent exploits.
Are most linux users using SELinux? I found it rather restricting. It's not unlike using a locked down windows machine. Optional security doesn't assure all users of the platform benefit from it, same problem as Windows.
I was saying that the anti-exploit code is inferior, like the DEP and ASLR... and I stand by tthat. It is. The notion of Linux security is merely de facto based on the small userbase. If it ever became even remotely popular, the facade would collapse.
Personally, I found CDE to be an abomination. KDE or Gnome are much better.
Right, but imagine your workplace needed consistent documents and the ability to create graphics and such. Practical groupware... etc. For any task that requires attractive and functional documents and media, I can't think of a single UNIX application that is even competitive with the commercial Mac/Windows alternatives. Inferior tools means less productive employees and ugly documents, this erodes the professional quality of documents and opresentations within a workplace, making your business appear shoddy. It's not worth it.
You're right on a technical level, you really are. But wouldn't that make playing video on the web more like it was in the web 1.0 era? People would have to stay on top of codecs and go surf for these sorts of things. I believe flash won out originally because it was a seamless solution for the end users-- one plugin to rule them all.
Honestly, the solution you're suggesting is not unlike the way Silverlight/Moonlight handles media-- except that it does have a default/preferred codec.
Why, you could circumvent the lack of a video tag on IE (or anything else) by using the pluggable codec support in Silverlight 3 to provide a Theora codec. ;) And that won't require any proprietary tools and very little code- just (if the browser is IE, load the following silverlight control, point it to the codec and your theora video)
We might as well just keep using the object tag to embed media files and let the system figure out what's supposed to run it, if we're going to use system codecs. On Windows, WMP will do it, on Linux, mplayer (or gstreamer if the user is a sadomasochist), and on mac it will be Quicktime. I mean, it's progressive, in an absolutely regressive sort of way.
I second (or fourth?) this advice. Game studios aren't going to care about what obscure technologies you're working on, they're going to care that you found the right tools and started pumping out products. They want to see you get through projects from start to finish. Essentially, they want to know that you'll be adaptable to the technologies they use internally and have the attention span to focus on what building games is really about.
The fun part of game creation is the first moment you think of a new idea, the rest is just painstaking and insane hard work, top to bottom. The game industry is not for the faint of heart, but I can tell you that there's no feeling in the world better than shipping a AAA game. :)
That is a direct result of MS training people (however unintentionally) that they should run as Administrator and (intentionally) that they need no technical knowledge to use and maintain Windows.
Microsoft has been advising users not to run as Administrator for as long as they've been doing multi-user systems, actually... but users prefer the single user system model, since it's simpler. UAC came from the realization that people were just going to run as Adminstrator anyway. It had to be locked down. In Windows 7, it's pretty smooth at this point.
I do find it humorous that you see MAC as inadequate when applied to Linux but somehow virtuous when applied to Windows (several years later). It's just as much a retrofitted old idea from Multics in NT. In both cases, I'd say it was retrofitted because it was a useful idea. I am glad to hear that MS has finally addressed the shatter attack.
I don't care when the features showed up. It isn't relevant. They are competing now and they are competing with NT 6. The point is that you can't just razzle-dazzle me with names of Linux features and expect me to be impressed. You're not talking about SELinux, you're talking about MAC. It's not NX bit, it's DEP, it's not anti-exploit code, it's ASLR. These are generic security features supported in many systems.
The fact of the matter is that Microsoft is doing a great job implementing some of these features. They're doing a better job than the unix people are in many cases. And just because they didn't exist before didn't mean you couldn't run a rock solid and secure NT system. The mighty infallible UNIX security model is a joke and it always will be as long as you have POSIX support... at least Microsoft has control over their architecture, so their system doesn't have to be built on 1970's technology and ideas-- I mean, just look at PowerShell, a fully object oriented CLI/Shell that uses objects instead of strings. See? Progress.
If they're able to get equivalent security model out of their system and offer sane driver API's, proper and documented use of the PCI/PC Specification, correct ACPI, modern graphics technology, usable sound, superior development tools, and a usable office solution, then the license costs should be simply moot. So I need a properly maintained network with an admin or two? All my employees will be more productive. I think the TCO argument is well in the bag here.
UNIX systems are and always have been a nightmare, especially for the regular workstation user. Maybe it'll be a different case when the UNIX world figures out how to write a proper GUI. The last complete and usable desktop solution offered by that community was CDE, and it's really aging now.
So you can argue better theoretical security several years ago... why would that convince me to inflict the terrible terrible productivity applications from the open source world on my workplace? Where's the rest of the TCO picture, here? Every single one of the employees are more expensive than almost all the software licenses combined.
Yes, I wish Mozilla would instead concentrate on advertising the "user doesn't need a mouthful of cocks at all times" feature. This is an area in which Chrome and especially Safari will simply never be able to compete.
Yeah!! You show The Man! He'll never be able to use chemtrails to read your mind!
Do you mean to tell me that Firefox 3.5 succeeded at failing to defeat Safari 4 at the one thing it does well, Javascript performance? Wow!
It also succeeded at failing to defeat Opera and Safari at the only other thing Firefox does well, standards support! They still fail the Acid 3!
Does it also succeed in failing to defeat Opera and IE 8 at page-load speed?
INCREDIBLE!
I think the only thing Firefox does better than other browsers at this point is attracting frothy-mouthed morons to shout their message from the tree tops and aggressively attack users of other browsers. So technically speaking, its only major strength is its wacky collection of extensions... just like IE 6! Welcome to mediocrity, Mozilla!