Wrong. Win95 doesn't leave DOS in memory. It moves DOS into a protected mode virtual machine. Win16 apps NEVER call into "real" DOS, they call into a DOS that is running in the (seperate) Win16 virtual machine. All of these run in the Win32 virtual machine. The exit to DOS trick isn't a trick, it unloads Windows, puts the sytem into real mode, and moves DOS back into regular memory. This version of DOS that runs when you "exit to DOS" is the REAL DOS. It runs in real-mode. It doesn't run on top of a VM. What MS has decided to do is remove this DOS. You seem to think that Win95 runs on top of DOS. This is a myth. If it really did run on top of DOS, then most system calls would result in a mode switch. That's simply not the case. The only time Win95 makes calls to DOS is to support old real mode drivers. (Like the original Win95 printer drivers.) There is no malacious intent in here. They figured that nobody needed real-mode DOS anymore (since most DOS apps can run in the virtual machine), and booting into real-mode first interfered with driver loading and made the system unstable, so the got rid of it. No malacious intent here, just some common sense. And it WILL interefer with programs like loadlin or the BeOS loader. They depend on rebooting into DOS mode so they can kick DOS out of memory. Just as these programs won't work on NT, they won't work on ME.
I believe that is true for NT. But at least Win 95 used to run more on top of DOS more than use it only as a loader. There are some evidence for that, like that DOS protected mode memory managers (EMM386 or third party like QEMM) are needed for loading Win 95, and I have seen few times Win 95 crash to a dos prompt! >>>>>>>>>> Win95 can crash to a DOS prompt, because it is still in there. However, that doesn't imply that it runs on top of DOS. Also, memory managers aren't needed for loading Win95, and infact it is reccomended that they are NOT used because it interferes with Win95 managing its own memory. Of course they can be used, but that only applies to when you've got apps that run in the DOS VM and need the services of the memory manager. For example, I had this old DOS-mode pinball game that needed a memory manager to run. However, my Win95 system has no memory managers in it's autoexec, so I doubt they are running. Win95 really isn't DOS in a pretty face. Its mostly Win 3.1 in a pretty face. DOS is still there, but it isn't used. DOS is a real-mode operating system. Unless you are running really old soft/hardware, then your system is running in protected-mode most of the time. The amount of DOS in Win95 and 98 is the same (nearly none.) There aren't really any tweeks to the kernel in Win98 aside from some to the memory manager. As for command.com, read up on some articles around the release of Win95. (From REAL magazines, like BYTE, not ZDnet.) The Win95 architecture runs application in 3 ways. First, there is the system VM for 32 bit applications. Then there are two other VMs (which are Win32 applications and run in the system VM.) There is one VM for Win16 (Win 3.x) applications, and there is one or more (each DOS app runs in it's own VM) VMs for DOS applications. In every case, the system is running in protected mode.
OTOH while Win 98 still needs EMM386 (I haven't tried QEMM on it) I have always got the comfy BSOD when it crashes so maybe there is less DOS in it. >>>>>>>>> I'm running Win98 without EMM386. According to Peter Norten (Norten's Inside the PC, Page 294) "Because of this, you need not specify EMM386.EXE in your CONFIG.SYS file, and you should not specify HIMEM.SYS-both of these will actually diminish Windows 95's capability to internally manager memory" As for QEMM, it is simply a replacement for the aformentioned EMM386.EXE and should not be used in Win9x.
Whoa, cool off. I was kidding. It's a joke, laugh, ha ha...
The DOS prompt is so woefully underpowered, it doesn't deserve to share the same name as beasts such as the UNIX shells. (This coming from a person who likes Windows better than Linux, so you know its not some anit-MS bull!)
Okay, now I can officially say that about half the posters on this topic (even the +4's and +5's) have no clue what they are talking about. To put it succinctly
1) It's bfs, not BeOSfs, don't EVER forget that;)
2) MS is not removing DOS comptability. It is simply removing the ability to boot into real-mode DOS. This is a feature that most people don't even use anymore. Unless you've got really old hardware drivers that need real-mode DOS (very rare) you're machine is in protected mode most of the time. Unless you use the "Reboot into DOS" option, you're not missing anything.
3) It is not removing the command line. From inside Windows, everything will still be the same. The DOS prompt will run on top of a VM (virtual machine) just like it always has, you can run DOS applications from the VM, just like you always have.
4) MS is not removing 16bit compatibility. 16bit compatbility has nothing to do with DOS support. Win9x/ME is still the "24bit" 16/32bit kernel.
5) Stability will only improve slightly. In previous versions of Windows, Windows booted DOS first, which then booted the Win95 kernel. WinME boots straight into Win95, so any stability problems that take down you're machine at startup or shutdown will be alleviated.
6) This will have no effect on bfs/vfat or UMSDOS. Those don't actually depend on DOS. The only thing that will change is that you can't use loadlin anymore.
7) This is not some marketing gimmick, or some way to supress CLI users, or anything like that. DOS mode conflicted with some stuff, it made the machine very vunerable at boot, and very few apps actually needed "pure" DOS mode anymore (as opposed to the VM DOS mode, just so we're clear) so Microsoft decided to get rid of it. It certainly doesn't deserve a nearly 700-post barrage on/.
Wrong. WinME doesn't use WOW. WOW is only for "pure" 32bit versions of NT. 16bit processes are handled by the same "24bit" Win95 kernel. 8 bit processes (just like the 16bit ones) are run in a virtual machine just like they always have. The only thing that has changed is that it is no longer possible to take a Windows95 machine into pure real mode DOS (without VM's or protections running.)
It's called a VM. If you knew how Windows worked, you wouldn't say stupid stuff like that. Command.com (the DOS prompt) is (and always HAS) run under a virtual machine in Windows 95. The machine operates in protected mode, and run a VM to run the DOS prompt. This is different from the straight DOS prompt which is ACTUAL real mode DOS. Windows 95 has NEVER run on top of DOS. The only DOS code (IE non protected mode) is some funky old drivers. The DOS prompt by itself was the last bits of DOS Win95 had. It was real mode DOS in which the entire machine operated in real mode without a VM. They didn't go out of their way to turn it off, they just got rid of it entirely.
A) DOS is gone from WinME. DOS never really was in Win95, it was present, but DOS apps ran in a VM. What Windows ME does is boot straight into protected mode because it helps in booting quickly, and the machine is very vunerable in this state. Also, it gets rid of uneeded code. B) The legacy 16bit stuff is still there. 8 bit stuff has been gone for a long time. Most Win95 machines rarely switch into real mode unless you're running some funky hardware or software. However, all the 16bit stuff is still there. C) It doesn't get rid of the CLI. Just like NT, WinME still has a DOS prompt. D) You call the DOS prompt a CLI?
This is a great product. If you've ever seen the communities that have sprung up over Ti's 8x line, then you'll know what uses these things have been put to. There are people that write custom OSs for TI's calculators (around 8-10 MHz on 32-128 KB of RAM.) People figure out how to make the thing generate sound through the serial port, how to overclock it, how to make it drive a radio, how to write custom assembly for it, all kinds of neat things. It brings a hackerish feeling that you really don't get these days. Programming the TIs is like programming those old computers. You get to program directly to the hardware, you have to come up with tricks to get the thing to do what you want, you have to find nifty uses for the built in hardware. That's real hacking. What really makes them so popular is that they're cheap and flexible. I really hope that HP manages to keep the price of this thing under $150, or else it really won't work. There are already other HPCs on the market, and this product needs to be different. Since it addresses the main problems with TI's series (poor screen resolution, slow screen response, and limited CPU power) this product could really be cool if priced appropriatly.
1) Whether or not the X Extension is trying to compete with Berlin, it is. Simply by their existance, they are competing.
2) What people don't seem to understand is that OSS isn't the magic bullet. Sure there is a near infinate number of developers, but the realities of the market are still present. There is only so much public interest, whether or not a product is Open Source. Public interest is ultimately what defines who wins and who loses. Great projects die because they don't have enough public interest. Say the X extension results in a rendering interface even half as good as Berlin. Now you've got a less superior product. Certianly a more bloated product (because it incurs the bloat of regular X, plus it's own protocol) and one that really is cobled together and "hackish." However, it has backwards comptability, and by that sole trait will win. What happens then? Probably Berlin dies. By stealing Berlin's turf, X leaves Berlin with no reason for existance.
3) Competition isn't the magic bullet either. Competition only works when to products are compatible. Take a look around the software world instead of thinking idealistically. There are a lot of graphics programs competing. Why? Because they are all comptabile with each other. Then, look at Office suites. Notice how much fewer products there are in this space. Why, less comptability. Take a look at operating systems? Notice the fact that there are no competing products. Why? No comptability. When there is no comptability, userbase is everything. Look all over the software world. Everywhere you can see products suceeding soley based on userbase. Wordperfect is a better word processor than Word? Who won? OS/2 is better than Windows. Who won? Why? There is always a winner and a loser, eventually. OSS slows the ending, but doesn't stop it. Ultimately one will win. By introducing competition to an interesting product, then the OSS community is hurting itself. Things would be different if both were competing on the same level, but they aren't. Soley because X has more comptability, it will win. No matter which is the superior product. By supporting one or the other, there are bigger implications than just one project getting more code time. After all, code time is infinate. Interest is not. By drawing interest away from Berlin, the X extension is almost guarenting its demise.
The question of exactly how much devices can merge is moot. There really isn't a limit. It's essentially asking exactly how much people can innovate. As people have learned, there really is NO limit to innovation. Screen size limitations are beaten by things like screens in glasses, or that technology that uses a laser to draw images on the retina, input limitations are overcome by voice control, maybe even direct connections between the brain and the hardware. There really isn't a limit to where you can take convergance. Convergance in the electronic world probably won't stop in a certain place like it has with appliances. With appliances, the limits of the technology are real. A toaster or a fridge will always have to be at least a certian size, you'll always need the same amount of energy to toast something, you'll never be able to fit a blender into a fridge. That's what prevents convergance in most other household objects from continuing. However, there are no such limitations in electronic convergance, becase all electronics are simply mediums for information. There are no physical limitations on something that carries information. For example, when you buy that brand new spanking DVD player, what do you buy it for? You certainly don't want the hardware, you want to play a DVD. You don't even want to play the DVD, you want to watch the movie. Hell, you don't even want to watch the movie, you want to stimulate you retina's into thinking you're watching a movie, etc. Because the actual "product" in the case of electronics is simply information, ocnvergance can continue until there is no medium left between the information and the consumer.
I didn't consider it a troll spam. This guy had a reasonable point on the surface. Why does/. say one thing about a company, but through it's advertisements (which imply a sponsering of the product) say something else?
Ah, but here is the beauty of it. At least you know that/. isn't being influenced by it's advertisers. I mean you see this on MaximumPC all the time. People post and advertisement for some product, and on the very next page is a review that spanks the product to all hell. It's a show of journalistic integrity, not hipocracy. A site's advertisements doesn't (and SHOULDN'T) having anything to do with the site's content.
How 'bout this? Stop being so petty! I know programmers are jaded, but hey, even we appreciate a nifty sounding product. Plus, it helps you remember. X-Box SDK, Java SDK, SDL-SDK. How are you going to stand out in the crowd?
That's a little unfair. If I go to the post about KDE 1.92's release, and ask for a filter to get rid of KDE/Linux stuff, then I'm a -1. Asking to filter out Microsoft stuff, and it's untouched.
The hipocracy's so thick you can cut it with a knife.
This is quite an interesting position for NVIDIA to be in. Observations
1) NVIDIA bases all its chips on the same driver model. This means that all its chips have similar registers/control sets.
2) NVIDIA doesn't release register-level specs.
3) Console developer's are almost required to give out register-level specs. The game developers love the ability to program directly to the metal (as shown by the fact that this Win2K runs apps in Ring0) and they do a much better job of finding uses for the hardware than the console developers do.
Point: If NVIDIA gives out the specs to this thing, then people can get to work writing driveres for non-Linux/Windows OSs. If NVIDIA doesn't give away specs, then they will get trounced by PS2 developers who find clever tricks to make the hardware do stuff it wasn't designed to. (For example, Crash3 on PSX uses transparency. Technically, PSX doesn't DO transparency.) Maybe they'll release the SDK under NDA?
In response to the first comment, I think I got off on a tangent talking about GNOME. My point was that a good interface should be responsive. Engineering all this "neato" features may be more sexy, but in the end, 99% of people won't use these features, but WILL notice that their interface is slower.
As for Microsoft, don't get me wrong. I am a big Microsoft fan, and I think it would be wise for Linux projects to emulate some of their practices. However, the fact remains that Windows by and large is heavily deteriorated by feature bloat. All this "whiz-bang" features that really aren't necessary hurt the environement. However, that is forgivable in Microsoft's case, they have upgrades to sell. However, projects like GNOME don't have that agenda.
As for COM, it is a gift from god. COM and DirectX are two technologies sorely missing from Linux. However, CORBA and COM are two entirely different beasts. Wheras COM is a (relativly) simple API, CORBA is quite a beast. Look at the two APIs. You can actually write something COM-like in a week or two. Most of the stuff is simply a standard and the C++ compiler handles most of it. However, implementing CORBA in a week or two is impossible. It is simply too complex. And in that complexity lies the problem. COM (not DCOM, the distributed version) is simple enough to allow you to use it for something like an API, or even reusable objects like stacks, etc. No way in hell you'd use CORBA for an API (oh wait, isn't that what they're doing in Berlin-land? Another stupid feature-bloat idea.) A CORBA-based stack object? I'd like to see that! In GNOME, CORBA is going to be much less usefull then another system would have been. It takes to many resources, and I predict that it will never gain the acceptance that COM has.
Are you kidding? Konqueror is even slower? On NT Explorer (NOT IE, regular Explorer) pops right up. Same thing with Tracker (BeOS.) However, Konqueror takes just as long as IE to load. And when I simply want to move a file, it really isn't worth it. My point (which you seem to miss) is that feature bloat is BAD. There is no real reason to integrate web-browsing and file browsing. There is no reason to draw your interface with XML. If you took regular Windows Explorer, added back and forward buttons, and let the user specify a custom background picture, then 99% of people wouldn't notice that it wasn't IE. Same thing with Mozilla. If you rip out the XML renderer as UI thing and replaced it with a conventional themable UI, 99% of people wouldn't notice the difference except the fact that it was more responsive. The problem with GNOME is that it is subscribing to the same feature bloat as Microsoft is. The difference is that GNOME doesn't have to. MS needs to sell it's software. It needs to put in features to make it so you have to upgrade very two years. GNOME doesn't have to do that. All it has to do is make the best DE possible. Sure it sounds cool on paper that GNOME is so feature rich, you can run objects on remote UNIX servers via CORBA, that the file manager is based on a powerful XML renderer which is also a CORBA component, etc. However, what does that gain you? How many of you have access to high performance remote CORBA servers? How many people are going to write custom XUL files for Mozilla? How many people will notice (or care) that XUL is so much more powerful than a conventional themable UI. All people will notice is that CORBA takes more memory than COM, and that Mozilla's UI is very unresponsive. Rich, powerful components have no place in a core system like GNOME. The simple stuff should be clean, fast, and modular. Take COM/DCOM for example. The standard COM can't handle remote connections. However, install DCOM (and put up with the resulting increase in memory use) and you've got something that come close ot CORBA in power. (People|Companies|Projects|Groups) in GNOME's position really shouldn't program for that 1% of the population that needs the extra features. They should program their app simple and fast for the 99% and make it modular so the other 1% can extend it if they need to. GNOME is currently running in the tens of megs. For a desktop environment that is simply too much. I can see why they do it, of course. Writing simple software isn't sexy. The allure of supporting every possible feature is probably just to great. But is DOES result in less than perfect software.
You know what would be really cool? A graphics chip that accelerated rendering of something like this. Currently, vector graphics aren't really that feasible for use all over the system because of their performance problems. Even MacOS X still uses bitmaps for most elements. Plus, as anyone who has tried software transparency knows, it is really slow for large images. However, graphics accelerators these days have a nearly limitless number of transistors (from the chip designer's point of view) and it shouldn't be too difficult to design chips that would accelerate something like SVG. Additionally, a lot of 3D these days is very similar to what a 2D vector API would need (the rasterization of curves and lines, compositing using blends, etc) and with some more hardware, it should be feasible to put it on a chip. What that would really enable is the large scale use of vector graphics all over the environment. People who have used MacOS X already know how nifty it can look, and a more large-scale implementation is sure to offer many new possibilities. (One of which could be the ability to keep resolution and image size independant so each one can be customized to suit one's tastes.) Any hardware designers out there? How hard exactly WOULD this be?
The problem is that it's UGLY. An interface not only needs to be functional but RESPONSIVE. Even on my 300MHz computer, KFM still crawls compared to Win95's explorer. Sure KFM looks a little better, but it really doesn't matter. Writing bloated code and hoping proc's will get fast enough is something best left to Microsoft. If something can be good looking and functional without taking a huge amount of resources, then it should be programmed that way. Every program should be coded with a correct balance. The current mantra is (add features, speed be damned!) That's what lead to using CORBA in the DE and Gecko in the file manager. They are very feature powerful, but most of those features are really useless. Does anybody actually take advantage of the HTML customizablity of their folders? File previews are nice, but that is very simple to implement (WITHOUT a powerful XML rendering engine like Gecko!) In the end, you can replace Gecko with some nice bitmaps in the corner, and 99% of people won't be able to tell the difference. If you code for that other 1%, then you'll end up with a crappier product. Ever wonder WHY Windows is so bloated? Not so much sloppy coding, but feature bloat. OSS projects don't need feature bloat. They don't HAVE to make people upgrade to new versions because of new (usually unused) feature. GNOME these days is reeking of Microsoft. I mean only MS would go and put a rich, powerful, distrubted framework like CORBA into the DE. I mean how many people NEED their AbiWord component distributed over a world-wide network?
Slashdot Mirror
Sorry, I mean 16bit real-mode DOS process. The person before me had reffered to this as 8bit, and I wasn't paying attention.
Wrong. Win95 doesn't leave DOS in memory. It moves DOS into a protected mode virtual machine. Win16 apps NEVER call into "real" DOS, they call into a DOS that is running in the (seperate) Win16 virtual machine. All of these run in the Win32 virtual machine. The exit to DOS trick isn't a trick, it unloads Windows, puts the sytem into real mode, and moves DOS back into regular memory. This version of DOS that runs when you "exit to DOS" is the REAL DOS. It runs in real-mode. It doesn't run on top of a VM. What MS has decided to do is remove this DOS. You seem to think that Win95 runs on top of DOS. This is a myth. If it really did run on top of DOS, then most system calls would result in a mode switch. That's simply not the case. The only time Win95 makes calls to DOS is to support old real mode drivers. (Like the original Win95 printer drivers.) There is no malacious intent in here. They figured that nobody needed real-mode DOS anymore (since most DOS apps can run in the virtual machine), and booting into real-mode first interfered with driver loading and made the system unstable, so the got rid of it. No malacious intent here, just some common sense. And it WILL interefer with programs like loadlin or the BeOS loader. They depend on rebooting into DOS mode so they can kick DOS out of memory. Just as these programs won't work on NT, they won't work on ME.
I believe that is true for NT. But at least Win 95 used to run more on top of DOS more than use it only as a loader. There are some evidence for that, like that DOS protected mode memory managers (EMM386 or third party like QEMM) are needed for loading Win 95, and I have seen few times Win 95 crash to a dos prompt!
>>>>>>>>>>
Win95 can crash to a DOS prompt, because it is still in there. However, that doesn't imply that it runs on top of DOS. Also, memory managers aren't needed for loading Win95, and infact it is reccomended that they are NOT used because it interferes with Win95 managing its own memory. Of course they can be used, but that only applies to when you've got apps that run in the DOS VM and need the services of the memory manager. For example, I had this old DOS-mode pinball game that needed a memory manager to run. However, my Win95 system has no memory managers in it's autoexec, so I doubt they are running. Win95 really isn't DOS in a pretty face. Its mostly Win 3.1 in a pretty face. DOS is still there, but it isn't used. DOS is a real-mode operating system. Unless you are running really old soft/hardware, then your system is running in protected-mode most of the time. The amount of DOS in Win95 and 98 is the same (nearly none.) There aren't really any tweeks to the kernel in Win98 aside from some to the memory manager. As for command.com, read up on some articles around the release of Win95. (From REAL magazines, like BYTE, not ZDnet.) The Win95 architecture runs application in 3 ways. First, there is the system VM for 32 bit applications. Then there are two other VMs (which are Win32 applications and run in the system VM.) There is one VM for Win16 (Win 3.x) applications, and there is one or more (each DOS app runs in it's own VM) VMs for DOS applications. In every case, the system is running in protected mode.
OTOH while Win 98 still needs EMM386 (I haven't tried QEMM on it) I have always got the comfy BSOD when it crashes so maybe there is less DOS in it.
>>>>>>>>>
I'm running Win98 without EMM386. According to Peter Norten (Norten's Inside the PC, Page 294)
"Because of this, you need not specify EMM386.EXE in your CONFIG.SYS file, and you should not specify HIMEM.SYS-both of these will actually diminish Windows 95's capability to internally manager memory" As for QEMM, it is simply a replacement for the aformentioned EMM386.EXE and should not be used in Win9x.
Whoa, cool off. I was kidding. It's a joke, laugh, ha ha...
The DOS prompt is so woefully underpowered, it doesn't deserve to share the same name as beasts such as the UNIX shells. (This coming from a person who likes Windows better than Linux, so you know its not some anit-MS bull!)
Okay, now I can officially say that about half the posters on this topic (even the +4's and +5's) have no clue what they are talking about. To put it succinctly
/.
1) It's bfs, not BeOSfs, don't EVER forget that;)
2) MS is not removing DOS comptability. It is simply removing the ability to boot into real-mode DOS. This is a feature that most people don't even use anymore. Unless you've got really old hardware drivers that need real-mode DOS (very rare) you're machine is in protected mode most of the time. Unless you use the "Reboot into DOS" option, you're not missing anything.
3) It is not removing the command line. From inside Windows, everything will still be the same. The DOS prompt will run on top of a VM (virtual machine) just like it always has, you can run DOS applications from the VM, just like you always have.
4) MS is not removing 16bit compatibility. 16bit compatbility has nothing to do with DOS support. Win9x/ME is still the "24bit" 16/32bit kernel.
5) Stability will only improve slightly. In previous versions of Windows, Windows booted DOS first, which then booted the Win95 kernel. WinME boots straight into Win95, so any stability problems that take down you're machine at startup or shutdown will be alleviated.
6) This will have no effect on bfs/vfat or UMSDOS. Those don't actually depend on DOS. The only thing that will change is that you can't use loadlin anymore.
7) This is not some marketing gimmick, or some way to supress CLI users, or anything like that. DOS mode conflicted with some stuff, it made the machine very vunerable at boot, and very few apps actually needed "pure" DOS mode anymore (as opposed to the VM DOS mode, just so we're clear) so Microsoft decided to get rid of it. It certainly doesn't deserve a nearly 700-post barrage on
It's not emulation. It is a virtual machine. kind of like like VMWare except limited to DOS.
Wrong. WinME doesn't use WOW. WOW is only for "pure" 32bit versions of NT. 16bit processes are handled by the same "24bit" Win95 kernel. 8 bit processes (just like the 16bit ones) are run in a virtual machine just like they always have. The only thing that has changed is that it is no longer possible to take a Windows95 machine into pure real mode DOS (without VM's or protections running.)
It's called a VM. If you knew how Windows worked, you wouldn't say stupid stuff like that. Command.com (the DOS prompt) is (and always HAS) run under a virtual machine in Windows 95. The machine operates in protected mode, and run a VM to run the DOS prompt. This is different from the straight DOS prompt which is ACTUAL real mode DOS. Windows 95 has NEVER run on top of DOS. The only DOS code (IE non protected mode) is some funky old drivers. The DOS prompt by itself was the last bits of DOS Win95 had. It was real mode DOS in which the entire machine operated in real mode without a VM. They didn't go out of their way to turn it off, they just got rid of it entirely.
A) DOS is gone from WinME. DOS never really was in Win95, it was present, but DOS apps ran in a VM. What Windows ME does is boot straight into protected mode because it helps in booting quickly, and the machine is very vunerable in this state. Also, it gets rid of uneeded code.
B) The legacy 16bit stuff is still there. 8 bit stuff has been gone for a long time. Most Win95 machines rarely switch into real mode unless you're running some funky hardware or software. However, all the 16bit stuff is still there.
C) It doesn't get rid of the CLI. Just like NT, WinME still has a DOS prompt.
D) You call the DOS prompt a CLI?
This is a great product. If you've ever seen the communities that have sprung up over Ti's 8x line, then you'll know what uses these things have been put to. There are people that write custom OSs for TI's calculators (around 8-10 MHz on 32-128 KB of RAM.) People figure out how to make the thing generate sound through the serial port, how to overclock it, how to make it drive a radio, how to write custom assembly for it, all kinds of neat things. It brings a hackerish feeling that you really don't get these days. Programming the TIs is like programming those old computers. You get to program directly to the hardware, you have to come up with tricks to get the thing to do what you want, you have to find nifty uses for the built in hardware. That's real hacking. What really makes them so popular is that they're cheap and flexible. I really hope that HP manages to keep the price of this thing under $150, or else it really won't work. There are already other HPCs on the market, and this product needs to be different. Since it addresses the main problems with TI's series (poor screen resolution, slow screen response, and limited CPU power) this product could really be cool if priced appropriatly.
1) Whether or not the X Extension is trying to compete with Berlin, it is. Simply by their existance, they are competing.
2) What people don't seem to understand is that OSS isn't the magic bullet. Sure there is a near infinate number of developers, but the realities of the market are still present. There is only so much public interest, whether or not a product is Open Source. Public interest is ultimately what defines who wins and who loses. Great projects die because they don't have enough public interest. Say the X extension results in a rendering interface even half as good as Berlin. Now you've got a less superior product. Certianly a more bloated product (because it incurs the bloat of regular X, plus it's own protocol) and one that really is cobled together and "hackish." However, it has backwards comptability, and by that sole trait will win. What happens then? Probably Berlin dies. By stealing Berlin's turf, X leaves Berlin with no reason for existance.
3) Competition isn't the magic bullet either. Competition only works when to products are compatible. Take a look around the software world instead of thinking idealistically. There are a lot of graphics programs competing. Why? Because they are all comptabile with each other. Then, look at Office suites. Notice how much fewer products there are in this space. Why, less comptability. Take a look at operating systems? Notice the fact that there are no competing products. Why? No comptability. When there is no comptability, userbase is everything. Look all over the software world. Everywhere you can see products suceeding soley based on userbase. Wordperfect is a better word processor than Word? Who won? OS/2 is better than Windows. Who won? Why? There is always a winner and a loser, eventually. OSS slows the ending, but doesn't stop it. Ultimately one will win. By introducing competition to an interesting product, then the OSS community is hurting itself. Things would be different if both were competing on the same level, but they aren't. Soley because X has more comptability, it will win. No matter which is the superior product. By supporting one or the other, there are bigger implications than just one project getting more code time. After all, code time is infinate. Interest is not. By drawing interest away from Berlin, the X extension is almost guarenting its demise.
The question of exactly how much devices can merge is moot. There really isn't a limit. It's essentially asking exactly how much people can innovate. As people have learned, there really is NO limit to innovation. Screen size limitations are beaten by things like screens in glasses, or that technology that uses a laser to draw images on the retina, input limitations are overcome by voice control, maybe even direct connections between the brain and the hardware. There really isn't a limit to where you can take convergance. Convergance in the electronic world probably won't stop in a certain place like it has with appliances. With appliances, the limits of the technology are real. A toaster or a fridge will always have to be at least a certian size, you'll always need the same amount of energy to toast something, you'll never be able to fit a blender into a fridge. That's what prevents convergance in most other household objects from continuing. However, there are no such limitations in electronic convergance, becase all electronics are simply mediums for information. There are no physical limitations on something that carries information. For example, when you buy that brand new spanking DVD player, what do you buy it for? You certainly don't want the hardware, you want to play a DVD. You don't even want to play the DVD, you want to watch the movie. Hell, you don't even want to watch the movie, you want to stimulate you retina's into thinking you're watching a movie, etc. Because the actual "product" in the case of electronics is simply information, ocnvergance can continue until there is no medium left between the information and the consumer.
But wouldn't it be better to go help Bernlin Consortium instead of try to put lipstick on a pig?
I didn't consider it a troll spam. This guy had a reasonable point on the surface. Why does /. say one thing about a company, but through it's advertisements (which imply a sponsering of the product) say something else?
Ah, but here is the beauty of it. At least you know that /. isn't being influenced by it's advertisers. I mean you see this on MaximumPC all the time. People post and advertisement for some product, and on the very next page is a review that spanks the product to all hell. It's a show of journalistic integrity, not hipocracy. A site's advertisements doesn't (and SHOULDN'T) having anything to do with the site's content.
How 'bout this? Stop being so petty! I know programmers are jaded, but hey, even we appreciate a nifty sounding product. Plus, it helps you remember. X-Box SDK, Java SDK, SDL-SDK. How are you going to stand out in the crowd?
That's a little unfair. If I go to the post about KDE 1.92's release, and ask for a filter to get rid of KDE/Linux stuff, then I'm a -1. Asking to filter out Microsoft stuff, and it's untouched.
The hipocracy's so thick you can cut it with a knife.
Huh? What exactly does your post have to do with what he said?
This is quite an interesting position for NVIDIA to be in. Observations
1) NVIDIA bases all its chips on the same driver model. This means that all its chips have similar registers/control sets.
2) NVIDIA doesn't release register-level specs.
3) Console developer's are almost required to give out register-level specs. The game developers love the ability to program directly to the metal (as shown by the fact that this Win2K runs apps in Ring0) and they do a much better job of finding uses for the hardware than the console developers do.
Point: If NVIDIA gives out the specs to this thing, then people can get to work writing driveres for non-Linux/Windows OSs. If NVIDIA doesn't give away specs, then they will get trounced by PS2 developers who find clever tricks to make the hardware do stuff it wasn't designed to. (For example, Crash3 on PSX uses transparency. Technically, PSX doesn't DO transparency.) Maybe they'll release the SDK under NDA?
In response to the first comment, I think I got off on a tangent talking about GNOME. My point was that a good interface should be responsive. Engineering all this "neato" features may be more sexy, but in the end, 99% of people won't use these features, but WILL notice that their interface is slower.
As for Microsoft, don't get me wrong. I am a big Microsoft fan, and I think it would be wise for Linux projects to emulate some of their practices. However, the fact remains that Windows by and large is heavily deteriorated by feature bloat. All this "whiz-bang" features that really aren't necessary hurt the environement. However, that is forgivable in Microsoft's case, they have upgrades to sell. However, projects like GNOME don't have that agenda.
As for COM, it is a gift from god. COM and DirectX are two technologies sorely missing from Linux. However, CORBA and COM are two entirely different beasts. Wheras COM is a (relativly) simple API, CORBA is quite a beast. Look at the two APIs. You can actually write something COM-like in a week or two. Most of the stuff is simply a standard and the C++ compiler handles most of it. However, implementing CORBA in a week or two is impossible. It is simply too complex. And in that complexity lies the problem. COM (not DCOM, the distributed version) is simple enough to allow you to use it for something like an API, or even reusable objects like stacks, etc. No way in hell you'd use CORBA for an API (oh wait, isn't that what they're doing in Berlin-land? Another stupid feature-bloat idea.) A CORBA-based stack object? I'd like to see that! In GNOME, CORBA is going to be much less usefull then another system would have been. It takes to many resources, and I predict that it will never gain the acceptance that COM has.
Are you kidding? Konqueror is even slower? On NT Explorer (NOT IE, regular Explorer) pops right up. Same thing with Tracker (BeOS.) However, Konqueror takes just as long as IE to load. And when I simply want to move a file, it really isn't worth it. My point (which you seem to miss) is that feature bloat is BAD. There is no real reason to integrate web-browsing and file browsing. There is no reason to draw your interface with XML. If you took regular Windows Explorer, added back and forward buttons, and let the user specify a custom background picture, then 99% of people wouldn't notice that it wasn't IE. Same thing with Mozilla. If you rip out the XML renderer as UI thing and replaced it with a conventional themable UI, 99% of people wouldn't notice the difference except the fact that it was more responsive. The problem with GNOME is that it is subscribing to the same feature bloat as Microsoft is. The difference is that GNOME doesn't have to. MS needs to sell it's software. It needs to put in features to make it so you have to upgrade very two years. GNOME doesn't have to do that. All it has to do is make the best DE possible. Sure it sounds cool on paper that GNOME is so feature rich, you can run objects on remote UNIX servers via CORBA, that the file manager is based on a powerful XML renderer which is also a CORBA component, etc. However, what does that gain you? How many of you have access to high performance remote CORBA servers? How many people are going to write custom XUL files for Mozilla? How many people will notice (or care) that XUL is so much more powerful than a conventional themable UI. All people will notice is that CORBA takes more memory than COM, and that Mozilla's UI is very unresponsive. Rich, powerful components have no place in a core system like GNOME. The simple stuff should be clean, fast, and modular. Take COM/DCOM for example. The standard COM can't handle remote connections. However, install DCOM (and put up with the resulting increase in memory use) and you've got something that come close ot CORBA in power. (People|Companies|Projects|Groups) in GNOME's position really shouldn't program for that 1% of the population that needs the extra features. They should program their app simple and fast for the 99% and make it modular so the other 1% can extend it if they need to. GNOME is currently running in the tens of megs. For a desktop environment that is simply too much. I can see why they do it, of course. Writing simple software isn't sexy. The allure of supporting every possible feature is probably just to great. But is DOES result in less than perfect software.
You know what would be really cool? A graphics chip that accelerated rendering of something like this. Currently, vector graphics aren't really that feasible for use all over the system because of their performance problems. Even MacOS X still uses bitmaps for most elements. Plus, as anyone who has tried software transparency knows, it is really slow for large images. However, graphics accelerators these days have a nearly limitless number of transistors (from the chip designer's point of view) and it shouldn't be too difficult to design chips that would accelerate something like SVG. Additionally, a lot of 3D these days is very similar to what a 2D vector API would need (the rasterization of curves and lines, compositing using blends, etc) and with some more hardware, it should be feasible to put it on a chip. What that would really enable is the large scale use of vector graphics all over the environment. People who have used MacOS X already know how nifty it can look, and a more large-scale implementation is sure to offer many new possibilities. (One of which could be the ability to keep resolution and image size independant so each one can be customized to suit one's tastes.) Any hardware designers out there? How hard exactly WOULD this be?
Why doesn't /. ever get /.ed? I mean if they can take Apple's site out, that's some serious hardware /. is running on!
The problem is that it's UGLY. An interface not only needs to be functional but RESPONSIVE. Even on my 300MHz computer, KFM still crawls compared to Win95's explorer. Sure KFM looks a little better, but it really doesn't matter. Writing bloated code and hoping proc's will get fast enough is something best left to Microsoft. If something can be good looking and functional without taking a huge amount of resources, then it should be programmed that way. Every program should be coded with a correct balance. The current mantra is (add features, speed be damned!) That's what lead to using CORBA in the DE and Gecko in the file manager. They are very feature powerful, but most of those features are really useless. Does anybody actually take advantage of the HTML customizablity of their folders? File previews are nice, but that is very simple to implement (WITHOUT a powerful XML rendering engine like Gecko!) In the end, you can replace Gecko with some nice bitmaps in the corner, and 99% of people won't be able to tell the difference. If you code for that other 1%, then you'll end up with a crappier product. Ever wonder WHY Windows is so bloated? Not so much sloppy coding, but feature bloat. OSS projects don't need feature bloat. They don't HAVE to make people upgrade to new versions because of new (usually unused) feature. GNOME these days is reeking of Microsoft. I mean only MS would go and put a rich, powerful, distrubted framework like CORBA into the DE. I mean how many people NEED their AbiWord component distributed over a world-wide network?
Yea, I had bought Opera for Windows. As for the sites I visit I mostly go to bebits, betips, benews, and Be's developer website.