Unisys used to have that moto about the LZW patent until it came close to expiring, when suddenly free LZW programs instantly required that $5,000 registration to use LZW output on a website. Sure, they might be playing the part of the good guy now, but who's to tell 5, 7 or 10 years down the road what they'll start doing.
These days, if it says "patented" or "patent pending", and it's not hardware, I just walk away from it. I'm not about to pad the pockets of those who enjoy patenting trivial software technology.
I can completely agree with you on this. Too often a dirty shutdown resulted in a "corrupt and unusable" file system that broke NT in strange and unusual ways. You'd end up with something that worked right 99% of the time any other time would simply "stop" working after an power fault. I have an Exchange server at work that hasn't worked right ever since a circuit breaker blew. This is by far my biggest complaint about NT... There's just too many unfixable problems, things that almost work, etc that can only be fixed by reinstalling (!!). I don't know about you, but on a server, this is simply not acceptable.
On a workstation, however, I have come to really appreciate NT, as it allows users to run all the Win32 programs they're familiar with, and gives them the desktop they're used to, while I can lock them out of making drastic changes to their workstation, thereby reducing my workload of remastering workstations (a time consuming process that consumes a minimum of 3 hours of my or my assistant's time).
I really believe you should choose whatever OS suites the task. For file servers, that's almost always Netware. Application servers are usually served best by NT. Internet related tasks are often more suited for an UNIX-like server. Hopefully this NDS announcement will help administrators tie all these solutions into a single framework, as my job as it currently stands requires 3 account creations for every single user in the organization, along with dealing with all the quirks with each of those admin tools. I have to admit that I really like Netware's administration tools -- they collect all the individual items of a user's account into one centralized tool, as well as information about all the servers, printer and volumes, as well as other network related information.
Unfortunately I've met a number of MCSE's that make the term incompetent seem inadequate. There's obviously something very wrong with the MCSE if people who can barely operate a computer properly can get a certification that says they're capable of planning, building and administering a network. I'm not saying that all MCSEs are this way, however there's a fairly large proportion of them that are.
NDS is more of a user/resource management system. It lets you organize users, printers, and servers into logical trees in any way your heart desires. You can build complex trees representing the organization the tree is for, and easily manage the users by moving them between locations and having permissions, login scripts, etc change according to their location. It's also available on a number of platforms, such as Windows NT, Netware (of course), and Sun Solaris. If you have a mixed network environment consisting of many different kinds of servers, it can simplify things greatly. Single sign-on is just a side effect of NDS deployment on all servers.
Honestly, there's very little in Word95/97/2000 that I couldn't have done way back in the Wordperfect for DOS 5.1 era. All Microsoft's done is make it easier to learn, and easier to get frustrated with when Word decides it knows what's best for you, and that what you're attempting to do is wrong. I know customers are screaming that they want products that do what they MEAN, not what they SAY, but when the product refuses to listen and you want a SPECIFIC thing to happen, it gets frustrating quickly.
You also have to realize that the NT kernel is a microkernel, which means NTOSKRNL doesn't do much of anything aside from memory management, whereas Linux uses a monolithic kernel. So, yes, you could say Win2000 has fewer bugs in it's "kernel" than Linux does.
On the other hands, if you mean kernel services, a lot of things run as a kernel process under Win2000/NT. The TCPIP system has had numerous bugs in NT4. The TCP sequence numbers, for example. It took Microsoft two hot fixes and several months to get it right. The first hotfix for the problem actually made it WORSE, making it even easier to spoof a connection to an NT machine. A number of NT core services run in the privledged Ring 0 on the intel platform for performance reasons, whereas most UNIX daemons are almost always implemented in user space (ring 3) where memory protection can occur. Microsoft's core selling feature has always been the speed at which it's supposed to operate. C2 certification was a joke (who runs a network server with networking capabilities disabled).
Check out Microsoft's support site about all the "known" bugs in Windows 2000. It's frightening. 187 known bugs affecting Windows 2000 to date, and it's not even shipped yet.
When flashing the ROM on a device, it's usually best to have as little going on as possible. One interruption in the flash process can render the device unusable. In DOS, you're pretty much always a single tasking OS, where nothing will interrupt you (especially if you CLI first). DOS also tends to run nearly flawlessly on even defective hardware or hardware with buggy BIOSes, whereas Linux can be much more finicky. DOS doesn't care about devices, memory protection, or the PCI bus. When dealing with a somewhat dangerous operation, such as a BIOS flash, it's best not to take any risks. After all, few people DON'T have a DOS bootdisk or Win9x startup disk laying around somewhere.
http://www.windows2000test.com Smell: A very large foot, immersed in a very large mouth.
Don't forget that Microsoft will claim the stench was caused by electrical storms in the area.
I just want to know what http://255.255.255.255 would smell like. And just think about the prospective market for those Glade "plugins" for IE and Netscape.
I, personally, have no problems with it being loaded on top of DOS. Unlike NT, I can EASILY bring up a Netware server that I installed an incorrect driver on. If it's a disk controller driver, I simply replace it in the C:\NWSERVER directory, and run SERVER.EXE. If it's not, I can start SERVER.EXE -na -ns and skip it's auto startup scripts. It's a very managable system.
When the same sort of thing happens on NT, you might as well format, since it's probably impossible to figure out what went wrong on it. Rescue disks are all but worthless. Windows2000 is a LITTLE better, but not much. There is a "troubleshooting" mode, but unless the system is mostly working, it probably won't help.
And for user administration, I think Netware beats just about everything out there, especially for a GUI. All information about any given user/print queue/volume/server is available through ONE tool -- nwadmin. Moving a user from one organizational unit to another instantly gives them access to all resources assigned to that OU by the administrator. You can set up logically constructed trees to organise all users and departments within the company. NT (even with Active Directory) and most unixes simply can't compare to this.
Don't forget that, from an archetural stand-point, Linux's design is pretty dated too -- monolithic kernel and all. But if it works, it works. At work, the netware 4.11 server we use often lasts a hundred days or so between hardware failures [I really DO need to fix that soon, especially since the Adaptec RAID controller just decided to eat the entire array for no good reason].
JIT is actually very different. A JIT compiler doesn't have to deal with things like a memory map that cannot change, self-modifying code, stack stuffing, hardware interrupts, etc. But this idea is far from new. The macintosh emulator for PC, Executor, used similar techniques. It was based on a dynamic recompiling CPU emulation core that would translate and simplify a series of instructions, cache the resulting instructions, and then execute them. These technique is only efficient when large amounts of code can be executed without interruption and without the requirement of cycle-level precision of timing. For 98% of the time, x86 software doesn't care or need to care about very precise CPU timings (there's too many different types of x86 CPUs out there to make it useful).
As an instruction set, the x86 is pretty bad, however it's easy to code for and easy to optimize for, which are it's biggests strengths. As a mid-layer API for this device, it was probably a good choice -- x86 recompiles well on RISCy machines with lots of registers. PowerPC and others probably wouldn't. They have a lot of registers themselves and are more complex (plus the wide variety of x86 clones means that most people will likely shy away from dangerous instructions, whereas since the PPC is VERY standardized, many software packages could rely on subtle bugs in the silicon of the PPC. Believe me, bugs are the hardest part of any hardware archetecture to emulate.)
ARDI has some fairly interesting whitepapers on their implementation of the 68k instruction set on x86. Keep in mind this is MUCH more difficult to do than the reverse. The 68k CPU has 16 registers, whereas x86 only has 8, etc.
The big problem with C is the same C source file compiled with the same compiler can often produce many wildly different results, and C doesn't solve the problem of hardware accesses, which almost always need to be done in a low-level language. This CPU/software will be beneficial to many companies due to the fact they will be able to reuse existing hardware, drivers and software with this. As long as the prices for the CPU get really low eventually, it could really lower the prices for PDA and hand-held computers. (Imagine playing a 3D accelerated game of Quake III on a hand-held machine)
"Copy sharing" isn't really going to make that big of a difference. For most people, watching a movie on your computer just isn't the same as watching it on a TV set. Whereas your monitor is probably only 15-17" in size, many people easily have 30-40" TV sets. Especially those interested in big A/V setups. Most people who want to "copy share" are more likely to do it by recording a DVD to VHS.
As of right now, getting a "perfect" rip of the MPEG2 data from a DVD is practically worthless given it's storage requirements (I don't know about you, but I often have enough trouble freeing 700MB to install a game, or burn a CD-R, let alone 4GB to rip the MPEG2 stream to disk), plus the fact DVD recordable technology is prohibitively expensive ($40 for media and $5000 for the recorder doesn't exactly make sense for piracy purposes. Gee, $20 for this DVD, or I can spend $45 to rent and pirate the same movie. Tough choice!). And given how paranoid the DVD forum seems to be about piracy of DVD video, they might try to go for artificially inflating the prices of recordable DVD medium.
I think the DVD Forum should start worrying less about DeCSS and start worrying more about raising the quality of released DVDs (some are so bad, the VHS version is better), increasing consumer acceptance of the technology and reducing the cost of the players and movies (DVDs are easily cheaper to produce than VHS cassettes, but usually cost between two and three times more).
You forget that before that Word has just taken a version number leap from version 2.0 to 6.0, because the old DOS version was at 5.0 and people assumed the DOS version Word 5.0 was newer than the Windows version Word for Windows 2.0. Microsoft's been playing the "Version Number Game" for far longer than anyone else. I doubt they'll stop anytime soon, either. Word is probably not more than 6 (major) versions old, and has already had 3 version marking changes made to it (2.0 -> 6.0, 6.0 -> 95, 97 -> 2000).
Uhg... Sickening... Getting a patent and using two freeware/sourceware programs as a reference (FIPS and presizer). From the way US5675769 reads, I'm not sure what exactly they're patenting. From the looks of it, they've covered everything from the partition operation to the error recovery situation to the PartitionMagic GUI. Oh, what a wonderful world this would be if software patents did not exist. How can anyone find any sanity in a patent system that takes 2-3 years to grant a patent that lasts for 17 years in an industry that can drastically change in less than 6 months?
Global warming may be occuring, but no one can say with accuracy if this is a natural phenominon or if it's caused by pollution. There are far too many variables, and most of the predictions are based on information that isnt' more than 200 years old and trying to relate it to 4 billion years of earth's history. The question is not if the world is getting warmer, but rather if we can really do anything about it. Reducing emissions of harmful gasses isn't a bad idea, but no one can say for certain at this point that it'll change anything.
IIS has crushed Netscape's web server platform, and, if you believe what M$ say, is a serious contender for the www server crown.
Not so! Apache is far more popular than IIS is, capturing a good 55% of the httpd server market, leaving IIS at a mere 22% (Netscape is stuck at about 5%). ASP is not nearly as widespread as Microsoft would like you to believe. http://www.netcraft.net has all the stats. Apache has actually had the lead since around 1996, and although usage stats have fluxuated, Apache hasn't lost much ground in the past few months.
You bet they do! When Microsoft SQL server's performance was dismal, NT added scatter/gather I/O in the next service pack. When IIS was serving web pages too slow, NT added ISAPI. Look at kernel32.dll, user32.dll, gdi32.dll, krnl386.exe, user.exe and gdi.exe sometime in a hex editor. It's scary how many "exported" function calls are there that aren't documented by ANYONE. I doubt you'll find docs on SMapLS_IP_EBP_## or Callback## function or perhaps the BEAR285 and BEAR3062 functions, at least not Microsoft documentation. These functions must exist for SOME reason, but what they do is almost anyone's guess.
I get the feeling that even if MS was broken up, that the OS division would eventually grow large enough to gobble up other other divisions. I get the distinct feeling that this is too late to do any good... Microsoft has successfully crippled Borland's hold on the compiler/RDBS market, Corel/Wordperfect Corp's wordprocessor market and Netscape's webbrowser market.
Microsoft hurt Borland by the fact that all expanded APIs for Windows always came out for Visual C++ first, and they left it to Borland to reimplement the interfaces in their own products. After all, why would you want to use Borland C++ 5.0 when you can't use DirectX. Wordperfect was pretty much killed because of buggy releases, and the fact they were slow to market a Win32 version of their wordprocessors. Then there's things like Microsoft SQL server -- when the developers needed a kernel function to improve performance of queries (scatter/gather IO), it magically became part of Service Pack 3. When the Internet suddenly became popular, Microsoft decided it was not permissible to use NT workstation to run a web server, and that you need to buy NT server instead, which just HAPPENS to come with a "free" web server. Regardless of if any of these things they've done are illegal or not, Microsoft has repeatedly killed off supperior products by using these dubious means and replaced them with their own.
There's very little left except niche markets like sys utils, etc. The only real major product that hasn't felt Microsoft's influence is Quicken, that Microsoft couldn't even kill by giving "free" copys of Microsoft Money away with Windows95.
DeCSS has got to be the worst way to do that kind of conversion. First, you need that 4GB of space to copy and decrypt the stream, then you need space to hold the temporary files during compression and the compressed file, too. I have a feeling that most people who would do something like that would hook a DVD player up to a MPEG or other kind of compressor device. I'll bet that decrypting, software decoding the MPEG2 stream, downscaling the frames and recompressing the video stream would be prohibitly time consuming.
Doesn't matter anyway. No matter what kind of protection you employ in these things, the pirates will ALWAYS find a way around and a way to defeat the protection. The discovery of DeCSS gained pirates very little or nothing, because if they want to duplicate the discs, they've always been able to do that without decrypting. The ripping pirates can still rip and encode movies, they don't get anything new or better from this either.
I really don't see what the big deal is. Especially since none of the defendants did the reverse engineering or anything. The only piracy related situation I can see DeCSS being useful in is if you happen to have 4GB of space sitting around and want to "extend" your rental period for a DVD by copying it to your harddrive. Pointless, yes, and it's not like it's going to be perm storage.
CNN rarely bothers to learn both sides of the story before airing it, especially when the DVD forum is so willing to talk about what it's "lost", and there's not really a central organization on the other side of the fence. Besides, the more time they take to research, the more likely one of their competitors will beat them to the story.
Besides, this lawsuit is pointless. The DVD industry won't dry up and cease to exist because of DeCSS, and the DVD forum, no matter how hard they try, cannot make CSS a secret again, now that the cat's out of the bag.
One of the biggest problems with software is the fact these days it seems that it's both patentable and copyrightable. I don't think there's a single industry out there that can acquire both sets of protection for their products. I have no problem with software copyrights. In fact, I think they're generally a good thing. But software patents bug me. What would happen if suddenly books could be patented, and music could be patented, and TV programmes could be patented. It sounds insane, but it's really no stupider than what's already going on.
I suppose things wouldn't be so bad if patent protection was limited to a very short time on software (say three to five years max). Copyright protection's limitations are also a little extreme for software, as well. For an industry that didn't really exist 20 years ago, protection for 50 years seems a little excessive, especially when you consider the fact in most cases the original product probably won't be sold ever again after 5-7 years after the original publication. There might eventually come a time when 20 year life cycles for software make sense, but that doesn't look like it's gonna happen anytime in the foreseeable future -- the personal computer industry is still rapidly growing and changing. Few software companies can claim they made it to where they are without copying from anyone else, and the ones who do claim that, are probably lying.
You can get the source code for the Corel Explorer from Corels FTP site at ftp:// ftp.corel.com/pub/linux/CorelLinux/source/corellin ux-1.0/corel/source/[It's part of the kde-corel-source tarball]. Good luck getting to build on anything but Debian with the rest of the Corel QT/KDE packages, though. I seriously doubt it's distribution-independant. It might even require the patches to the XF86 servers Corel has up there.
Embrace and extend isn't inherently bad. The time it becomes bad is when you don't give any information on how it's "extended", which simply isn't possible with GPL or QPL'ed code. The reason Microsoft's embrace and extend is painful is they extend an existing protocol, then keep all "extentions" proprietary so no one can copy it in their own implementations, or the extensions are so complex, and of little value to anyone else, so no one bothers implementing them, but are REQUIRED for Microsoft apps to interact with.
Besides, I thought Corel changed the library name of their "bugfix/enhanced" version of the KDE and QT libs to prevent serious conflicts.
It would be a really Bad Thing for users to believe these were official versions of QT and KDE, and started blaming Troll Tech or the KDE team for the fact some QT/KDE apps won't run because of the bugfixes. Since the QPL/GPL requires all changes to the code be made available, these bugfixes could quite easily be folded back into the offical QT or KDE if they are deemed acceptable quality.
It actually depends on what your aim is... Quite often it's difficult or impossible to make money off those ""Naive" home users" or developer-types you speak of, as the support costs are often high enough to seriously cut into profits, or there's very little profit to speak of. They really only serve as a way to get your name out. Corporate sales is usually where the money is because they're usually willing to spend more for better support, and if their end impression is positive, willing to buy more things from you.
I somehow get the feeling that once Corel gets the bugs worked out of the GUI tools, they might start selling "server" versions of Corel Linux, perhaps packed with a SSL HTTPd, and the other technologies that can't be free right now (NDS, perhaps). Their NT-like "event viewer" (log viewer) is nice, and the user manager is okay (although I much perfer Netware's user management tool to the NT-style one in Corel Linux -- a tree based system is the way to go, IMO. It makes managing a hundred or more users easier, especially when there's a high number of transient employees who work for only 1-2 months, and then never return, or switch departments many times.)
I think Corel is on the road to trying at getting the Corporate market, and I can't really blame them. In fact, I think they'd be crazy not to do it. If they get "user-friendly" admin tools, they can pitch it to management types as "Look! It's as easy to admin as NT, and much more stable!", and sysadmin, who have to do all the real work, can stick with the CLI tools they know and love. It's sorta like a best-of-both-worlds type scene. Especially if they can pull it off (more so if they can pull it off without the serious bugs their Windows software tends to suffer from).
They're a Canadian-based company, so I guess I should cheer for them! =P
I don't think it's fixed, or has even been updated in quite some time... It also did the same thing on my Rage128GL based system. From what I could see, Corel Linux doesn't much like ISA (anything) cards, and you have to configure them all with the debian module management tools. It also repeatedly tried to install the ne2k-pci.o driver for my D-Link DFE-530TX, which should use a via-rhine.o driver instead. Took me a while to figure out that editing the file that was refering to my PCI NIC as a ne2k wasn't the way to fix it, and instead I had to add an entry to the/etc/pcimap.inf file with the PCI IDs. Although this wasn't too difficult for me, a new user would likely be hopelessly lost. This same pcimap.inf file makes some sweeping generalizations about some manufacturers (such as ATI adaptors not matching the list of IDs should use the XF86_Mach64 driver... Which is incorrect, since Rage128 gag with that driver)
The retail versions come with a little commercial/licensed software, including some Bitstream/Type1 fonts, BRU, Corel Wordperfect, Civ:CTP, OSS sound drivers, etc. For that reason, they the download and purchaseable versions are somewhat different. The differences between the downloadable and standard editions are minor at worst, and for the most part you can download all those components elsewhere for free anyway. The deluxe version has some commercial software that cannot be downloaded, such as BRU and the full version of Corel Wordperfect.
Slashdot Mirror
These days, if it says "patented" or "patent pending", and it's not hardware, I just walk away from it. I'm not about to pad the pockets of those who enjoy patenting trivial software technology.
On a workstation, however, I have come to really appreciate NT, as it allows users to run all the Win32 programs they're familiar with, and gives them the desktop they're used to, while I can lock them out of making drastic changes to their workstation, thereby reducing my workload of remastering workstations (a time consuming process that consumes a minimum of 3 hours of my or my assistant's time).
I really believe you should choose whatever OS suites the task. For file servers, that's almost always Netware. Application servers are usually served best by NT. Internet related tasks are often more suited for an UNIX-like server. Hopefully this NDS announcement will help administrators tie all these solutions into a single framework, as my job as it currently stands requires 3 account creations for every single user in the organization, along with dealing with all the quirks with each of those admin tools. I have to admit that I really like Netware's administration tools -- they collect all the individual items of a user's account into one centralized tool, as well as information about all the servers, printer and volumes, as well as other network related information.
Unfortunately I've met a number of MCSE's that make the term incompetent seem inadequate. There's obviously something very wrong with the MCSE if people who can barely operate a computer properly can get a certification that says they're capable of planning, building and administering a network. I'm not saying that all MCSEs are this way, however there's a fairly large proportion of them that are.
NDS is more of a user/resource management system. It lets you organize users, printers, and servers into logical trees in any way your heart desires. You can build complex trees representing the organization the tree is for, and easily manage the users by moving them between locations and having permissions, login scripts, etc change according to their location. It's also available on a number of platforms, such as Windows NT, Netware (of course), and Sun Solaris. If you have a mixed network environment consisting of many different kinds of servers, it can simplify things greatly. Single sign-on is just a side effect of NDS deployment on all servers.
Honestly, there's very little in Word95/97/2000 that I couldn't have done way back in the Wordperfect for DOS 5.1 era. All Microsoft's done is make it easier to learn, and easier to get frustrated with when Word decides it knows what's best for you, and that what you're attempting to do is wrong. I know customers are screaming that they want products that do what they MEAN, not what they SAY, but when the product refuses to listen and you want a SPECIFIC thing to happen, it gets frustrating quickly.
On the other hands, if you mean kernel services, a lot of things run as a kernel process under Win2000/NT. The TCPIP system has had numerous bugs in NT4. The TCP sequence numbers, for example. It took Microsoft two hot fixes and several months to get it right. The first hotfix for the problem actually made it WORSE, making it even easier to spoof a connection to an NT machine. A number of NT core services run in the privledged Ring 0 on the intel platform for performance reasons, whereas most UNIX daemons are almost always implemented in user space (ring 3) where memory protection can occur. Microsoft's core selling feature has always been the speed at which it's supposed to operate. C2 certification was a joke (who runs a network server with networking capabilities disabled).
Check out Microsoft's support site about all the "known" bugs in Windows 2000. It's frightening. 187 known bugs affecting Windows 2000 to date, and it's not even shipped yet.
When flashing the ROM on a device, it's usually best to have as little going on as possible. One interruption in the flash process can render the device unusable. In DOS, you're pretty much always a single tasking OS, where nothing will interrupt you (especially if you CLI first). DOS also tends to run nearly flawlessly on even defective hardware or hardware with buggy BIOSes, whereas Linux can be much more finicky. DOS doesn't care about devices, memory protection, or the PCI bus. When dealing with a somewhat dangerous operation, such as a BIOS flash, it's best not to take any risks. After all, few people DON'T have a DOS bootdisk or Win9x startup disk laying around somewhere.
Smell: A very large foot, immersed in a very large mouth.
Don't forget that Microsoft will claim the stench was caused by electrical storms in the area.
I just want to know what http://255.255.255.255 would smell like. And just think about the prospective market for those Glade "plugins" for IE and Netscape.
When the same sort of thing happens on NT, you might as well format, since it's probably impossible to figure out what went wrong on it. Rescue disks are all but worthless. Windows2000 is a LITTLE better, but not much. There is a "troubleshooting" mode, but unless the system is mostly working, it probably won't help.
And for user administration, I think Netware beats just about everything out there, especially for a GUI. All information about any given user/print queue/volume/server is available through ONE tool -- nwadmin. Moving a user from one organizational unit to another instantly gives them access to all resources assigned to that OU by the administrator. You can set up logically constructed trees to organise all users and departments within the company. NT (even with Active Directory) and most unixes simply can't compare to this.
Don't forget that, from an archetural stand-point, Linux's design is pretty dated too -- monolithic kernel and all. But if it works, it works. At work, the netware 4.11 server we use often lasts a hundred days or so between hardware failures [I really DO need to fix that soon, especially since the Adaptec RAID controller just decided to eat the entire array for no good reason].
As an instruction set, the x86 is pretty bad, however it's easy to code for and easy to optimize for, which are it's biggests strengths. As a mid-layer API for this device, it was probably a good choice -- x86 recompiles well on RISCy machines with lots of registers. PowerPC and others probably wouldn't. They have a lot of registers themselves and are more complex (plus the wide variety of x86 clones means that most people will likely shy away from dangerous instructions, whereas since the PPC is VERY standardized, many software packages could rely on subtle bugs in the silicon of the PPC. Believe me, bugs are the hardest part of any hardware archetecture to emulate.)
ARDI has some fairly interesting whitepapers on their implementation of the 68k instruction set on x86. Keep in mind this is MUCH more difficult to do than the reverse. The 68k CPU has 16 registers, whereas x86 only has 8, etc.
The big problem with C is the same C source file compiled with the same compiler can often produce many wildly different results, and C doesn't solve the problem of hardware accesses, which almost always need to be done in a low-level language. This CPU/software will be beneficial to many companies due to the fact they will be able to reuse existing hardware, drivers and software with this. As long as the prices for the CPU get really low eventually, it could really lower the prices for PDA and hand-held computers. (Imagine playing a 3D accelerated game of Quake III on a hand-held machine)
As of right now, getting a "perfect" rip of the MPEG2 data from a DVD is practically worthless given it's storage requirements (I don't know about you, but I often have enough trouble freeing 700MB to install a game, or burn a CD-R, let alone 4GB to rip the MPEG2 stream to disk), plus the fact DVD recordable technology is prohibitively expensive ($40 for media and $5000 for the recorder doesn't exactly make sense for piracy purposes. Gee, $20 for this DVD, or I can spend $45 to rent and pirate the same movie. Tough choice!). And given how paranoid the DVD forum seems to be about piracy of DVD video, they might try to go for artificially inflating the prices of recordable DVD medium.
I think the DVD Forum should start worrying less about DeCSS and start worrying more about raising the quality of released DVDs (some are so bad, the VHS version is better), increasing consumer acceptance of the technology and reducing the cost of the players and movies (DVDs are easily cheaper to produce than VHS cassettes, but usually cost between two and three times more).
You forget that before that Word has just taken a version number leap from version 2.0 to 6.0, because the old DOS version was at 5.0 and people assumed the DOS version Word 5.0 was newer than the Windows version Word for Windows 2.0. Microsoft's been playing the "Version Number Game" for far longer than anyone else. I doubt they'll stop anytime soon, either. Word is probably not more than 6 (major) versions old, and has already had 3 version marking changes made to it (2.0 -> 6.0, 6.0 -> 95, 97 -> 2000).
Uhg... Sickening... Getting a patent and using two freeware/sourceware programs as a reference (FIPS and presizer). From the way US5675769 reads, I'm not sure what exactly they're patenting. From the looks of it, they've covered everything from the partition operation to the error recovery situation to the PartitionMagic GUI. Oh, what a wonderful world this would be if software patents did not exist. How can anyone find any sanity in a patent system that takes 2-3 years to grant a patent that lasts for 17 years in an industry that can drastically change in less than 6 months?
Global warming may be occuring, but no one can say with accuracy if this is a natural phenominon or if it's caused by pollution. There are far too many variables, and most of the predictions are based on information that isnt' more than 200 years old and trying to relate it to 4 billion years of earth's history. The question is not if the world is getting warmer, but rather if we can really do anything about it. Reducing emissions of harmful gasses isn't a bad idea, but no one can say for certain at this point that it'll change anything.
IIS has crushed Netscape's web server platform, and, if you believe what M$ say, is a serious contender for the www server crown.
Not so! Apache is far more popular than IIS is, capturing a good 55% of the httpd server market, leaving IIS at a mere 22% (Netscape is stuck at about 5%). ASP is not nearly as widespread as Microsoft would like you to believe. http://www.netcraft.net has all the stats. Apache has actually had the lead since around 1996, and although usage stats have fluxuated, Apache hasn't lost much ground in the past few months.
You bet they do! When Microsoft SQL server's performance was dismal, NT added scatter/gather I/O in the next service pack. When IIS was serving web pages too slow, NT added ISAPI. Look at kernel32.dll, user32.dll, gdi32.dll, krnl386.exe, user.exe and gdi.exe sometime in a hex editor. It's scary how many "exported" function calls are there that aren't documented by ANYONE. I doubt you'll find docs on SMapLS_IP_EBP_## or Callback## function or perhaps the BEAR285 and BEAR3062 functions, at least not Microsoft documentation. These functions must exist for SOME reason, but what they do is almost anyone's guess.
Microsoft hurt Borland by the fact that all expanded APIs for Windows always came out for Visual C++ first, and they left it to Borland to reimplement the interfaces in their own products. After all, why would you want to use Borland C++ 5.0 when you can't use DirectX. Wordperfect was pretty much killed because of buggy releases, and the fact they were slow to market a Win32 version of their wordprocessors. Then there's things like Microsoft SQL server -- when the developers needed a kernel function to improve performance of queries (scatter/gather IO), it magically became part of Service Pack 3. When the Internet suddenly became popular, Microsoft decided it was not permissible to use NT workstation to run a web server, and that you need to buy NT server instead, which just HAPPENS to come with a "free" web server. Regardless of if any of these things they've done are illegal or not, Microsoft has repeatedly killed off supperior products by using these dubious means and replaced them with their own.
There's very little left except niche markets like sys utils, etc. The only real major product that hasn't felt Microsoft's influence is Quicken, that Microsoft couldn't even kill by giving "free" copys of Microsoft Money away with Windows95.
Doesn't matter anyway. No matter what kind of protection you employ in these things, the pirates will ALWAYS find a way around and a way to defeat the protection. The discovery of DeCSS gained pirates very little or nothing, because if they want to duplicate the discs, they've always been able to do that without decrypting. The ripping pirates can still rip and encode movies, they don't get anything new or better from this either.
I really don't see what the big deal is. Especially since none of the defendants did the reverse engineering or anything. The only piracy related situation I can see DeCSS being useful in is if you happen to have 4GB of space sitting around and want to "extend" your rental period for a DVD by copying it to your harddrive. Pointless, yes, and it's not like it's going to be perm storage.
Besides, this lawsuit is pointless. The DVD industry won't dry up and cease to exist because of DeCSS, and the DVD forum, no matter how hard they try, cannot make CSS a secret again, now that the cat's out of the bag.
One of the biggest problems with software is the fact these days it seems that it's both patentable and copyrightable. I don't think there's a single industry out there that can acquire both sets of protection for their products. I have no problem with software copyrights. In fact, I think they're generally a good thing. But software patents bug me. What would happen if suddenly books could be patented, and music could be patented, and TV programmes could be patented. It sounds insane, but it's really no stupider than what's already going on.
I suppose things wouldn't be so bad if patent protection was limited to a very short time on software (say three to five years max). Copyright protection's limitations are also a little extreme for software, as well. For an industry that didn't really exist 20 years ago, protection for 50 years seems a little excessive, especially when you consider the fact in most cases the original product probably won't be sold ever again after 5-7 years after the original publication. There might eventually come a time when 20 year life cycles for software make sense, but that doesn't look like it's gonna happen anytime in the foreseeable future -- the personal computer industry is still rapidly growing and changing. Few software companies can claim they made it to where they are without copying from anyone else, and the ones who do claim that, are probably lying.
You can get the source code for the Corel Explorer from Corels FTP site at ftp:// ftp.corel.com/pub/linux/CorelLinux/source/corellin ux-1.0/corel/source/[It's part of the kde-corel-source tarball]. Good luck getting to build on anything but Debian with the rest of the Corel QT/KDE packages, though. I seriously doubt it's distribution-independant. It might even require the patches to the XF86 servers Corel has up there.
Embrace and extend isn't inherently bad. The time it becomes bad is when you don't give any information on how it's "extended", which simply isn't possible with GPL or QPL'ed code. The reason Microsoft's embrace and extend is painful is they extend an existing protocol, then keep all "extentions" proprietary so no one can copy it in their own implementations, or the extensions are so complex, and of little value to anyone else, so no one bothers implementing them, but are REQUIRED for Microsoft apps to interact with.
Besides, I thought Corel changed the library name of their "bugfix/enhanced" version of the KDE and QT libs to prevent serious conflicts.
It would be a really Bad Thing for users to believe these were official versions of QT and KDE, and started blaming Troll Tech or the KDE team for the fact some QT/KDE apps won't run because of the bugfixes. Since the QPL/GPL requires all changes to the code be made available, these bugfixes could quite easily be folded back into the offical QT or KDE if they are deemed acceptable quality.
It actually depends on what your aim is... Quite often it's difficult or impossible to make money off those ""Naive" home users" or developer-types you speak of, as the support costs are often high enough to seriously cut into profits, or there's very little profit to speak of. They really only serve as a way to get your name out. Corporate sales is usually where the money is because they're usually willing to spend more for better support, and if their end impression is positive, willing to buy more things from you.
I somehow get the feeling that once Corel gets the bugs worked out of the GUI tools, they might start selling "server" versions of Corel Linux, perhaps packed with a SSL HTTPd, and the other technologies that can't be free right now (NDS, perhaps). Their NT-like "event viewer" (log viewer) is nice, and the user manager is okay (although I much perfer Netware's user management tool to the NT-style one in Corel Linux -- a tree based system is the way to go, IMO. It makes managing a hundred or more users easier, especially when there's a high number of transient employees who work for only 1-2 months, and then never return, or switch departments many times.)
I think Corel is on the road to trying at getting the Corporate market, and I can't really blame them. In fact, I think they'd be crazy not to do it. If they get "user-friendly" admin tools, they can pitch it to management types as "Look! It's as easy to admin as NT, and much more stable!", and sysadmin, who have to do all the real work, can stick with the CLI tools they know and love. It's sorta like a best-of-both-worlds type scene. Especially if they can pull it off (more so if they can pull it off without the serious bugs their Windows software tends to suffer from).
They're a Canadian-based company, so I guess I should cheer for them! =P
I don't think it's fixed, or has even been updated in quite some time... It also did the same thing on my Rage128GL based system. From what I could see, Corel Linux doesn't much like ISA (anything) cards, and you have to configure them all with the debian module management tools. It also repeatedly tried to install the ne2k-pci.o driver for my D-Link DFE-530TX, which should use a via-rhine.o driver instead. Took me a while to figure out that editing the file that was refering to my PCI NIC as a ne2k wasn't the way to fix it, and instead I had to add an entry to the /etc/pcimap.inf file with the PCI IDs. Although this wasn't too difficult for me, a new user would likely be hopelessly lost. This same pcimap.inf file makes some sweeping generalizations about some manufacturers (such as ATI adaptors not matching the list of IDs should use the XF86_Mach64 driver... Which is incorrect, since Rage128 gag with that driver)
The retail versions come with a little commercial/licensed software, including some Bitstream/Type1 fonts, BRU, Corel Wordperfect, Civ:CTP, OSS sound drivers, etc. For that reason, they the download and purchaseable versions are somewhat different. The differences between the downloadable and standard editions are minor at worst, and for the most part you can download all those components elsewhere for free anyway. The deluxe version has some commercial software that cannot be downloaded, such as BRU and the full version of Corel Wordperfect.