And you're ignoring the fact that.NET doesn't magically make the object model compatible -- it basically forces apps to use the C# object model. The ars article gets fairly caught up in the marketing that Microsoft has put out on.NET, without addressing the limitations inherent in MSIL. For the other side of the coin, you should read http://www.javalobby.org/clr.html, which is the perspective on.NET's components marketing claims from the perspective of a Java engineer. It is, of course, biased, but he does have good technical arguments.
Re:Why MS Chose to create .NET for FreeBSD
on
What is .NET?
·
· Score: 1
Once the source is provided for FreeBSD, anybody can take it and port it to their favorite Unix-like OS, such as, Linux or Solaris.
Not if it's "shared source". MS's "Shared Source" licenses are a look-but-don't-touch style license. You can't do anything with it, and you may not even be allowed to compile it. Of course, Microsoft could still shock people and release it under the BSD/X11 license. I doubt it'll happen, though.
I mean accessing a pgsql database features without a backend pgsql database server. Unless I'm completely misunderstanding, ecpg is simply a pre-processor to make using SQL queries to a backend SQL database easier. I'm talking about something closer to Sleepycat's Berkeley DB or the new features of MySQL 4.0, where the application actually calls the library (which is just the stripped down version of the server) to create the database, create the tables, and so on.
There are good reasons for putting data in a conventional server-based SQL database, but there's also good reasons not to. There are a number of cases where another application (and even the database server itself) has no business accessing or potentially modifying the data. It's for things like this that I would like to have such an embeded version. For example, if I'm making a mail server, or something, perhaps bad things can happen if the data is not updated in a particular way, I don't want just any old app making changes, obviously. It also helps reduce namespace conflicts between programs, and so on.
I really wish there was an embedable version of PostgreSQL... It's a very good database, but it's sometimes a real pain to write a program that ties together a SQL database with anything else, unless it's a local-use only program. I know MySQL added this feature in 4.x (but their transaction support is too new, IMO).
I work at a computer store, and one of the locations' service department in the city used to take a hard drive with bad sectors anyway (and off-of-warranty), and open it up, and leave it running. Every week or something, they'd run scandisk on the drive, and have a pool on how many bad sectors they'd find. It seems that a drive has to be very badly damaged for the drive circuitry to completely give up on reading from the disk
One single case which is likely to cause an avalanche of gold rush "victims" seeking to sue the company to get compensation for their "pain and misery". The same sort of thing that happened with Carpal Tunnel Syndrome and keyboards/mice. Such a litagious and corrupt society we live in....
The DVD region system was a good idea, but it's poorly implemented. It's supposed to allow cheap DVDs to be sold in places like India without affecting the market in the US and Europe. Without it, DVDs would probably never be released in India at all, or they'd be released there at the same price as in the US and the middle class wouldn't be able to afford them.
No, the entire purpose of region encoding is the fact that the studios don't want to have to do world-wide releases of movies when they hit the theatres, because the movies are often released on DVD in North America at nearly the same time as the theatrical release of them overseas. It can be months after the premiere showing in North America, before they start showing it in other countries. Without region coding, they were woried that they would end up competing against themselves, and they would lose the maximum profit potential. They don't really care about cheap DVDs anywhere (besides, from all reports, region 1 [North American] DVDs are consistently cheaper than any other region).
It's all about the maximization of profits and nothing else. It's just a sad thing that the movie (and music) studios don't understand that if they'd just reduce the price on these things a bit, they would stand to sell better. Instead they want to squeeze every penny out of their customers.
In terms of power and computation, the Atari 2600 was far less powerful than the Z80. Far closer would be the original Nintendo Entertainment System (which used a 6502) or Sega Master System (used Z80). The Gameboy has also used it since it was introduced (the GBA is the first to break tradition and use an ARM CPU instead).
Every generation has it's luddites, and somehow they always seem to make it into government. I've seen Canadian MPs shouting about the evils of the Internet, I've seen American senators do the same, along with damn near every other country that is similarily well developed. It just never ceases to amaze me how these luddites end up always getting elected and opening their big mouths about things they know little or nothing about, always invoking "for the CHILDREN" as the reason for their insanity.
This type of "optimization" would be best acchieved by having the installer detect the host CPU and install the correct kernel package. There's really only a few distinct classes of CPUs that Linux supports for x86 -- 386, 486, Pentium (and clones), PPro/PII/Celeron, PIII, and so on. Having about 7-8 kernel packages in a distribution wouldn't be too difficult. I believe RedHat (and others) do exactly this already.
He probably means PCI/AGP -- they're the only ones that would be messed up enough to change their vendorid/productid every boot. Kudzu did "Plug-and-pray" for both ISA PnP and PCI devices.
No, they don't give all specs to XFree86. They have withheld information about their cards in the past, and continue to do so today. I believe they still do not provide information to developers on the iDCT and motion compensation in the Rage128 and higher cards. While this is far less problematic than the information that nvidia is withholding on their cards, it's still far from being able to use the card to it's fullest (some future version of XFree86 will feature XvMC - X Video Motion Compensation). Both vendors are willing to give up anything they think will not help there competition (for ATI, everything but the video bits, for nvidia, everything but the memory bits), but fiercely guard everything else.
I would assume that enterprise agreements mean volume licenses. If you only had one developer that needed the produce (or even two), chances are you'd just buy the boxed version. After a certain number of copies (usually two to three), it becomes much more economical to use the volume licensing, at which point you'd be subject to these type of provisions for just about any type of licensed software you might buy (Microsoft's stuff comes to mind, as they have similar provisions).
Well, yeah...IIS3 SUCKED. It had the worst interface ever. 4 wasn't bad. 5 actually got things right from a GUI POV. Every version only has gotten better.
I still don't care for any of these. Although that might have more to do with the fact that everytime I install a hotfix or service pack for Windows or Exchange server, it resets some of the IIS settings and messes up a working configuration. Win2K SP1 and SP2, and Exchange 2K SP1 both insist on turning on "Integrated Windows Authentication" on all exchange web services whenever you install them. Unfortunately, SSL and NTLMSSP auth don't mix well, and every single time I apply these, Outlook Webaccess completely stops working, until I remember that the stupid service pack "fixed" my configuration, and I need to unfix it back.
There is nothing I hate more than software that second guesses my decisions without asking me (if there would be a legitimate reason I would want to change that setting at this time -- asking me because it thinks I made a mistake is equally frustrating).
Part of the problem is that the mere fact that such functionality exists is nearly completely hidden from casual view. The other problem is the scripting stuff is all located on MSDN, which says to many people, that this stuff is only for developers. The third problem I have with all this is some of the interfaces are just too difficult to find documentation on, or implement every concievable function you could ever want, EXCEPT the one you urgently need.
I believe you will need to use the Kerberos PAM modules as well for Active Directory. Microsoft's MSDN also has source code for adding/removing users from AD, and changing passwords via Kerberos. I've never tried it myself, but it's supposed to work.
I think you just hit on the difference between what random data means in the computer research arena, and what it means in press releases. Whereas we would take random data to mean something that is unpredictable, and non-repeating, the press release language may just mean random file types.
Macrovision is an analogue copy protection technology and has nothing to do with the Digital Millenium Copyright Act. Technology to defeat it has existed long before the DMCA, and has significant non-infringing uses.
I think a distinction to be made here is that in the Macrovision case, the copy-protection scheme predated the hardware to beat it, so that it could legitimately be argued that the hardware was designed specifically to defeat Macrovision copy protection.
I doubt that. The reason Macrovision worked was because they mess with the H/V synch of the signal. The modification that prevented Macrovision from corrupting the copy would've also helped recording from poor input sources. Take video stabilizers, for instance, which have a secondary effect of erasing Macrovision corruption while also cleaning up a video signal and preventing the loss of synch.
Will this end up like the VHS market where VHS recorders started intentionally mis-recording Macrovision protected content, despite the fact they had fixed the original flaw that allowed macrovision copy protection to work? Or will the DVD drive manufacturers stand up to the recording industry?
If perl hurts your eyeballs, you don't understand it.
Perl makes it easy to write a quick hack that looks ugly but fixes a real problem. The hurting eyeballs factor only comes in when that quick hack gets progressively promoted to a full mission-critical program with little or no planning. There are well written, structured perl apps, and there's the unstructured ugly ones. Most people have seen more of the ugly ones.
You're assuming the Microsoft assertion that anytime you use any GPL code in any project, you must release all code that uses that GPLed code, which is completely untrue. There's only one real time when you need to release -- when you link the code of the GPLed product to your own. Calling external GPLed programs is perfectly legitimate, as is running proprietary programs on a GPLed kernel. The only time you need to release changes would be if you modified either the GPLed kernel or one of the GPLed programs you were using.
Quite frankly this argument is getting old. If you copy someone else's code and don't abide by their licensing terms, you're no different from the person who makes unlicensed copies of proprietary software. Most Linux libraries that provide common functionality are LGPLed (which is allowed to be linked against a proprietary binary program) or can be licensed from their author under different terms for a sum of money. The GPL really is no more viral than a Microsoft EULA -- you either agree to be bound by it's terms or you cease the infringing activity.
Slashdot Mirror
And you're ignoring the fact that .NET doesn't magically make the object model compatible -- it basically forces apps to use the C# object model. The ars article gets fairly caught up in the marketing that Microsoft has put out on .NET, without addressing the limitations inherent in MSIL. For the other side of the coin, you should read http://www.javalobby.org/clr.html, which is the perspective on .NET's components marketing claims from the perspective of a Java engineer. It is, of course, biased, but he does have good technical arguments.
There are good reasons for putting data in a conventional server-based SQL database, but there's also good reasons not to. There are a number of cases where another application (and even the database server itself) has no business accessing or potentially modifying the data. It's for things like this that I would like to have such an embeded version. For example, if I'm making a mail server, or something, perhaps bad things can happen if the data is not updated in a particular way, I don't want just any old app making changes, obviously. It also helps reduce namespace conflicts between programs, and so on.
I really wish there was an embedable version of PostgreSQL... It's a very good database, but it's sometimes a real pain to write a program that ties together a SQL database with anything else, unless it's a local-use only program. I know MySQL added this feature in 4.x (but their transaction support is too new, IMO).
I work at a computer store, and one of the locations' service department in the city used to take a hard drive with bad sectors anyway (and off-of-warranty), and open it up, and leave it running. Every week or something, they'd run scandisk on the drive, and have a pool on how many bad sectors they'd find. It seems that a drive has to be very badly damaged for the drive circuitry to completely give up on reading from the disk
One single case which is likely to cause an avalanche of gold rush "victims" seeking to sue the company to get compensation for their "pain and misery". The same sort of thing that happened with Carpal Tunnel Syndrome and keyboards/mice. Such a litagious and corrupt society we live in....
It's all about the maximization of profits and nothing else. It's just a sad thing that the movie (and music) studios don't understand that if they'd just reduce the price on these things a bit, they would stand to sell better. Instead they want to squeeze every penny out of their customers.
yuo dotn spel reel gud, do yuo? Danm spel chek!
In terms of power and computation, the Atari 2600 was far less powerful than the Z80. Far closer would be the original Nintendo Entertainment System (which used a 6502) or Sega Master System (used Z80). The Gameboy has also used it since it was introduced (the GBA is the first to break tradition and use an ARM CPU instead).
Every generation has it's luddites, and somehow they always seem to make it into government. I've seen Canadian MPs shouting about the evils of the Internet, I've seen American senators do the same, along with damn near every other country that is similarily well developed. It just never ceases to amaze me how these luddites end up always getting elected and opening their big mouths about things they know little or nothing about, always invoking "for the CHILDREN" as the reason for their insanity.
This type of "optimization" would be best acchieved by having the installer detect the host CPU and install the correct kernel package. There's really only a few distinct classes of CPUs that Linux supports for x86 -- 386, 486, Pentium (and clones), PPro/PII/Celeron, PIII, and so on. Having about 7-8 kernel packages in a distribution wouldn't be too difficult. I believe RedHat (and others) do exactly this already.
He probably means PCI/AGP -- they're the only ones that would be messed up enough to change their vendorid/productid every boot. Kudzu did "Plug-and-pray" for both ISA PnP and PCI devices.
No, they don't give all specs to XFree86. They have withheld information about their cards in the past, and continue to do so today. I believe they still do not provide information to developers on the iDCT and motion compensation in the Rage128 and higher cards. While this is far less problematic than the information that nvidia is withholding on their cards, it's still far from being able to use the card to it's fullest (some future version of XFree86 will feature XvMC - X Video Motion Compensation). Both vendors are willing to give up anything they think will not help there competition (for ATI, everything but the video bits, for nvidia, everything but the memory bits), but fiercely guard everything else.
I would assume that enterprise agreements mean volume licenses. If you only had one developer that needed the produce (or even two), chances are you'd just buy the boxed version. After a certain number of copies (usually two to three), it becomes much more economical to use the volume licensing, at which point you'd be subject to these type of provisions for just about any type of licensed software you might buy (Microsoft's stuff comes to mind, as they have similar provisions).
There is nothing I hate more than software that second guesses my decisions without asking me (if there would be a legitimate reason I would want to change that setting at this time -- asking me because it thinks I made a mistake is equally frustrating).
Part of the problem is that the mere fact that such functionality exists is nearly completely hidden from casual view. The other problem is the scripting stuff is all located on MSDN, which says to many people, that this stuff is only for developers. The third problem I have with all this is some of the interfaces are just too difficult to find documentation on, or implement every concievable function you could ever want, EXCEPT the one you urgently need.
I believe you will need to use the Kerberos PAM modules as well for Active Directory. Microsoft's MSDN also has source code for adding/removing users from AD, and changing passwords via Kerberos. I've never tried it myself, but it's supposed to work.
I think you just hit on the difference between what random data means in the computer research arena, and what it means in press releases. Whereas we would take random data to mean something that is unpredictable, and non-repeating, the press release language may just mean random file types.
Gah. I suppose I should've known better than assume logic in lawmakers decisions.
Macrovision is an analogue copy protection technology and has nothing to do with the Digital Millenium Copyright Act. Technology to defeat it has existed long before the DMCA, and has significant non-infringing uses.
Will this end up like the VHS market where VHS recorders started intentionally mis-recording Macrovision protected content, despite the fact they had fixed the original flaw that allowed macrovision copy protection to work? Or will the DVD drive manufacturers stand up to the recording industry?
Perl makes it easy to write a quick hack that looks ugly but fixes a real problem. The hurting eyeballs factor only comes in when that quick hack gets progressively promoted to a full mission-critical program with little or no planning. There are well written, structured perl apps, and there's the unstructured ugly ones. Most people have seen more of the ugly ones.
Quite frankly this argument is getting old. If you copy someone else's code and don't abide by their licensing terms, you're no different from the person who makes unlicensed copies of proprietary software. Most Linux libraries that provide common functionality are LGPLed (which is allowed to be linked against a proprietary binary program) or can be licensed from their author under different terms for a sum of money. The GPL really is no more viral than a Microsoft EULA -- you either agree to be bound by it's terms or you cease the infringing activity.