This is a battle royale over distribution of streaming media. In this case, Microsoft's format is already open.
Microsoft's format is already open? All I've ever seen is MS offering to license the technology (for royalties, of course) to manufacturers of portable music players. In fact, if the format was open, it'd probably undermine Microsoft's ability to sell the "Digital Rights Management" to studios. The end result is they need to kill both RealPlayer and WinAMP to take over as the dominant media player, and making the format of "Windows Media Audio" an open format would allow their competitors to support this just as easily as Microsoft, so all they open is the APIs to their proprietary technology. If you want to playback on a supported version of Windows or MacOS, fine, but everything else is offlimits.
No, people out in the maritime provinces say things like "arfter" (after) and "ruff" (roof) and just generally use too many r's. People on the prairies tend to mispronounce "bury" (e.g. rhyming with jury instead of berry). I'm not sure where "aboot" comes from. All I know is that everytime I hear someone in Manitoba say "aboot", I feel the need to hit them very, very hard.
The problem is that companies like Tivo have PATENTS on this kind of technology, and Open Source companies (or individuals) can't usually afford to pay for an unrestricted license that would allow them to distribute such a work. While someone such as myself, who lives in Canada, a country that doesn't recognise software patents, a large number of people who live in countries like the United States will not have this option. Such a patent would make even an end user that uses such a program a criminal, regardless of if they knew they were breaking the law by using the program or not.
I guess you don't bother to follow Microsoft's list of "Best Practices", then. Because of the fact you need to reduce the security level to run IIS on a machine that's either a BDC or PDC, your webserver can become quite a risk to the security of the rest of your network. MS recommends that webservers be run on a standalone server so that it can take advantage of the "local" accounts on that machine, rather than relying on a domain user.
As for having Windows 2000 share that much, you probably don't want to set it up that way. The way MS SQL Server was designed, it will take every last byte of memory and can be a real hog for CPU cycles if it has work it believes it has to do.
It may or may not be developed by Microsoft, but they contract out production to other companies. Many of the SideWinder joypads, for example, are made by Mitsumi (according to the circuit board) with Microsoft's name the visible outside name. In fact, I seriously doubt that Microsoft will build X-Boxes themselves, either -- that, too, will probably fall to one of these other hardware companies.
I didn't say they weren't enforcing it. But up until very recently, if your program was free and generated LZW compressed GIFs, Unisys had no problems with that (you didn't have to pay royalties). Although they didn't rescind on that completely, if you want to use LZW compressed GIFs from such a "non-licensed" program, Unisys is being an ass and enforcing licensing for that. That action, however, IS recent.
Yet somehow that reminds me of Unisys's stance on their LZW compression patent up until very recently when they discovered they could extort more money from people by not having such a lax restriction. "You want GIFs on your site produced by a non-licensed generator? Pay us $5,000 now or face legal action!". They're not making any promises that they won't go after royalties from free distribution places at a later date, so I still wouldn't trust them.
Windows 2000 also just installs SLOWLY. Compared to NT 4.0, where you'd configure partitioning, it'd spend 15-20 minutes copying files (coffee break), come back, answer more questions, and the final stage took about 15-20 seconds to copy the rest of the files, then you reboot. Windows 2000 is very different. You have to answer the partitioning questions, then waste 5-10 minutes waiting for it to format, only to come back later to answer a dozen more questions, and wait another half hour for it to copy and configure all the installed software. And if you're setting up a 2000 server with active directory, it gets worse as you have to wait for it to reboot again, and then run adpromo.exe to make it a domain controller, and wait for that to install.
Yeah, except you still reboot after all of them. Realistically, there was little that prevented them from doing this before, besides a self-placed restriction in some of the hot fixes that require that nothing else have been installed without a reboot before installing it. qchain is mostly only useful for patching newly installed machines. Rarely is it useful on machines that are already in production. I do find it rather unusual that with Windows NT 4.0 and NTFS you could replace files that are in use by renaming the file that's being used to some new name, and then creating a new one in it's place. Windows 2000 removed this "feature", forcing you into recovery console mode if you want to do the same thing.
I would however debate if Windows is now stable enough for mission-critical servers -- my experience with Windows 2000 and Active Directory has been that although it doesn't crash, it doesn't act or react consistently, and strange things can go wrong with little apparent reason or fix. Just as an example, take the group policy feature in Windows 2000. If there's any rights granted to specific users and one or more of those users are deleted, the group policies will stop being applied properly.
Maybe Windows 2004 (whatever MS decides to name that) will be ready, but 2000 definately isn't, and XP isn't likely to be ready for completely mission critical use either (in cases like this, redundancy doesn't fix anything, because if one server has the problem, the other will too). Windows 2000 (and probably XP) are much better as workstation OS's than NT4.0 or Win9x, and if you need to use Windows, Win2000 is probably your best bet. There's still rough edges on Windows 2000 that are nearly completely unexplainable or produce obscure error messages (even the much maligned UNIX error message, "printer is on fire", gives more accurate information on what's wrong to an administrator than some of the eventlog messages that Windows 2000 can generate).
I think the GUI would be one of the last reasons why you'd need two windows machines to fill the role of 1 UNIX-ish machine. Lets take webserving for instance. With Microsoft IIS (and most other Windows-based webservers), to apply a security fix, you must reboot the entire server for the change to take effect -- this can be upwards of 15 minutes per boot depending on the system, and hardware. With Apache on any type of UNIX, you can simply install the new version, and only the service needs to be restarted, which means a downtime of about 5 seconds for small sites, and sometimes 60 seconds for very large virtual hosting sites. Indeed with Linux, the only thing you really need to do a full reboot for is a kernel change or upgrade (and most kernel upgrades aren't really required for security reasons). Many of Microsoft's other server products are the same. Exchange, for example, can be upgraded to a new service pack without rebooting the server, but the Exchange services are all stopped during the upgrade (which again can take 10 minutes or more).
So, once again, it comes down to the reboot factor. Microsoft spent a lot of time removing reboots from Windows 2000 and still completely missed these ones. IMO, you really shouldn't need to reboot a system unless you're changing either a core driver (say like a storage controller that you booted off of), or some core kernel components. There's no real reason why something like the TCP/IP stack in Windows 2000 can't be changed without a reboot, since it's a loadable driver (tcpip.sys), not core to the OS, and the system would have no valid reason for blowing up if it was unloaded for a short period.
The only parts of the MS Windows interface that are bitmap based are icon and toolbar images. Window decorations, 'widgets' and everything else is constructed via the use of the TrueType scalable font 'Marlett'. Now, unless Microsoft changed this in Windows XP as part of the skinnable UI, this should scale reasonably well to high dpi. I suppose you could always just use ugly up-scaling of the rest of the bitmapped elements.
Far worse would be the effect to webpages, since most are tuned far too specificly to the average DPI of most monitor/OS combos.
The major problem you'll face isn't authentication (which pam_krb can handle), it's the fact that the stock Microsoft Active Directory LDAP directory doesn't have the needed schema to support UNIX clients without the Services for Unix 2.0 package (in which case, you could just use NIS). To complete the package, you need to have both nss_ldap and pam_ldap, however the format of Active Directory doesn't match the format that nss_ldap expects.
Somewhere on MS's site is some code for adding/deleteing/modifying users from Active Directory along with a script that could cull the users from active directory and allow you to import that into a passwd file (minus the passwords, which would be authenticated via kerberos). This is certainly not for the faint-at-heart, and I looked into it for a while and eventually gave up on that idea because it was simply too complex and too likely to break for no apparent reason.
Re:And in a completely unrelated story...
on
Duct Tape
·
· Score: 1
No, no, no. It's not TV that's doing it -- it's those damned computer games and rap music that's doing it!
I can list one way to fix ICANN -- throw out the remaining original members. They finished their initial task of getting the @Large members in there, it's time for them to step back and let the final four members be elected (perhaps by by giving votes to ICANN sponsors this time against those four seats). I think that would be an adequate blend -- 5 voted by internet users worldwide, and 4 voted by those who pay the money to keep ICANN running. Add to this, some kind of charter that prevents the kind of abuse that's going on right now by the old 4 members, and you might end up with a functional body for these matters.
Actually, it fits Microsoft's "Embrace and Extend" moto better than anything else. It mutates an otherwise healthy protocol until it chokes off the original healthy protocol.
Problem is that ICANN is trying to change it without really trying to change it. Despite the introduction of these new TLDs, they are still giving trademark holders a monopoly over them because of the domain name dispute arbitration via WIPO. What difference would it make if we had 2 TLDs or 2,000 TLDs if there will be no variety between them? Unless there's strict restrictions on what TLDs can be used for what purposes (like.edu, for example), the abuse of the domain name system will continue forever.
I think most of this would break down the moment you have a CVS conflict in the commits, since for a short time, there is no authorative version of the file to be signed. The server (or client, I don't know) instead injects the diff at the point in the source where the conflict occured. Given a compromised server, it's possible that arbitrary user code could be inserted.
Then by all means propose a new one! While Outlook's interface is pretty damn bad, it still seems to be one of the cleanest (have you ever used Novell Groupwise?). Evolution isn't even yet beta quality software, and the developers are very much open to suggestions on how to improve the user interface. Complain if you must, but complaints with suggestions on how to fix are much more helpful.
The only drop in CD sales that the recording industry has seen definitively is of CD singles (which are a waste of money for most people). CD album sales, and everything else increased and have increased each and every year. The only place they managed to notice a drop in CD sales was in some universities and colleges. What is responsible for this is anyone's guess, and it seems that everyone does. The RIAA claims CD singles sales are down due to MP3 and Napster, but it could just as easily be that people are buying full albums instead (which is why sales are up). A decrease in buying around colleges could be that students have less money. The facts are that there is not enough isolated data available to make an informed decision.
It doesn't really matter either way. Napster, Aimster, MP3.com, etc are not in the RIAA clique, and hence must be destroyed by any and all means possible.
The GPL license is not viral, and any sense in which you claim that it is becomes mere FUD and is just plain wrong. This is the height of Microsoft marketing trying to associate evil with the GPL.
The so-called viral aspect of the GPL is the one thing that Microsoft rarely regulates -- the ability to link (even dynamically) your application against their copyright protected library for free, whereas the GPL suggests that if you do this, your entire work must then be covered by the GPL. This is one aspect that has, in the past, been misunderstood by a number of developers and is important to recognise.
Re:Directed towards investors, not customers
on
Mundie Responds
·
· Score: 2
Craig Mundie spent a disportionate amount of his speech making vague accusations about open source and it's viability, while touting the amount of success that they've had. It's obvious this is aimed at businesses who may be considering supporting either open source systems, or developing open source software of their own. By cutting off the supply of new programs, Microsoft probably hopes to smother development of the Linux and other similar platforms.
Throughout the speech and followup, he goes to great lengths to explain his perceived weaknesses of the GPL, instead of focusing on the strengths of conventional commercial development. Rather he usually just ends up saying that conventional commercial licenses should be used because they're "proven". However, this neglects the fact that few software companies, outside of either specialized vertical markets, or games are doing all that well as Microsoft gobbles up more and more consumer dollars.
There are other parts of the commentary that is nothing but pure fear mongering. For example, this statement: "Some of the tension I see between the GPL and strong business models is by design, and some of it is caused simply because there remains a high level of legal uncertainty around the GPL--uncertainty that translates into business risk." While, he certainly is welcome to his opinion, I think this kind of statement is despicable, since it serves only one purpose -- to create fear, uncertianty and doubt. He conveniently forgets that even if the GPL is ruled to be unenforcable, the only result is that you'd be unable to redistribute GPL covered works in any form aside from the normal fair-use exceptions. That is far less uncertainty than most commercial licenses that disclaim all warranties, prohibit rights such as reverse engineering, and sometimes even from disclosing certain details about the software, such as performance, flaws, bugs and capabilities.
Sounds like he's advocating "Political Correctness". Gotta make sure that everything I say can't be taken as slander against someone, and that no one ever gets offended, and that my opinions that oppose anyone else are suppressed.
It's certainly a nice idea, but rather misguided. It's generating traffic that people who do maintain and check firewall logs would rather not deal with, and doesn't fix the core problem -- machines that aren't kept up-to-date with security fixes. You'd think that with all the press these self-replicating worms are getting that people'd be more vigilant about updating their systems. Hell, I was gone for a week and was nervous about not having the systems constantly up-to-date.
The only version of the BSD license that's not officially sanctioned by RMS as being 'non-free' are the ones that include advertising clauses. He prefers calling the BSD style licenses without the advertising clause XFree86 style, and those with the clause BSD.
The original clause is:
3. All advertising materials mentioning features or use of this software must display the following acknowledgement:
This product includes software developed by the University of California, Berkeley and its contributors.
The rational behind not using licenses that include this clause is quite sound, you can see some arguments for omiting that clause at http://www.gnu.org/philosophy/bsd.html (don't worry, it doesn't say the BSD license sucks and everyone should use GPL). You could imagine how such a clause could cause problems if, perhaps, you had a product that included 40-50 of programs with those kinds of licenses -- the required statements could outnumber actual ad-space.
Slashdot Mirror
Microsoft's format is already open? All I've ever seen is MS offering to license the technology (for royalties, of course) to manufacturers of portable music players. In fact, if the format was open, it'd probably undermine Microsoft's ability to sell the "Digital Rights Management" to studios. The end result is they need to kill both RealPlayer and WinAMP to take over as the dominant media player, and making the format of "Windows Media Audio" an open format would allow their competitors to support this just as easily as Microsoft, so all they open is the APIs to their proprietary technology. If you want to playback on a supported version of Windows or MacOS, fine, but everything else is offlimits.
No, people out in the maritime provinces say things like "arfter" (after) and "ruff" (roof) and just generally use too many r's. People on the prairies tend to mispronounce "bury" (e.g. rhyming with jury instead of berry). I'm not sure where "aboot" comes from. All I know is that everytime I hear someone in Manitoba say "aboot", I feel the need to hit them very, very hard.
The problem is that companies like Tivo have PATENTS on this kind of technology, and Open Source companies (or individuals) can't usually afford to pay for an unrestricted license that would allow them to distribute such a work. While someone such as myself, who lives in Canada, a country that doesn't recognise software patents, a large number of people who live in countries like the United States will not have this option. Such a patent would make even an end user that uses such a program a criminal, regardless of if they knew they were breaking the law by using the program or not.
As for having Windows 2000 share that much, you probably don't want to set it up that way. The way MS SQL Server was designed, it will take every last byte of memory and can be a real hog for CPU cycles if it has work it believes it has to do.
It may or may not be developed by Microsoft, but they contract out production to other companies. Many of the SideWinder joypads, for example, are made by Mitsumi (according to the circuit board) with Microsoft's name the visible outside name. In fact, I seriously doubt that Microsoft will build X-Boxes themselves, either -- that, too, will probably fall to one of these other hardware companies.
I didn't say they weren't enforcing it. But up until very recently, if your program was free and generated LZW compressed GIFs, Unisys had no problems with that (you didn't have to pay royalties). Although they didn't rescind on that completely, if you want to use LZW compressed GIFs from such a "non-licensed" program, Unisys is being an ass and enforcing licensing for that. That action, however, IS recent.
Yet somehow that reminds me of Unisys's stance on their LZW compression patent up until very recently when they discovered they could extort more money from people by not having such a lax restriction. "You want GIFs on your site produced by a non-licensed generator? Pay us $5,000 now or face legal action!". They're not making any promises that they won't go after royalties from free distribution places at a later date, so I still wouldn't trust them.
Windows 2000 also just installs SLOWLY. Compared to NT 4.0, where you'd configure partitioning, it'd spend 15-20 minutes copying files (coffee break), come back, answer more questions, and the final stage took about 15-20 seconds to copy the rest of the files, then you reboot. Windows 2000 is very different. You have to answer the partitioning questions, then waste 5-10 minutes waiting for it to format, only to come back later to answer a dozen more questions, and wait another half hour for it to copy and configure all the installed software. And if you're setting up a 2000 server with active directory, it gets worse as you have to wait for it to reboot again, and then run adpromo.exe to make it a domain controller, and wait for that to install.
I would however debate if Windows is now stable enough for mission-critical servers -- my experience with Windows 2000 and Active Directory has been that although it doesn't crash, it doesn't act or react consistently, and strange things can go wrong with little apparent reason or fix. Just as an example, take the group policy feature in Windows 2000. If there's any rights granted to specific users and one or more of those users are deleted, the group policies will stop being applied properly.
Maybe Windows 2004 (whatever MS decides to name that) will be ready, but 2000 definately isn't, and XP isn't likely to be ready for completely mission critical use either (in cases like this, redundancy doesn't fix anything, because if one server has the problem, the other will too). Windows 2000 (and probably XP) are much better as workstation OS's than NT4.0 or Win9x, and if you need to use Windows, Win2000 is probably your best bet. There's still rough edges on Windows 2000 that are nearly completely unexplainable or produce obscure error messages (even the much maligned UNIX error message, "printer is on fire", gives more accurate information on what's wrong to an administrator than some of the eventlog messages that Windows 2000 can generate).
So, once again, it comes down to the reboot factor. Microsoft spent a lot of time removing reboots from Windows 2000 and still completely missed these ones. IMO, you really shouldn't need to reboot a system unless you're changing either a core driver (say like a storage controller that you booted off of), or some core kernel components. There's no real reason why something like the TCP/IP stack in Windows 2000 can't be changed without a reboot, since it's a loadable driver (tcpip.sys), not core to the OS, and the system would have no valid reason for blowing up if it was unloaded for a short period.
Far worse would be the effect to webpages, since most are tuned far too specificly to the average DPI of most monitor/OS combos.
Just change the formatting to either "code" or "plain old text" instead of HTML.
Somewhere on MS's site is some code for adding/deleteing/modifying users from Active Directory along with a script that could cull the users from active directory and allow you to import that into a passwd file (minus the passwords, which would be authenticated via kerberos). This is certainly not for the faint-at-heart, and I looked into it for a while and eventually gave up on that idea because it was simply too complex and too likely to break for no apparent reason.
No, no, no. It's not TV that's doing it -- it's those damned computer games and rap music that's doing it!
I can list one way to fix ICANN -- throw out the remaining original members. They finished their initial task of getting the @Large members in there, it's time for them to step back and let the final four members be elected (perhaps by by giving votes to ICANN sponsors this time against those four seats). I think that would be an adequate blend -- 5 voted by internet users worldwide, and 4 voted by those who pay the money to keep ICANN running. Add to this, some kind of charter that prevents the kind of abuse that's going on right now by the old 4 members, and you might end up with a functional body for these matters.
Actually, it fits Microsoft's "Embrace and Extend" moto better than anything else. It mutates an otherwise healthy protocol until it chokes off the original healthy protocol.
Problem is that ICANN is trying to change it without really trying to change it. Despite the introduction of these new TLDs, they are still giving trademark holders a monopoly over them because of the domain name dispute arbitration via WIPO. What difference would it make if we had 2 TLDs or 2,000 TLDs if there will be no variety between them? Unless there's strict restrictions on what TLDs can be used for what purposes (like .edu, for example), the abuse of the domain name system will continue forever.
I think most of this would break down the moment you have a CVS conflict in the commits, since for a short time, there is no authorative version of the file to be signed. The server (or client, I don't know) instead injects the diff at the point in the source where the conflict occured. Given a compromised server, it's possible that arbitrary user code could be inserted.
Then by all means propose a new one! While Outlook's interface is pretty damn bad, it still seems to be one of the cleanest (have you ever used Novell Groupwise?). Evolution isn't even yet beta quality software, and the developers are very much open to suggestions on how to improve the user interface. Complain if you must, but complaints with suggestions on how to fix are much more helpful.
It doesn't really matter either way. Napster, Aimster, MP3.com, etc are not in the RIAA clique, and hence must be destroyed by any and all means possible.
The so-called viral aspect of the GPL is the one thing that Microsoft rarely regulates -- the ability to link (even dynamically) your application against their copyright protected library for free, whereas the GPL suggests that if you do this, your entire work must then be covered by the GPL. This is one aspect that has, in the past, been misunderstood by a number of developers and is important to recognise.
Throughout the speech and followup, he goes to great lengths to explain his perceived weaknesses of the GPL, instead of focusing on the strengths of conventional commercial development. Rather he usually just ends up saying that conventional commercial licenses should be used because they're "proven". However, this neglects the fact that few software companies, outside of either specialized vertical markets, or games are doing all that well as Microsoft gobbles up more and more consumer dollars.
There are other parts of the commentary that is nothing but pure fear mongering. For example, this statement: "Some of the tension I see between the GPL and strong business models is by design, and some of it is caused simply because there remains a high level of legal uncertainty around the GPL--uncertainty that translates into business risk." While, he certainly is welcome to his opinion, I think this kind of statement is despicable, since it serves only one purpose -- to create fear, uncertianty and doubt. He conveniently forgets that even if the GPL is ruled to be unenforcable, the only result is that you'd be unable to redistribute GPL covered works in any form aside from the normal fair-use exceptions. That is far less uncertainty than most commercial licenses that disclaim all warranties, prohibit rights such as reverse engineering, and sometimes even from disclosing certain details about the software, such as performance, flaws, bugs and capabilities.
Sounds like he's advocating "Political Correctness". Gotta make sure that everything I say can't be taken as slander against someone, and that no one ever gets offended, and that my opinions that oppose anyone else are suppressed.
It's certainly a nice idea, but rather misguided. It's generating traffic that people who do maintain and check firewall logs would rather not deal with, and doesn't fix the core problem -- machines that aren't kept up-to-date with security fixes. You'd think that with all the press these self-replicating worms are getting that people'd be more vigilant about updating their systems. Hell, I was gone for a week and was nervous about not having the systems constantly up-to-date.
The original clause is:
3. All advertising materials mentioning features or use of this software must display the following acknowledgement:
This product includes software developed by the University of California, Berkeley and its contributors.
The rational behind not using licenses that include this clause is quite sound, you can see some arguments for omiting that clause at http://www.gnu.org/philosophy/bsd.html (don't worry, it doesn't say the BSD license sucks and everyone should use GPL). You could imagine how such a clause could cause problems if, perhaps, you had a product that included 40-50 of programs with those kinds of licenses -- the required statements could outnumber actual ad-space.