![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
Gates: 'Linux makes interoperability harder'
- Thread starter DR_K13
- Start date
Just more FUD from Billy and his gang.
A few things that come to mind when hearing this:
1) Microsoft's web publishing tools create non-standard, non-interoperable, non-compliant pages.
2) My machine running NT 4.0 will not interoperate all applications that were made for Windows '95/DOS, and Office '97 won't open many of my Office 2003 documents.
3) FreeBSD can read my NTFS partitions without problems, but XP has no clue that my UFS partitions even exist.
Of course I could be wrong if he meant interoperaility with virus', games, and a easy to use Windows Update to fix a few of the issues.
A few things that come to mind when hearing this:
1) Microsoft's web publishing tools create non-standard, non-interoperable, non-compliant pages.
2) My machine running NT 4.0 will not interoperate all applications that were made for Windows '95/DOS, and Office '97 won't open many of my Office 2003 documents.
3) FreeBSD can read my NTFS partitions without problems, but XP has no clue that my UFS partitions even exist.
Of course I could be wrong if he meant interoperaility with virus', games, and a easy to use Windows Update to fix a few of the issues.
A few thoughts:
Does this mean Windows can do all of the above flawlessly? Abso-freaking-lutely not. However, it doesn't have toit just has to remain "good enough" to the ones writing the checks. This is how Apple has the creative (image, video, publishing) market cornered, by being an "industry standard" based on years of "good enough" performance, and both Microsoft and any Linux distro group would have to work three times as hard to even make a dent in those industries. That's just the way it is.
Please note that I'm not claiming Gates' letter wasn't a lot of hyperbole, but what he did is take a much smaller issue regarding integration and "playing nice" and made a larger deal of it than he should have. This is markedly so because as some have pointed out the same arguments could be made about Microsoft. However, if you guys want to address the points within the letter then by all means do so, but with a warning: don't be too wuick to dismiss the points and don't react with knee-jerk accusations about ten-year-old products not cutting the mustard. An argument can be made against Gates' statements, but toeing any platform lines is just going to sound like frothing zealotry and not really make the hard points for anyone.
- Linux still has no working standard for dealing with interoperations with a business Microsoft network. Yes, LDAP works "okay" and samba is doable enough, but the methodology of integration is a hodge-podge of choose-your-own-adventure-type packages, some that have more success than others but none which are industry standard. This is fine if a machine is to be a server operating under the domain tree for files or intranet or firewall, but creates significant issues when implementing them as workstations. It's not that it can't be done (it sure as hell can), but that the worth of doing so will either break even due to support licensing costs or be greater due rebuilding interoperability (from nearly scratch).
- There are no "work-critical" applications available to mitigate costs of implementation, and plenty of "killer apps" that would be lostno matter what anyone says, there is no *nix equivalent to the Office/Outlook/Exchange solution (anyone who thinks Outlook is "just an e-mail client" knows jack shit about Outlook).
- Linux has strengths mostly on its ties to Unix, which make it a strong contender in the server room. This is only going to be seen directly by IT or IS, and be practically transparent to other sections of any company or corporation. Until Linux can display a solid, standardized (and organized) offering of solutions in the office and production rooms, there is very little evidence to provide to the ones paying the budget that an "if it ain't broke" methodology isn't the best, safest course of action.
Does this mean Windows can do all of the above flawlessly? Abso-freaking-lutely not. However, it doesn't have toit just has to remain "good enough" to the ones writing the checks. This is how Apple has the creative (image, video, publishing) market cornered, by being an "industry standard" based on years of "good enough" performance, and both Microsoft and any Linux distro group would have to work three times as hard to even make a dent in those industries. That's just the way it is.
Please note that I'm not claiming Gates' letter wasn't a lot of hyperbole, but what he did is take a much smaller issue regarding integration and "playing nice" and made a larger deal of it than he should have. This is markedly so because as some have pointed out the same arguments could be made about Microsoft. However, if you guys want to address the points within the letter then by all means do so, but with a warning: don't be too wuick to dismiss the points and don't react with knee-jerk accusations about ten-year-old products not cutting the mustard. An argument can be made against Gates' statements, but toeing any platform lines is just going to sound like frothing zealotry and not really make the hard points for anyone.
Curiously, Apple might just be the right people to make the unix tools viable drop-in replacements. It's already possible to replace windows servers with OS X servers in many cases, and I imagine they are working to increase the number of roles it can take on.
I must say that it would be interesting if Apple took a sizeable chunk out of the windows server market with the same tools as linux.
(The final proof that the presentation and polish is more important than the purchase price?)
I must say that it would be interesting if Apple took a sizeable chunk out of the windows server market with the same tools as linux.
(The final proof that the presentation and polish is more important than the purchase price?)
I wouldn't say more important, but creating an abstraction for the user that does not require the lack of abstraction that the expert may desire does make a huge difference.HHunt said:Curiously, Apple might just be the right people to make the unix tools viable drop-in replacements. It's already possible to replace windows servers with OS X servers in many cases, and I imagine they are working to increase the number of roles it can take on.
I must say that it would be interesting if Apple took a sizeable chunk out of the windows server market with the same tools as linux.
(The final proof that the presentation and polish is more important than the purchase price?)
I'd predict that if Apple does something like that, it will be through adding APIs similar to what they did with OS X to make a mach kernel that can do Unix-y stuff while retaining the popular Apple interface.
NewBlackDak
[H]ard|Gawd
- Joined
- Aug 31, 2003
- Messages
- 1,322
We've replaced our Appletalk, server, a windows server, and a novell server with a single dual G5 Xserver. It does everything those machines did, and better in some places. It's even faster than all three of those machines were. Apple's extremely simple tools and configuration made it tons easier. I get the *NIX like backend, so it's easy for me to setup/script stuff we need. I get Apple's easyness with front-end stuff.
All arguments about Windows interoperability are pretty much junk. After all, of course an all-Windows network is going to be able to interoperate quite well.
Where I work, we have an all *nix network. We use Linux and Solaris on our servers, and Linux and Mac OS X on our workstations. For most of the masses, we use Linux-based thin-client machines.
We have zero interoperability problems in our organization. We have a single sign-on system that works across multiple versions of operating systems on different hardware platforms. I'd love to see a Windows machine interoperate well with a server room full of Sparc machines. Remember that not everyone wants to interoperate with Windows or a Windows network. I work in telecommunications, which means nearly everything we run uses Sun hardware and Solaris for the voice side of the house, and Linux for the data side.
We pay no licensing fees for any of the software running on the data side of the house, yet we have over 150 servers. This includes our 35-server mail cluster, which pushes over 2 million emails an hour, including full spam and virus checking on those emails. We pay no licensing fees for our databases or our web servers, and we damn sure don't pay for "Client Access Licenses" for any of our file servers.
While we could interoperate with Windows, we have zero need to do so, as Windows would be more of a hinderance than a help in any case.
Next time you think about Windows being interoperable with other systems, try getting a Windows box to authenticate against a RADIUS server or a TACACS box, or try using an NFS file server. The only thing Windows is designed to work with well is Windows. How is that for interoperability?
Where I work, we have an all *nix network. We use Linux and Solaris on our servers, and Linux and Mac OS X on our workstations. For most of the masses, we use Linux-based thin-client machines.
We have zero interoperability problems in our organization. We have a single sign-on system that works across multiple versions of operating systems on different hardware platforms. I'd love to see a Windows machine interoperate well with a server room full of Sparc machines. Remember that not everyone wants to interoperate with Windows or a Windows network. I work in telecommunications, which means nearly everything we run uses Sun hardware and Solaris for the voice side of the house, and Linux for the data side.
We pay no licensing fees for any of the software running on the data side of the house, yet we have over 150 servers. This includes our 35-server mail cluster, which pushes over 2 million emails an hour, including full spam and virus checking on those emails. We pay no licensing fees for our databases or our web servers, and we damn sure don't pay for "Client Access Licenses" for any of our file servers.
While we could interoperate with Windows, we have zero need to do so, as Windows would be more of a hinderance than a help in any case.
Next time you think about Windows being interoperable with other systems, try getting a Windows box to authenticate against a RADIUS server or a TACACS box, or try using an NFS file server. The only thing Windows is designed to work with well is Windows. How is that for interoperability?