The article basically supports Microsoft's claim. Technology goes in phases.
IBM was the Mainframe phase, DEC was the Mini phase, and Microsoft/Dell road the PC phase. Each of these companies have had their Golden years, followed by a plateau, or in the case of DEC a major decline.
The problem is you are still thinking in terms of the PC Phase and worried about Microsoft there. What you need to be doing is trying to uncover what the next big thing is, because that's where Microsoft will stumble.
Basically I don't think Microsoft has anything to fear from you.
"I have many friends from East Germany who would find the comparison highly offensive. "
I agree. The highly modded post I saw yesterday comparing the use of free software to the struggle for Civil Rights was even more offensive.
The real turning point here seems to be that the OSS zealots have become even more desperate. They're going off the deep end and lashing out with bizarre hateful arguments.
"The major difference between the two solutions is that in the first solution I am not sponsoring a foreign corporation"
I assume you must live in Taiwan then, since in both examples you are supporting the Taiwanese electronics industry by purchasing a computer.
Or perhaps it's a combination of Taiwan, Korea, China and Malaysia. Regardless you are benefiting from globalism... i.e. that computer cost you far less because it was imported from Taiwan. The same is true for software as well as other products such as automotives, etc.
Your post completely minimalizes the contributions of men such as Martin Luther King Jr., Susan B. Anthony and many others who have struggled for Civil Rights in America.
Free Software written by a bunch of rich white boys is not in the least bit similar to the struggles that have taken place in the United States with regards to obtaining equal opportunity.
Open source software is already being considered. It's also being rejected.
So now the OSS community wants to go one step forward in mandating, so they write up this bill which requires you to provide justification whenever you don't use open source software.
Justification means paperwork. It also means a review committee most likely, what's the point of paperwork if nobody is going to read it?
I know how government operates. I know how issues are spun. I know for a fact that if the bill stated justification must be provided when choosing open source software, the slashbots would be up in arms. You spin this as "consideration", and I'll point out the truth which is "mandating".
"However, this feature was removed after Windows 2000 RC2 due to technical issues."
Not so much technical as priorities. They discovered that they were essentially rewriting all the code in SQL Server and that it wasn't signifigantly faster than SQL Server. It was decided it was simply a waste of time.
Like others have pointed out SQL Server, Oracle, etc. already maintain their own caches of data in memory. So on SQL that accesses the same chunks of data, the response is really quite fast.
"I know whenever I see an app that advertises itself as cross-platform, but turns out to be written in Java, I tend to kind of dismiss it."
Sadly, you are correct.
I was recently evaluating IM solutions, and every one I encountered that was written in Java was sold as "cross-platform". After evaluation it quickly became apparent it sucked for one reason or another.(usually kludgey UI)
The factors you bring up are simply not signifigant towards the results. I've been doing stress testing of internal applications for my employer for a while now. I don't know much about testing embedded systems, but I do know enterprise systems.
The issues you bring up might result in a variation in results in the order of 1%. That's just not statistically signifigant for the purpose of this test.
"And on your comments about the business relationships: it would be one thing for Microsoft to have paid some lab to do the test. It's quite another when MicroSoft and the Lab's parent company are so deeply in bed together that the benefit of the doubt disappears completely."
I still see no validity to this claim until you can prove otherwise.
"They disabled last access time updating under windows. They didn't under Linux. This is enough to account for these differences, I suspect."
Ok, you have a hypothesis. That's the first step.
Now prove whether it is true or not. Otherwise you are simply following a pattern of Argument by Assertion.
http://www.io.com/~jwtlai/illogic.html
This is a particularly easy argument for you to back up, since we are talking about something which can be measure objectively rather than just an opinion.
Re:Flawed testing methodology / conflict of intere
on
What's Microsoft Up To?
·
· Score: 2, Interesting
"Second of all, "repeatability" is meaningless in terms of determining statistically significant results."
I believe you are confusing social science with physical science. Benchmarking a computer system is a physical science, the system should behave in a deterministic fashion provided you have properly identified all inputs.
It's like measuring the length of a 2x4. You do so twice, solely to verify your results. You don't need to sample the lengths of many 2x4's to understand the pattern of behavior which applies to the length of a 2x4.
"I could probably go on, but this should be enough.."
Since the methodology used to conduct the benchmark has been published, the only legitimate complaint you can make is to reproduce the benchmark and show exactly how they misconfigured the system.
What I see in your post is idle speculation. The attempt to claim conflict of interest may have relevance only if there has been a history of conflict of interest influencing test results in the past. Unfortunately for the sake of your argument that has not been the case.
In the world of benchmarking, the key is that the documentation have enough information in it so that the results can be reproduced by other parties.
From reading the veritest report this appears to be the case.
In other words... If you doubt the benchmarks, reproduce them and state what you think Veritest did wrong.
If you'll recall history. When Mindcraft showed IIS was faster than Apache on Linux, there was much gnashing of teeth with similar charges. As it turned out, Mindcraft was vindicated when the results were verified by PCWeek(?).
Unfortunately the Linux zealots never apologized to Mindcraft. The Linux adults went and fixed the problems identified in the kernel.
They used Compaq servers... These are typical of the kind of servers running in data centers.
Can you point to me servers from Compaq, Dell, IBM, etc. running AMD processors in 4-way configurations?
Re:File server shoot-out? You're kidding, right?
on
What's Microsoft Up To?
·
· Score: 2, Insightful
"Of course, this is just one part of Microsoft's strategy against Linux and OSS. But I'm pretty sure that this salvo will fall on deaf ears."
You're right. Nobody has been listening to the Samba people claim their server was faster.
This was more of a pride thing to Microsoft. They saw an article showing their server was not the fastest one out there, and so they went to their developers and said "fix this."
And so they did. Now it's not just faster, it's signifigantly faster.
I've been working with IIS now for about 6 years and I've never seen the Metabase become corrupted. I imagine it could happen if your machine crashed while updating the metabase, but how common is that?
No, the main reason people are familiar with the Metabase is that's where the IIS configuration settings are located and it's far easier to write a script to modify them in the metabase than to use the UI. Especially when you're doing this to hundreds of web servers.
Now that's interesting. First you want a way to schedule Windows Update, and now you complain that Windows Update doesn't get all the patches.
Seems like you should have properly defined your problem to begin with.
"And I wasn't talking about the task scheduler, I was referring to the option to automatically remotely update machines (completely) without having to do it on every box."
Well if you want to spend money you can go with HFNetChk pro. If you don't want to spend money you can download psexec from sysinternals.com and script your own solution.
OleDB is the data access technology. ADO is the object-oriented, vendor-neutral specification for transparent data persistence. It guarantees important levels of support for transaction management, identity management, and caching.
Slashdot Mirror
It'd be like the dot com companies...
...
1. Buy $100 bills
2. Reprint as $20s
3.
4. Profit!
It certainly does make a surge if you stroke it long and hard enough.
I don't think you read the article.
The article basically supports Microsoft's claim. Technology goes in phases.
IBM was the Mainframe phase, DEC was the Mini phase, and Microsoft/Dell road the PC phase. Each of these companies have had their Golden years, followed by a plateau, or in the case of DEC a major decline.
The problem is you are still thinking in terms of the PC Phase and worried about Microsoft there. What you need to be doing is trying to uncover what the next big thing is, because that's where Microsoft will stumble.
Basically I don't think Microsoft has anything to fear from you.
Actually spreading Fear, Uncertainty and Doubt seems to be the hallmark of the Open Source world these days... :(
"I have many friends from East Germany who would find the comparison highly offensive. "
I agree. The highly modded post I saw yesterday comparing the use of free software to the struggle for Civil Rights was even more offensive.
The real turning point here seems to be that the OSS zealots have become even more desperate. They're going off the deep end and lashing out with bizarre hateful arguments.
"because the cost of the paperwork to justify buying proprietary is too high ? "
Because the path of least resistance is using open source software.
*THAT* is the point of using the justification phrasology in the bill. You are naive to believe otherwise.
"there is no extra added overhead of paperwork"
That is not the way this bill is worded.
"The major difference between the two solutions is that in the first solution I am not sponsoring a foreign corporation"
I assume you must live in Taiwan then, since in both examples you are supporting the Taiwanese electronics industry by purchasing a computer.
Or perhaps it's a combination of Taiwan, Korea, China and Malaysia. Regardless you are benefiting from globalism... i.e. that computer cost you far less because it was imported from Taiwan. The same is true for software as well as other products such as automotives, etc.
Your argument overall is very naive.
"It needs plenty of care and feeding, so it's better that the IT budget gets spent on people to do that instead of supporting huge licensng fees."
NO! NO! NO! A thousand times NO!
I used to work for state government, and what you are describing is EXACTLY the mentality which has balooned the budget problems.
You are encouraged, nay rewarded, to not do things efficiently but rather to guarantee your job position.
You're just encouraging bad behavior.
Your post completely minimalizes the contributions of men such as Martin Luther King Jr., Susan B. Anthony and many others who have struggled for Civil Rights in America.
Free Software written by a bunch of rich white boys is not in the least bit similar to the struggles that have taken place in the United States with regards to obtaining equal opportunity.
Your post is insulting and insensitive.
Open source software is already being considered. It's also being rejected.
So now the OSS community wants to go one step forward in mandating, so they write up this bill which requires you to provide justification whenever you don't use open source software.
Justification means paperwork. It also means a review committee most likely, what's the point of paperwork if nobody is going to read it?
I know how government operates. I know how issues are spun. I know for a fact that if the bill stated justification must be provided when choosing open source software, the slashbots would be up in arms. You spin this as "consideration", and I'll point out the truth which is "mandating".
"However, this feature was removed after Windows 2000 RC2 due to technical issues."
Not so much technical as priorities. They discovered that they were essentially rewriting all the code in SQL Server and that it wasn't signifigantly faster than SQL Server. It was decided it was simply a waste of time.
Like others have pointed out SQL Server, Oracle, etc. already maintain their own caches of data in memory. So on SQL that accesses the same chunks of data, the response is really quite fast.
"I know whenever I see an app that advertises itself as cross-platform, but turns out to be written in Java, I tend to kind of dismiss it."
Sadly, you are correct.
I was recently evaluating IM solutions, and every one I encountered that was written in Java was sold as "cross-platform". After evaluation it quickly became apparent it sucked for one reason or another.(usually kludgey UI)
The factors you bring up are simply not signifigant towards the results. I've been doing stress testing of internal applications for my employer for a while now. I don't know much about testing embedded systems, but I do know enterprise systems.
The issues you bring up might result in a variation in results in the order of 1%. That's just not statistically signifigant for the purpose of this test.
"And on your comments about the business relationships: it would be one thing for Microsoft to have paid some lab to do the test. It's quite another when MicroSoft and the Lab's parent company are so deeply in bed together that the benefit of the doubt disappears completely."
I still see no validity to this claim until you can prove otherwise.
"They disabled last access time updating under windows. They didn't under Linux. This is enough to account for these differences, I suspect."
Ok, you have a hypothesis. That's the first step.
Now prove whether it is true or not. Otherwise you are simply following a pattern of Argument by Assertion.
http://www.io.com/~jwtlai/illogic.html
This is a particularly easy argument for you to back up, since we are talking about something which can be measure objectively rather than just an opinion.
"Second of all, "repeatability" is meaningless in terms of determining statistically significant results."
I believe you are confusing social science with physical science. Benchmarking a computer system is a physical science, the system should behave in a deterministic fashion provided you have properly identified all inputs.
It's like measuring the length of a 2x4. You do so twice, solely to verify your results. You don't need to sample the lengths of many 2x4's to understand the pattern of behavior which applies to the length of a 2x4.
"I could probably go on, but this should be enough.."
Since the methodology used to conduct the benchmark has been published, the only legitimate complaint you can make is to reproduce the benchmark and show exactly how they misconfigured the system.
What I see in your post is idle speculation. The attempt to claim conflict of interest may have relevance only if there has been a history of conflict of interest influencing test results in the past. Unfortunately for the sake of your argument that has not been the case.
In the world of benchmarking, the key is that the documentation have enough information in it so that the results can be reproduced by other parties.
From reading the veritest report this appears to be the case.
In other words... If you doubt the benchmarks, reproduce them and state what you think Veritest did wrong.
If you'll recall history. When Mindcraft showed IIS was faster than Apache on Linux, there was much gnashing of teeth with similar charges. As it turned out, Mindcraft was vindicated when the results were verified by PCWeek(?).
Unfortunately the Linux zealots never apologized to Mindcraft. The Linux adults went and fixed the problems identified in the kernel.
They used Compaq servers... These are typical of the kind of servers running in data centers.
Can you point to me servers from Compaq, Dell, IBM, etc. running AMD processors in 4-way configurations?
"Of course, this is just one part of Microsoft's strategy against Linux and OSS. But I'm pretty sure that this salvo will fall on deaf ears."
You're right. Nobody has been listening to the Samba people claim their server was faster.
This was more of a pride thing to Microsoft. They saw an article showing their server was not the fastest one out there, and so they went to their developers and said "fix this."
And so they did. Now it's not just faster, it's signifigantly faster.
That's the nature of competition!
Have you considered looking into a different career?
You might want to keep your options open if you're going to bet against successful technologies.
I've been working with IIS now for about 6 years and I've never seen the Metabase become corrupted. I imagine it could happen if your machine crashed while updating the metabase, but how common is that?
No, the main reason people are familiar with the Metabase is that's where the IIS configuration settings are located and it's far easier to write a script to modify them in the metabase than to use the UI. Especially when you're doing this to hundreds of web servers.
What's amazing is that not one of the claims made in this parent post has any factual evidence to substantiate them.
I love this...
"This article addresses what many people consider to be open source's weekpoint. It is however it's strongest point."
This could have been written by the Iraqi Minister of Information...
Welcome to FUD 101, I guess.
Ahh perhaps if you actually wrote some code using ADO you might begin to understand the differences.
kexec still does a reboot, even if it's just a soft boot instead of a hard one.
Now that's interesting. First you want a way to schedule Windows Update, and now you complain that Windows Update doesn't get all the patches.
Seems like you should have properly defined your problem to begin with.
"And I wasn't talking about the task scheduler, I was referring to the option to automatically remotely update machines (completely) without having to do it on every box."
Well if you want to spend money you can go with HFNetChk pro. If you don't want to spend money you can download psexec from sysinternals.com and script your own solution.
"How's that for another complaint."
Check and Mate
You've confused ADO with OleDB.
OleDB is the data access technology. ADO is the object-oriented, vendor-neutral specification for transparent data persistence. It guarantees important levels of support for transaction management, identity management, and caching.
Sounds like JDO.