And what I'm telling you is that your not doing an apples to apples comparison. Try running a poorly supported piece of hardware under Windows (one where you can't just double click on some driver). You've never tried to use a non windows piece of hardware on a windows computer so you don't really have anything to compare your experience with Linux too. You might be able to get it sort of kind of work under limited conditions but not very smoothly.
You can far more easily get something to 1/2 work under Linux. 6 months from now it will be a different story because the drivers will be integrated and everything will just self install. If 3 Linux distributions don't support your video card than most likely your video card was added to a recent version of X, for example.
Don't forget the hardware you are talking about was built for windows and shipped with windows. If you bought an AMD Sun you wouldn't be shocked if you had driver problems under Windows. Pretty much there is a 6 mo delay between the time hardware comes out and the time distributions support it. That's pretty good if you think about. No one else is even close to Linux in supporting foreign hardware.
Having to pick a distro based on quick adoption of hardware is likely to make you unhappy in a whole lot of other areas. You may just want to wait.
'm betting the court would find the lack of an independant physical irrelevant to the procedings.
I disagree. Assume for a moment the suit was regarding a procedure. A party to the suit (the doctor) had you sign a contract waiving your right to sue while acting as your council. They would probably argue there is a conflict of interest. The same way a realtor has to be ultra careful if they represent buyer and seller or a lawyer can't even think of representing two sides.
Generally doesn't matter. It's up to the parties of the contract to get their own legal counsel to explain the contract.
Generally the courts have found that if one side cannot be reasonable be expected to understand a contract.... Further this is being one in a way to discourage the patient from getting council (in the doctor's office). If the doctor said something like "I want to mail these agreements to your attorney what's his address" and then the patient decided to sign on the spot...
Not actually the case. The arbitration company would be supposedly independent, and the client would be the insurance company.
And the insurance company is providing a service to the doctor who is a party to the suit. No court is going to see the insurance company or their chosen firm as neutral.
The whole thing is in market speak at this point. Who knows what a "strategic marketing agreement with Real Networks running Helix-Basic on the Xandros Server". It could just mean they agree to bundle or it could mean more.
And probably 0 courts in the United States would consider such a contract binding. "Yes your honor I agreed to binding arbitration on a medical practice without an independent physical, without my attorney being present without having this clauses clear explained, with no information provided about the arbitration proceedings with the arbitrator being an employee of the doctor..." Yep that's informed consent.
I don't know about this chip but on the thinkpads that wasn't the case. The chips were able to perform one encryption/decryption without actually knowing the proper prime (hardcoded). Lets assume they used RSA to make this easier: the chip knows pq, e. M^e is encrypted and (M^e)^f is decrypted. The chip has the (^f) function (mod pq) hardcoded in. It doesn't actually "know" f.
Lets say your harddrive is encrypted, there are 3 primes needed to be known to get to it:
one is generated by a pass-phrase one is stored on the harddrive (encypted using the chips built in) one is stored in the chip (semi permanent)
Only by using the chip can you read the hard drive.
Most (all?) security schemes don't work well against against someone being able to 1) take physical possession of your computers 2) hold your employees at gunpoint 3) cross check against physical assets 4) be able to repeat as often as needed
You can even go earlier than the 386 Desqview days:
By OS/2 1.1 (providing you were running an OS/2 BBS) you could get that performance. Microsoft own LAN product (which was really good) was OS/2 based. Of course Xenix wasn't bad either in driving hardware.
As for your idea of the WPS as a Linux WindowManager violates a core principle of application separation. There are lots of global variables in a WPS instance (including behaviors) so app A can change app B's whole desktop experience. That makes a great environment for customization and for cool hack apps but security is just terrible. This wasn't much a problem because most people weren't using several different OS/2 apps at a time (they had one at most and then used a bunch of Dos and Windows apps) if they had been....
You mentioned below you were talking about artists. Well for the artistic community:
1 - PDF rendering for the desktop (taking WYSIWYG one step further). This was something Lotus was moving in the direction of with AmiPro, that TeX compuset... is the master of that is extremely important for production print / graphics design
2 - System wide multimedia standard that is used for virtually every app fully integrated. Advantages are obvious
3 - High quality OpenGL (including kernel video card acceleration) for 3D graphics work
When the Mac came out there was no harddrive. A year and a half later the 10m harddrive was a $2000 option that almost no one got. Windows programs loaded into memory and rarely used the hard-drive until about windows 95. I'm pretty sure you could windows 286 to run off floppies if you tried. Systems like the amiga 500 didn't even have a harddrive.
Windows 3.1 networking was so bad that a 9600 bps modem could flood the stack unless you were single tasking. That's why OS/2 was so popular with the few people who had to use ethernet it was just impossible under Windows 3.1 (workgroups made this a little better). These notebooks have to be networkable.
First off it ran with 128megs of ram. What you are forgetting is that the "libraries" were in ROM on those computers. Also the fact is that these apps were being written in hand coded assembler. I wrote programs that did useful stuff that took less than 1k. But I had access to huge library of subroutines for free and I could easily count the bytes my code was using (and I mean that literally).
We are talking 128 megs. Why run OpenOffice on that? OpenOffice offers a high end integrated office suite meant to be basically a clone of Microsoft Office Professional. The people who are paying $100 for a laptop would probably be better served by a low end office suite that was much more intuitive. I gotta figure they probably don't have all that much office experience.
The $100 laptop they can use expert system administrators to build their kernels. Heck they could hire RedHat and thus get Alan Cox to do it personally. Eric Raymond (who wrote some of the early kernel config programs) supports the $100 laptop project.
So no I'd say its as far this goes its the opposite end of the spectrum.
Unforunately too much open source software relies on recompilation and/or patching to get the setup you need, whereas the Windows equivalent software can be reconfigured in 10 seconds using a GUI.
That is just plain nonsense. Unix apps tend to be far more configurable since they utilize human readable config files and scripts. Windows programs conversely are difficult for the developers themselves to recompile with different settings due to the fragility of the Visual Studio environment.
Oracle 10 (I can address different versions if you like) uses a very complex strategy to determine what the best sequence is for a SQL computation:
1) Automatically generate statistics as you use tables 2) If those statistics are good enough then using the statistics estimate costs for various strategies 3) If those statistics are not good enough then use a rule based system
It used to be that hints were used for rule based. However there are used for times where statistics are not likely to reflect the query. There are examples given but its basically an engine override. Hopefully you know what you are doing when you override the statistical data.
Wow between a bad CNN article and a bad editor this article makes no sense at all. OK Oracle has been on Linux for years. The CNN article that got linked to is about Oracle porting their cluster file system over to Linux. This is a filesystem that is just a little faster because it removes some redundancy. To quote Oracle:
Cluster Filesystem Options for Running Oracle Oracle RAC technology already provides features such as load balancing, redundancy, failover, scalability, caching, and locking, so there is a duplication of functions that occurs when Oracle datafiles reside on a block device with a traditional Linux filesystem such as ext2/ext3. Performance decreases in this case because caching by Oracle as well as the filesystem drains memory resources. As of this writing, in addition to third-party cluster filesystems, there are four filesystem options for running Oracle RAC. They are, in order of recommendation by Oracle:
1) Oracle Automatic Storage Management 2) Oracle Cluster File System 3) Network Filesystem 4) Raw devices.
1) Easy to learn 2) Easy to use 3) Feature rich 4) Fast 5) Has integration with multiple languages 6) Has a good quality development environment
I terms of criteria like (1) and (2) IMHO QT is the single best one out there. I think I dealt with the thousands of dollars vs. business model in the parent, there really is only a fairly narrow group of people that can't either go with one of the 5 options I listed. .
Or maybe they are trying to get Linus to understand that while he is "god" as far as the one particular piece of software goes when it comes to things that effect many hundreds of pieces of software he should try and act in a cooperative manner.
I'm strongly of the opinion that insurance companies started counter suing these idiots for wrongful suit. Once insurance companies make it clear to ambulance chasing lawyers that they are perfectly willing to spend $300k to avoid settlement for $100k; and perfectly clear to people that they will spend $100k to take get a $50k judgement they won't get sued. I think they have been penny wise and pound foolish and created this situation.
I'd agree that doctors have substantially higher costs than other professions. My point was that $100k is not a good salary, and regarding your post that your stats were off.
It is much easier to get into a computer science post-doc program than med school.
The comparison would be a residency. And I don't know about that.
1) I think its probably harder to go to say MI in compsci than harvard in medicine 2) Without question retention rates are lower so it appears that getting the PhD is harder than getting the MD. It also takes longer 3) Getting the residency is probably about as hard as staying in 4) Getting the post docs seems about equal AFAIKT.
Slashdot Mirror
And what I'm telling you is that your not doing an apples to apples comparison. Try running a poorly supported piece of hardware under Windows (one where you can't just double click on some driver). You've never tried to use a non windows piece of hardware on a windows computer so you don't really have anything to compare your experience with Linux too. You might be able to get it sort of kind of work under limited conditions but not very smoothly.
You can far more easily get something to 1/2 work under Linux. 6 months from now it will be a different story because the drivers will be integrated and everything will just self install. If 3 Linux distributions don't support your video card than most likely your video card was added to a recent version of X, for example.
Don't forget the hardware you are talking about was built for windows and shipped with windows. If you bought an AMD Sun you wouldn't be shocked if you had driver problems under Windows. Pretty much there is a 6 mo delay between the time hardware comes out and the time distributions support it. That's pretty good if you think about. No one else is even close to Linux in supporting foreign hardware.
Having to pick a distro based on quick adoption of hardware is likely to make you unhappy in a whole lot of other areas. You may just want to wait.
'm betting the court would find the lack of an independant physical irrelevant to the procedings.
I disagree. Assume for a moment the suit was regarding a procedure. A party to the suit (the doctor) had you sign a contract waiving your right to sue while acting as your council. They would probably argue there is a conflict of interest. The same way a realtor has to be ultra careful if they represent buyer and seller or a lawyer can't even think of representing two sides.
Generally doesn't matter. It's up to the parties of the contract to get their own legal counsel to explain the contract.
Generally the courts have found that if one side cannot be reasonable be expected to understand a contract.... Further this is being one in a way to discourage the patient from getting council (in the doctor's office). If the doctor said something like "I want to mail these agreements to your attorney what's his address" and then the patient decided to sign on the spot...
Not actually the case. The arbitration company would be supposedly independent, and the client would be the insurance company.
And the insurance company is providing a service to the doctor who is a party to the suit. No court is going to see the insurance company or their chosen firm as neutral.
The whole thing is in market speak at this point. Who knows what a "strategic marketing agreement with Real Networks running Helix-Basic on the Xandros Server". It could just mean they agree to bundle or it could mean more.
And probably 0 courts in the United States would consider such a contract binding. "Yes your honor I agreed to binding arbitration on a medical practice without an independent physical, without my attorney being present without having this clauses clear explained, with no information provided about the arbitration proceedings with the arbitrator being an employee of the doctor..." Yep that's informed consent.
I don't know about this chip but on the thinkpads that wasn't the case. The chips were able to perform one encryption/decryption without actually knowing the proper prime (hardcoded). Lets assume they used RSA to make this easier:
the chip knows pq, e. M^e is encrypted and (M^e)^f is decrypted. The chip has the (^f) function (mod pq) hardcoded in. It doesn't actually "know" f.
Lets say your harddrive is encrypted, there are 3 primes needed to be known to get to it:
one is generated by a pass-phrase
one is stored on the harddrive (encypted using the chips built in)
one is stored in the chip (semi permanent)
Only by using the chip can you read the hard drive.
Most (all?) security schemes don't work well against against someone being able to
1) take physical possession of your computers
2) hold your employees at gunpoint
3) cross check against physical assets
4) be able to repeat as often as needed
You can even go earlier than the 386 Desqview days:
By OS/2 1.1 (providing you were running an OS/2 BBS) you could get that performance. Microsoft own LAN product (which was really good) was OS/2 based. Of course Xenix wasn't bad either in driving hardware.
As for your idea of the WPS as a Linux WindowManager violates a core principle of application separation. There are lots of global variables in a WPS instance (including behaviors) so app A can change app B's whole desktop experience. That makes a great environment for customization and for cool hack apps but security is just terrible. This wasn't much a problem because most people weren't using several different OS/2 apps at a time (they had one at most and then used a bunch of Dos and Windows apps) if they had been....
You mentioned below you were talking about artists. Well for the artistic community:
1 - PDF rendering for the desktop (taking WYSIWYG one step further). This was something Lotus was moving in the direction of with AmiPro, that TeX compuset... is the master of that is extremely important for production print / graphics design
2 - System wide multimedia standard that is used for virtually every app fully integrated. Advantages are obvious
3 - High quality OpenGL (including kernel video card acceleration) for 3D graphics work
etc...
I think I may be being stupid here but what do you mean by a "retrenchment strategy"?
As far as I can tell they did find 2 niches.
1) A dedicated Real Media Server. There aren't good options on Linux because RedHat, Suse etc... aren't in bed with Real
2) Integration with Xandros desktop management tools.
Windows, Amiga, and Apple
When the Mac came out there was no harddrive. A year and a half later the 10m harddrive was a $2000 option that almost no one got. Windows programs loaded into memory and rarely used the hard-drive until about windows 95. I'm pretty sure you could windows 286 to run off floppies if you tried. Systems like the amiga 500 didn't even have a harddrive.
Windows 3.1 networking was so bad that a 9600 bps modem could flood the stack unless you were single tasking. That's why OS/2 was so popular with the few people who had to use ethernet it was just impossible under Windows 3.1 (workgroups made this a little better). These notebooks have to be networkable.
First off it ran with 128megs of ram. What you are forgetting is that the "libraries" were in ROM on those computers. Also the fact is that these apps were being written in hand coded assembler. I wrote programs that did useful stuff that took less than 1k. But I had access to huge library of subroutines for free and I could easily count the bytes my code was using (and I mean that literally).
We are talking 128 megs. Why run OpenOffice on that? OpenOffice offers a high end integrated office suite meant to be basically a clone of Microsoft Office Professional. The people who are paying $100 for a laptop would probably be better served by a low end office suite that was much more intuitive. I gotta figure they probably don't have all that much office experience.
The $100 laptop they can use expert system administrators to build their kernels. Heck they could hire RedHat and thus get Alan Cox to do it personally. Eric Raymond (who wrote some of the early kernel config programs) supports the $100 laptop project.
So no I'd say its as far this goes its the opposite end of the spectrum.
Unforunately too much open source software relies on recompilation and/or patching to get the setup you need, whereas the Windows equivalent software can be reconfigured in 10 seconds using a GUI.
That is just plain nonsense. Unix apps tend to be far more configurable since they utilize human readable config files and scripts. Windows programs conversely are difficult for the developers themselves to recompile with different settings due to the fragility of the Visual Studio environment.
Oracle 10 (I can address different versions if you like) uses a very complex strategy to determine what the best sequence is for a SQL computation:
1) Automatically generate statistics as you use tables
2) If those statistics are good enough then using the statistics estimate costs for various strategies
3) If those statistics are not good enough then use a rule based system
It used to be that hints were used for rule based. However there are used for times where statistics are not likely to reflect the query. There are examples given but its basically an engine override. Hopefully you know what you are doing when you override the statistical data.
Well he's taking a softline on DRM which means a hardline on the GPLv3.
Fair enough by good I meant:
1) Easy to learn
2) Easy to use
3) Feature rich
4) Fast
5) Has integration with multiple languages
6) Has a good quality development environment
I terms of criteria like (1) and (2) IMHO QT is the single best one out there. I think I dealt with the thousands of dollars vs. business model in the parent, there really is only a fairly narrow group of people that can't either go with one of the 5 options I listed. .
Or maybe they are trying to get Linus to understand that while he is "god" as far as the one particular piece of software goes when it comes to things that effect many hundreds of pieces of software he should try and act in a cooperative manner.
I'm strongly of the opinion that insurance companies started counter suing these idiots for wrongful suit. Once insurance companies make it clear to ambulance chasing lawyers that they are perfectly willing to spend $300k to avoid settlement for $100k; and perfectly clear to people that they will spend $100k to take get a $50k judgement they won't get sued. I think they have been penny wise and pound foolish and created this situation.
I'd agree that doctors have substantially higher costs than other professions. My point was that $100k is not a good salary, and regarding your post that your stats were off.
Agreed. Lets say total debt is $400k that's worth about $40k / yr in salary.
OK I didn't know about Inkwell. Point taken that isn't so hard.
It is much easier to get into a computer science post-doc program than med school.
The comparison would be a residency. And I don't know about that.
1) I think its probably harder to go to say MI in compsci than harvard in medicine
2) Without question retention rates are lower so it appears that getting the PhD is harder than getting the MD. It also takes longer
3) Getting the residency is probably about as hard as staying in
4) Getting the post docs seems about equal AFAIKT.
So do you have any evidence here?