ReiserFS was especially designed for file systems with many small files, such as the/etc directory, a directory of text files, a web hosting directory, etc.
To imply it's obsolete because it may not be the best at uses other than those it was designed for does not work. You also have to show that the types of uses for which is was designed are no longer around.
Any admin who knew ahead of time the use for a system (which might not be the case for a desktop, granted, but should be for a server) would not blindly use one FS across all partitions. Either you'd weigh the pros and cons of a single general-purpose FS like ext2/ext3 carefully or you'd split your partitions among different FS types based on usage.
If they thought it was worthwhile but that the main project was nice enough, they wouldn't bother forking. Why pa to maintain it yourself when someone else is happily doing the work?
Hans going to prison is a game changer. Whether that means his FS doesn't get used or that someone else takes over development, his conviction and sentence will require changes.
Not just Customs, but the TSA will have them too. Everyone flying into the country, according to the Ninth Circuit, is subject to a search of their laptops.
Well, Microsoft and SCO have already tried their damnedest to make every Linux user an actual criminal for using MS and SCO patents and copyrighted code without permission. Not that they've been able to prove anything, nor that SCO actually ended up owning any patents or the code in question at all.
Well, if the hot patch went smoothly on all of the first 10%, then you could probably roll it out to the other 90% in a second step.
One could also have a four to ten node testing cluster that gets regularly load tested and intrusion tested which receives any updates before the big production cluster. That way, you test it on the testing cluster first, and can roll out to as many machines as you can afford to have down as the first part of the production update. Then, the rest of the systems could get the update.
You're still looking at a much faster turnaround time doing three steps with no lag as servers go idle than ten steps with that delay. I wouldn't suggest making a habit of hot-patching a running kernel just because it's faster, though. Weigh the benefits and risks.
My issue is that I've heard doctors say the patients demand the newest, best-marketed treatment even when older, cheaper drugs work just as well with the same or fewer side effects. If that's a major pattern, that's a problem.
Another problem is that doctors and their staffs spend a lot of time seeing patients and handling paperwork. They don't read the medical journals and spend hours debating like we'd like to think they do. Many of them get little more than the one-sheet of the tri-fold glossy for a new drug explaining the benefits and risks with percentages of side effect incidence and contraindications. Then they get free samples of the new drug, and they see that most of their patients do well on it.
Many decisions get made without considering that some patients do better on diuretics for blood pressure than the latest patented pills. Some people would rather have organizational skills training specifically for those with ADHD than be on a stimulant, but they don't realize their insurance will cover such a thing with a prescription.
My favorites are the drugs that have warnings that sound like coincidence. "Side effects include headache, fever, runny nose, coughing, and nausea, but were comparable to placebo" -- for a flu or common cold medicine. Well, gee, if you're treating an infection and the side effects sound an awful lot like the symptoms, then perhaps there's reason to think the side-effects aren't that important.
The support information usually is free both in the free speech and free beer senses. The problem is just that a mediocre man page and some sparse comments in the code is not want many people want. You're not paying technical support people and consultants to quote the manual to you. You're paying them for the fact that they actually know the information already and understand it better than you.
At least, that's the theory. Some of them are clueless, too, but hopefully those don't stick around in the business long.
If you didn't have the first customer paying you to implement it, you wouldn't have a spec designed by a target user. You wouldn't have a testing base that's paying you for the testing process. You wouldn't have someone to help you with the interface usability for free.
You'd be implementing in the dark, making your best guess what a potential customer might want and how they might want it. Then, you'd be lucky to get any customers.
Charging $1000 for the time to implement something and then charging $5 per copy or even $1 afterwards is better than selling it to the first user for $1000 as a work for hire, then not being allowed to make any more money. Lots of custom programming clients won't let you turn around and sell the same software at all. Many will negotiate with you that they'll let you maintain the copyright and license it as open source, although some will want a price break for that.
So, are you looking to get compensated for your time at an hourly rate better than what you'd get working in some IT department swapping backup tapes when you're not needed as a programmer and still be able to sell copies on the cheap, or are you wanting to take a chance of being a billionaire based on a product of which you may never sell a single copy?
No. Just... no. Let's use Slashdot's favorite line of analogies here.
A computer program is like the design of a car. There is no real-world material cost to the design. There's some initial layout of funds for the labor of the designers, different molds get made, and there are sometimes tweaks to the assembly line machines for different parts. Yet it's the building of the actual car that takes up the vast bulk of the labor and almost all the materials per unit.
The car companies don't sue when Ford builds a car that fits the same market segment as a Chevy or when Saturn releases a car that's almost visually indistinguishable from an Accord. The new cars work differently (although very similarly), sell through different dealers, and all compete for the same markets. They don't sue when you sell, buy, or install after-market parts or when you build another car (Shelby, Callaway, Monster Miata, etc) based on theirs. They know you're never going to take the bulk of their business, because to mass produce cars is an expensive proposition. Sure, they patent something once in a while, but that's typically for a specific way to do something, and not a general "saving fuel by running a leaner mixture" or something that's common sense.
What Bill Gates wants is to design the software, and get paid like a car for each copy. The whole basis of the Open Source movement and the FSF's Free Software movement, though, is that the duplication of software isn't at all like the duplication of a car. Once the software's written, the real cost is over. Making a copy is cheap.
So Bill Gate's problem with Open Source software, and Free Software in particular, is that people are using the advantage of cheap reproduction to move the compensation for software to the place where the cost is. They're charging to implement something, and then saying you can do with it what you want. He'd rather they keep charging as if there's significant cost associated with each copy, because that market practice is what makes him so rich.
In the US a doctor typically can write a prescription for a specific brand-name drug, a drug with the same active ingredients that's generic (if available), or any equivalent drug. The more options the doctor allows, the more options the pharmacist and the patient can discuss.
Many doctors I've been to will discuss the drug options with the patient and will prescribe an alternative drug or give the pharmacist an option if there's not a compelling reason to choose one drug over another.
The best way to get more options in prescriptions is to get a better doctor. However, some prescription drugs have side effect risks or other drawbacks that keep certain doctors away from them altogether.
His problem is he doesn't understand "customization", "support", or "good documentation" as competitive advantages, but anyone who's been a long-time customer of Microsoft should be able to tell you that.
If the license is invalid, there is no point in using it. The whole goal of arguing that it's invalid is to keep people from being able to use it. If the courts would have agreed, the license would be useless and someone would have had to rewrite the sections the court invalidated.
Maybe 2008 will be the year of the home server garden for people who have a bunch of free time.
The year of the home desktop garden is still years away, because Aunt Gertrude still can't grow her own squash and because not everyone's willing to pull weeds and read books on how to get rid of pests for themselves.
Probably, but how many? 10%? Your average upgrade time per box is 2.5 hours and the whole process takes 5 hours that way (assuming the same time to get the boxes idle and upgraded). It's an improvement, but it's still not as good as running the patch against all them at once without rebooting.
Sure, there are drawbacks either way, but being able to patch a running kernel is an option that may make sense. If the kernels so patched can be kept up a few weeks, then you have plenty of time to schedule other maintenance in the meantime.
Let me be clear that I'm not advocating this for every patch. Most machines don't have anything exploitable facing the world directly anyway thanks to firewalls, intrusion detection systems, closed ports on the systems themselves, good routing policies, etc. If you have machines that are being used as firewalls, routers, switches, or other infrastructure, though,you might be less comfortable with a long turnaround time for a critical security patch than with patching the kernel while it's running.
If you're using dictionary words and punctuation, that's going to be easier to find automatically for the same number of characters. More characters does make the problem harder, though, so it's a tradeoff.
If there's something so subtly wrong with your system that it fails from this, what's to say it won't test fine after rebooting and then fail when placed under load?
I hear Steve Jobs gets bitchy whenever someone says Woz was the brains. Oh, and I bet Trump gets pretty upset whenever people ask if the curtains match the Venetian blinds...
Slashdot Mirror
ReiserFS was especially designed for file systems with many small files, such as the /etc directory, a directory of text files, a web hosting directory, etc.
To imply it's obsolete because it may not be the best at uses other than those it was designed for does not work. You also have to show that the types of uses for which is was designed are no longer around.
Any admin who knew ahead of time the use for a system (which might not be the case for a desktop, granted, but should be for a server) would not blindly use one FS across all partitions. Either you'd weigh the pros and cons of a single general-purpose FS like ext2/ext3 carefully or you'd split your partitions among different FS types based on usage.
Oh, come on. Dating a confessed serial killer doesn't have a chance to create a reasonable doubt about someone's disappearance?
"irrelevant" means there's no way something makes a difference. To the contrary, I think you'd have to work very hard to dispel that doubt.
If they thought it was worthwhile but that the main project was nice enough, they wouldn't bother forking. Why pa to maintain it yourself when someone else is happily doing the work?
Hans going to prison is a game changer. Whether that means his FS doesn't get used or that someone else takes over development, his conviction and sentence will require changes.
What would be great would be to find out it's Knoppix or Trinity running ntcrack and such with no source.
Not just Customs, but the TSA will have them too. Everyone flying into the country, according to the Ninth Circuit, is subject to a search of their laptops.
Well, Microsoft and SCO have already tried their damnedest to make every Linux user an actual criminal for using MS and SCO patents and copyrighted code without permission. Not that they've been able to prove anything, nor that SCO actually ended up owning any patents or the code in question at all.
I'd say it's innovative to reimplement an entire OS from published specs rather than buying a source license like everyone else.
It was innocative reimplementing it all from published standards instead of buying a source license like everyone else was doing?
Well, if the hot patch went smoothly on all of the first 10%, then you could probably roll it out to the other 90% in a second step.
One could also have a four to ten node testing cluster that gets regularly load tested and intrusion tested which receives any updates before the big production cluster. That way, you test it on the testing cluster first, and can roll out to as many machines as you can afford to have down as the first part of the production update. Then, the rest of the systems could get the update.
You're still looking at a much faster turnaround time doing three steps with no lag as servers go idle than ten steps with that delay. I wouldn't suggest making a habit of hot-patching a running kernel just because it's faster, though. Weigh the benefits and risks.
My issue is that I've heard doctors say the patients demand the newest, best-marketed treatment even when older, cheaper drugs work just as well with the same or fewer side effects. If that's a major pattern, that's a problem.
Another problem is that doctors and their staffs spend a lot of time seeing patients and handling paperwork. They don't read the medical journals and spend hours debating like we'd like to think they do. Many of them get little more than the one-sheet of the tri-fold glossy for a new drug explaining the benefits and risks with percentages of side effect incidence and contraindications. Then they get free samples of the new drug, and they see that most of their patients do well on it.
Many decisions get made without considering that some patients do better on diuretics for blood pressure than the latest patented pills. Some people would rather have organizational skills training specifically for those with ADHD than be on a stimulant, but they don't realize their insurance will cover such a thing with a prescription.
My favorites are the drugs that have warnings that sound like coincidence. "Side effects include headache, fever, runny nose, coughing, and nausea, but were comparable to placebo" -- for a flu or common cold medicine. Well, gee, if you're treating an infection and the side effects sound an awful lot like the symptoms, then perhaps there's reason to think the side-effects aren't that important.
Yeah, so many doctors going to Broadway shows at $120 a ticket in Lesotho and Cambodia really pisses me off!
The support information usually is free both in the free speech and free beer senses. The problem is just that a mediocre man page and some sparse comments in the code is not want many people want. You're not paying technical support people and consultants to quote the manual to you. You're paying them for the fact that they actually know the information already and understand it better than you.
At least, that's the theory. Some of them are clueless, too, but hopefully those don't stick around in the business long.
If you didn't have the first customer paying you to implement it, you wouldn't have a spec designed by a target user. You wouldn't have a testing base that's paying you for the testing process. You wouldn't have someone to help you with the interface usability for free.
You'd be implementing in the dark, making your best guess what a potential customer might want and how they might want it. Then, you'd be lucky to get any customers.
Charging $1000 for the time to implement something and then charging $5 per copy or even $1 afterwards is better than selling it to the first user for $1000 as a work for hire, then not being allowed to make any more money. Lots of custom programming clients won't let you turn around and sell the same software at all. Many will negotiate with you that they'll let you maintain the copyright and license it as open source, although some will want a price break for that.
So, are you looking to get compensated for your time at an hourly rate better than what you'd get working in some IT department swapping backup tapes when you're not needed as a programmer and still be able to sell copies on the cheap, or are you wanting to take a chance of being a billionaire based on a product of which you may never sell a single copy?
No. Just... no. Let's use Slashdot's favorite line of analogies here.
A computer program is like the design of a car. There is no real-world material cost to the design. There's some initial layout of funds for the labor of the designers, different molds get made, and there are sometimes tweaks to the assembly line machines for different parts. Yet it's the building of the actual car that takes up the vast bulk of the labor and almost all the materials per unit.
The car companies don't sue when Ford builds a car that fits the same market segment as a Chevy or when Saturn releases a car that's almost visually indistinguishable from an Accord. The new cars work differently (although very similarly), sell through different dealers, and all compete for the same markets. They don't sue when you sell, buy, or install after-market parts or when you build another car (Shelby, Callaway, Monster Miata, etc) based on theirs. They know you're never going to take the bulk of their business, because to mass produce cars is an expensive proposition. Sure, they patent something once in a while, but that's typically for a specific way to do something, and not a general "saving fuel by running a leaner mixture" or something that's common sense.
What Bill Gates wants is to design the software, and get paid like a car for each copy. The whole basis of the Open Source movement and the FSF's Free Software movement, though, is that the duplication of software isn't at all like the duplication of a car. Once the software's written, the real cost is over. Making a copy is cheap.
So Bill Gate's problem with Open Source software, and Free Software in particular, is that people are using the advantage of cheap reproduction to move the compensation for software to the place where the cost is. They're charging to implement something, and then saying you can do with it what you want. He'd rather they keep charging as if there's significant cost associated with each copy, because that market practice is what makes him so rich.
In the US a doctor typically can write a prescription for a specific brand-name drug, a drug with the same active ingredients that's generic (if available), or any equivalent drug. The more options the doctor allows, the more options the pharmacist and the patient can discuss.
Many doctors I've been to will discuss the drug options with the patient and will prescribe an alternative drug or give the pharmacist an option if there's not a compelling reason to choose one drug over another.
The best way to get more options in prescriptions is to get a better doctor. However, some prescription drugs have side effect risks or other drawbacks that keep certain doctors away from them altogether.
His problem is he doesn't understand "customization", "support", or "good documentation" as competitive advantages, but anyone who's been a long-time customer of Microsoft should be able to tell you that.
If the license is invalid, there is no point in using it. The whole goal of arguing that it's invalid is to keep people from being able to use it. If the courts would have agreed, the license would be useless and someone would have had to rewrite the sections the court invalidated.
Maybe 2008 will be the year of the home server garden for people who have a bunch of free time.
The year of the home desktop garden is still years away, because Aunt Gertrude still can't grow her own squash and because not everyone's willing to pull weeds and read books on how to get rid of pests for themselves.
Probably, but how many? 10%? Your average upgrade time per box is 2.5 hours and the whole process takes 5 hours that way (assuming the same time to get the boxes idle and upgraded). It's an improvement, but it's still not as good as running the patch against all them at once without rebooting.
Sure, there are drawbacks either way, but being able to patch a running kernel is an option that may make sense. If the kernels so patched can be kept up a few weeks, then you have plenty of time to schedule other maintenance in the meantime.
Let me be clear that I'm not advocating this for every patch. Most machines don't have anything exploitable facing the world directly anyway thanks to firewalls, intrusion detection systems, closed ports on the systems themselves, good routing policies, etc. If you have machines that are being used as firewalls, routers, switches, or other infrastructure, though,you might be less comfortable with a long turnaround time for a critical security patch than with patching the kernel while it's running.
If you're using dictionary words and punctuation, that's going to be easier to find automatically for the same number of characters. More characters does make the problem harder, though, so it's a tradeoff.
Yeah, the jargon really is necessary to getting the point across. Everyone knows "quiesce", but nobody understands "quieten" or "simmer down". WTG!
If there's something so subtly wrong with your system that it fails from this, what's to say it won't test fine after rebooting and then fail when placed under load?
I hear Steve Jobs gets bitchy whenever someone says Woz was the brains. Oh, and I bet Trump gets pretty upset whenever people ask if the curtains match the Venetian blinds...
That still takes longer than just patching in place. Meanwhile, your systems have a critical security bug.