As for tested I meant 3rd party since that's what we are discussing. As an aside Microsoft is actually quite good at the Apple model and their software is mostly updated early.
As for Adobe... Adobe has all sorts of work arounds to low level hooks they had in old version of MacOS, the codebase is super fragile.... They are a terrible example. Adobe had huge problems with once in a while big changes as well.
Anyway you seem a bit bitter about Mail. I'm not sure what problems you've had. I find Mail rather stable and even use a 3rd party mail extension rather regularly. Similarly with Finder, though there the API shifts annually.
Why would continuous updates drive personal buyers away? Mac customers and iPhone customers like continuous updates. Windows customers don't mind continuous updates within versions. I don't see how this follows.
I buy a personal computer. Included with this computer is 1 year of Office 365. After that I pay $100 / year (or maybe a bit more) and I get updated OSes, plus file sharing, plus management from Azure for my cloud applications and accounts, plus the ability to use office, plus a secure environment for corporate documents from work, plus the virtual desktop support.... Or make it $35 / mo and I don't have to buy the personal computer at all, it is just included along with a hardware service plan...
I suspect most consumers would love a subscription model.
I don't agree with calling any of those badly designed. It was painful for early adopters and there were problems but ultimately the design of Vista was good, there were just bugs and incompatibilities. Vista was an important shift that made Windows 7 possible.
Windows 8 I think was an excellent idea. Microsoft just chickened out in not making touchscreen and movable hinges on laptops mandatory. The problem Microsoft had was that they sold a transitional OS designed to ease the transition to a new generation of hardware as just "the next version". Also if they didn't want to introduce the extra cost, it should never have been offered to the bottom of the market.
They care about what you are willing to pay for not what you like. Under the current model, you not upgrading means you not buying. If they can move you over to a service model instead of say $30 every 4 years you could be at $5/mo. Even if they lose 10,20,30,+ percent it is still worth it.
1) Apple releases a beta of the new OS 2) The people who write 3rd party software test against the OS, and if needed release a minor version upgrade 3) End user upgrades the minor version of their software automatically 4) When the OS is released all the software is compatible.
I think Microsoft is going to try and drive most of the small business software towards a continuous distribution model tied to Azure. They are also moving small business towards a managed services model. They are going to be arguably even better supported.
It is a huge cultural change but a doable cultural change.
That's pretty much how XNU (OSX) evolved. Microkernel with monolithic for some stuff. Though once in it gets harder to pull it out again so the movement has been mostly one way. My point was Hurd never made those compromises, so it wasn't a viable kernel.
how are they going to keep up with major Windows changes every few months?
They are going to have a full time compatibility group as part of their helpdesk. That's what was done in the 1990s, helpdesk was always working on the next version of upgrades and the staffed around it. Enterprises had to be upgrading their upgrading their applications regularly. The staff (remember this was staff not consultants) associated with the internal applications had to be prepping for the next versions and removing compatibility. Microsoft's stability after XP allowed their customer base to reduce their spending. They have created a culture around their operating system which is unable and unwilling to absorb even relatively minor changes. If they go to a model of continuous upgrades that disappears, change becomes the norm.
It is not undoable, enterprises just spend more and they get more.
I don't know. Say Microkernel costs you 60% of performance. You can always go microkernel and accept rolling back 5 years or so on CPU. But at the same time why pay that much performance wise for a different kernel design.
I suspect we are a long long way off from microkernel. Possibly the next entire generation of server OSes might do it since they will be running virtual compute on virtual machines.... with everything being wrapped the micro-kernel might be cheaper.
Not sure. 1994 the Open Source community was pretty small. I got support. The BSDs were pretty friendly back then too. But it is a good story about the community.
Other than Xenix what do you mean by Microsoft. Also you forgot SCO if you are including commercial Unixes for 386. Also one that gets forgotten about but was quite good in those early days was: http://en.wikipedia.org/wiki/C...
Your years are off. Around the time of the 386SX I had 20megs of RAM in my 386. Hard drives were hundreds of megs. $1000k 20mb drive is like 1985 or so. 1990 you are at like $10 / meg for storage and 1995 around $2 / meg.
Hurd was never a viable kernel. It never did much. Moreover micro-kernel design didn't work very well. QNX is really the operating system that got micro-kernels working well and they took it in non-desktop directions. On the XNU (OSX kernel) the kernel has moved more and more monolithic for performance, speed matters. Citrix is doing some interesting work on microkernels so we'll see.
Ultimately though, it is just too expensive to add even a few extra simple operations to most kernel functions. And that's why Hurd lost.
Fundamentally the BSDs and Linux were on par with one another in the mid 1990s. BSD386 was getting sued and that cost a year but it started ahead. Where Linux thrived though was it recruited from Unix users and Windows powerusers. Linux focused on the ability of Unix users the ability to run Unix at home, and focused on the ability for not particularly good system admins to setup servers. It aimed for ease of use. The BSDs conversely aimed to offer a Unix like environment for Intel/Western Digital Hardware; a free version of SCO.
Then of course came the licensing issue. Linux was already ahead by 1997. But GPL allowed companies whose primary goal was not to sell software (or at least the software in question) to cooperate safely. It turned out those companies were larger supporters of free software than companies who were making extensions to base packages.
I think the reason the Linux kernel was successful was that
a) It was very good on par with the BDS kernels originally b) It was tied to the Linux culture. c) The GPL
Linux culture beat the BSD culture. The GPL beat the BSD license. The kernel just went along for the ride originally, though of course it had to be good enough to not hamper Linux-OS. After the initial ride the commercial interest led to greater development for the Linux kernel than other kernels which led to it fitting more uses better and more commercial development and interest.... A self feeding cycle that does make the kernel a winning point for Linux-OS.
The 90% bracket was on individuals. The corporate tax rate was never that high.
Most likely what you are concerned about is that capital gains aren't taxed very much so it pays for corporations to hold income and not pass it on as dividends. Things like: taxing assets mark to market (i.e. you pay on what the asset is worth each year), lower unearned income taxes along with higher capital gains taxes would get the effect you are looking for.
Why the hell should companies, including tax payers (costs of running courts & all) have to pay for the USPTO's fuckups?
There are two different issues here.
For the USPTO to vet patents properly would require a substantially higher cost per patent. The tax payers through their elected representative did not allow for patent fees to go high enough to cover that cost nor subsidies to cover that cost. They are the most responsible party for the policies.
As for companies. Companies don't have to pay for the USPTO. What they do have is a situation where their patents are registered but unvetted. They have to understand what they bought. They didn't buy much more than a filing.
Either the USPTO didn't do their job right (incompetence gets you fired in the real world, but not if you're a bureaucrat apparently)
In the "real world" offering a lower quality product at a much better price is applauded. It doesn't get you fired it makes you rich.
#1 Richest man Bill Gates got their for cutting the cost of desktop software. #2 Warren Buffet got their for cutting the cost of running insurance companies #3 Larry Ellison got their for reducing administrative expenses
etc...
What does often get people fired in the real world is blaming others rather than owning your mistakes. If people want a better patent system where the USPTO vets patents rather than registers them, they should pay for that system and stop pretend the reason they aren't getting such a system for 95% off is because patent examiners are being lazy.
So far you've been wrong about just about everything. I know you think that prefacing nonsense with insults makes it an argument but it isn't.
Mainframes do precisely what you claim cannot be done and have done since long before Unix. So your categorical assertions of what can and can't be done are simply and obviously provably false. The fact that rather than admit this and engage you continue to be rude proves that you lack character as well as knowledge.
How can a service handle a situation when it is down? The services have to register in advance how to handle things. Moreover other services might still have issues.
B depends on C and C needs to reinitiate with B, but D is also talking to C. How does the new B signal C?
As for it being contrived that's one of the key issues in process management how to handle chains and stacks of processes. That doesn't happen much in the sysv world because sysv handles it so badly that everything ended up having to write its own process manager.
Mixed with the sarcasm and the editorial I'm having a tough time extracting the position you are advocating. Or more importantly what you are disagreeing with. The point above is that you can't sue patent examiners. Your position seems to be that patents should be subject to a more complex and hostile review (i.e. a much more expensive review). If that's your point I'd agree. Otherwise you are going to have to represent.
In 2011 Cray deployed a Linux supercomputer with 9600 GPUs and from that point on Linux has been able to handle essentially infinitely many. I think you may want to read up a bit.
First off you sound like you would be happier on BSD. You should however get a grip. Non systemd distributions exist.
The upsides of Linux in 2015 are in spaces like huge range of software, IaaS, supercomputing and embedded. If you want to run 1990s style traditional Unix servers with software available for BSD, BSD is probably the better choice.
Slashdot Mirror
As for tested I meant 3rd party since that's what we are discussing. As an aside Microsoft is actually quite good at the Apple model and their software is mostly updated early.
As for Adobe... Adobe has all sorts of work arounds to low level hooks they had in old version of MacOS, the codebase is super fragile.... They are a terrible example. Adobe had huge problems with once in a while big changes as well.
Anyway you seem a bit bitter about Mail. I'm not sure what problems you've had. I find Mail rather stable and even use a 3rd party mail extension rather regularly. Similarly with Finder, though there the API shifts annually.
Well if we are talking non paying customers then who cares if they are driven away?
Why would continuous updates drive personal buyers away? Mac customers and iPhone customers like continuous updates. Windows customers don't mind continuous updates within versions. I don't see how this follows.
I buy a personal computer. Included with this computer is 1 year of Office 365. After that I pay $100 / year (or maybe a bit more) and I get updated OSes, plus file sharing, plus management from Azure for my cloud applications and accounts, plus the ability to use office, plus a secure environment for corporate documents from work, plus the virtual desktop support .... Or make it $35 / mo and I don't have to buy the personal computer at all, it is just included along with a hardware service plan...
I suspect most consumers would love a subscription model.
I don't agree with calling any of those badly designed. It was painful for early adopters and there were problems but ultimately the design of Vista was good, there were just bugs and incompatibilities. Vista was an important shift that made Windows 7 possible.
Windows 8 I think was an excellent idea. Microsoft just chickened out in not making touchscreen and movable hinges on laptops mandatory. The problem Microsoft had was that they sold a transitional OS designed to ease the transition to a new generation of hardware as just "the next version". Also if they didn't want to introduce the extra cost, it should never have been offered to the bottom of the market.
They care about what you are willing to pay for not what you like. Under the current model, you not upgrading means you not buying. If they can move you over to a service model instead of say $30 every 4 years you could be at $5/mo. Even if they lose 10,20,30,+ percent it is still worth it.
Apple does this today it is not hard.
1) Apple releases a beta of the new OS
2) The people who write 3rd party software test against the OS, and if needed release a minor version upgrade
3) End user upgrades the minor version of their software automatically
4) When the OS is released all the software is compatible.
I think Microsoft is going to try and drive most of the small business software towards a continuous distribution model tied to Azure. They are also moving small business towards a managed services model. They are going to be arguably even better supported.
It is a huge cultural change but a doable cultural change.
That's pretty much how XNU (OSX) evolved. Microkernel with monolithic for some stuff. Though once in it gets harder to pull it out again so the movement has been mostly one way. My point was Hurd never made those compromises, so it wasn't a viable kernel.
They are going to have a full time compatibility group as part of their helpdesk. That's what was done in the 1990s, helpdesk was always working on the next version of upgrades and the staffed around it. Enterprises had to be upgrading their upgrading their applications regularly. The staff (remember this was staff not consultants) associated with the internal applications had to be prepping for the next versions and removing compatibility. Microsoft's stability after XP allowed their customer base to reduce their spending. They have created a culture around their operating system which is unable and unwilling to absorb even relatively minor changes. If they go to a model of continuous upgrades that disappears, change becomes the norm.
It is not undoable, enterprises just spend more and they get more.
I don't know. Say Microkernel costs you 60% of performance. You can always go microkernel and accept rolling back 5 years or so on CPU. But at the same time why pay that much performance wise for a different kernel design.
I suspect we are a long long way off from microkernel. Possibly the next entire generation of server OSes might do it since they will be running virtual compute on virtual machines.... with everything being wrapped the micro-kernel might be cheaper.
Not sure. 1994 the Open Source community was pretty small. I got support. The BSDs were pretty friendly back then too. But it is a good story about the community.
Other than Xenix what do you mean by Microsoft. Also you forgot SCO if you are including commercial Unixes for 386. Also one that gets forgotten about but was quite good in those early days was: http://en.wikipedia.org/wiki/C...
Your years are off. Around the time of the 386SX I had 20megs of RAM in my 386. Hard drives were hundreds of megs. $1000k 20mb drive is like 1985 or so. 1990 you are at like $10 / meg for storage and 1995 around $2 / meg.
Hurd was never a viable kernel. It never did much. Moreover micro-kernel design didn't work very well. QNX is really the operating system that got micro-kernels working well and they took it in non-desktop directions. On the XNU (OSX kernel) the kernel has moved more and more monolithic for performance, speed matters. Citrix is doing some interesting work on microkernels so we'll see.
Ultimately though, it is just too expensive to add even a few extra simple operations to most kernel functions. And that's why Hurd lost.
Fundamentally the BSDs and Linux were on par with one another in the mid 1990s. BSD386 was getting sued and that cost a year but it started ahead. Where Linux thrived though was it recruited from Unix users and Windows powerusers. Linux focused on the ability of Unix users the ability to run Unix at home, and focused on the ability for not particularly good system admins to setup servers. It aimed for ease of use. The BSDs conversely aimed to offer a Unix like environment for Intel/Western Digital Hardware; a free version of SCO.
Then of course came the licensing issue. Linux was already ahead by 1997. But GPL allowed companies whose primary goal was not to sell software (or at least the software in question) to cooperate safely. It turned out those companies were larger supporters of free software than companies who were making extensions to base packages.
I think the reason the Linux kernel was successful was that
a) It was very good on par with the BDS kernels originally
b) It was tied to the Linux culture.
c) The GPL
Linux culture beat the BSD culture. The GPL beat the BSD license. The kernel just went along for the ride originally, though of course it had to be good enough to not hamper Linux-OS. After the initial ride the commercial interest led to greater development for the Linux kernel than other kernels which led to it fitting more uses better and more commercial development and interest.... A self feeding cycle that does make the kernel a winning point for Linux-OS.
I've done it. When the CD is $14+ and the .mp3 is $5. At that spread I'll take the digital only option. Also for single songs.
The 90% bracket was on individuals. The corporate tax rate was never that high.
Most likely what you are concerned about is that capital gains aren't taxed very much so it pays for corporations to hold income and not pass it on as dividends. Things like: taxing assets mark to market (i.e. you pay on what the asset is worth each year), lower unearned income taxes along with higher capital gains taxes would get the effect you are looking for.
An example of what? That patent is legal because of bad patent law. That's not the USPTO that's congress.
There are two different issues here.
For the USPTO to vet patents properly would require a substantially higher cost per patent. The tax payers through their elected representative did not allow for patent fees to go high enough to cover that cost nor subsidies to cover that cost. They are the most responsible party for the policies.
As for companies. Companies don't have to pay for the USPTO. What they do have is a situation where their patents are registered but unvetted. They have to understand what they bought. They didn't buy much more than a filing.
In the "real world" offering a lower quality product at a much better price is applauded. It doesn't get you fired it makes you rich.
#1 Richest man Bill Gates got their for cutting the cost of desktop software.
#2 Warren Buffet got their for cutting the cost of running insurance companies
#3 Larry Ellison got their for reducing administrative expenses
etc...
What does often get people fired in the real world is blaming others rather than owning your mistakes. If people want a better patent system where the USPTO vets patents rather than registers them, they should pay for that system and stop pretend the reason they aren't getting such a system for 95% off is because patent examiners are being lazy.
So far you've been wrong about just about everything. I know you think that prefacing nonsense with insults makes it an argument but it isn't.
Mainframes do precisely what you claim cannot be done and have done since long before Unix. So your categorical assertions of what can and can't be done are simply and obviously provably false. The fact that rather than admit this and engage you continue to be rude proves that you lack character as well as knowledge.
How can a service handle a situation when it is down? The services have to register in advance how to handle things. Moreover other services might still have issues.
B depends on C and C needs to reinitiate with B, but D is also talking to C. How does the new B signal C?
As for it being contrived that's one of the key issues in process management how to handle chains and stacks of processes. That doesn't happen much in the sysv world because sysv handles it so badly that everything ended up having to write its own process manager.
Mixed with the sarcasm and the editorial I'm having a tough time extracting the position you are advocating. Or more importantly what you are disagreeing with. The point above is that you can't sue patent examiners. Your position seems to be that patents should be subject to a more complex and hostile review (i.e. a much more expensive review). If that's your point I'd agree. Otherwise you are going to have to represent.
http://www.top500.org/statisti...
In 2011 Cray deployed a Linux supercomputer with 9600 GPUs and from that point on Linux has been able to handle essentially infinitely many. I think you may want to read up a bit.
RedHat has had their own kernel for a quarter century. They don't need Linus' permission to introduce change into their kernel.
First off you sound like you would be happier on BSD. You should however get a grip. Non systemd distributions exist.
The upsides of Linux in 2015 are in spaces like huge range of software, IaaS, supercomputing and embedded. If you want to run 1990s style traditional Unix servers with software available for BSD, BSD is probably the better choice.
Wow wish I could mod you up. Your last paragraph is spot on.