I've had various VIA, ALi and SiS chipsets which had a poor reputation, and yet had very good stability under Linux... Similarly, Cyrix CPUs had a reputation for causing stability problems, and yet i had several cyrix based machines running for years without incident. One of the cyrix systems was given to me by someone who got sick of it's instability while running windows. His replacement system fared little better.
I'd say it was more down to drivers than hardware, windows systems often end up with a horrible mess of shoddily written drivers from multiple third parties that can often conflict with other, whereas linux *usually* has all the drivers built together in the same kernel. The only times i've had issues with linux were due to out of kernel drivers of questionable quality.
Perhaps, or perhaps the cpu/motherboard will just be thought of as a single component instead of discrete ones... You choose one, and then choose the other parts to go round it (ram etc)
Of those 3 companies, only AMD currently make a complete package that's likely to satisfy gamers... VIA only produce low power CPUs that don't compete on the gaming front, and Intel only produce low end videocards that would be no use for serious gaming. I would imagine nVidia's target would be producing videocards for use with Intel motherboards, and by OEMs such as Apple. Their linux drivers are also better than ATI's for the time being, tho that is rapidly changing.
Not much... You need to grow it indoors, if you grow it outside where it can be seen expect a visit from the cops. You need to give it lots of light to grow, which consumes a lot of power and generates heat, makes your electric bill abnormally high (which flags warning signs) and gives your house a large infrared footprint...
If you just grow a small amount i'm sure you could get away with it, the cops are more concerned with people who are producing and distributing large quantities than someone who has a small amount for their own use.
If they are starving, it's because they don't have sufficient resources to sustain their current population. If you let them starve, the population will contract to a sustainable level. If you give them food, the population will increase to even more unsustainable levels meaning you have to keep giving them food or face an even bigger level of starvation.
They really need to stop having so many kids, smaller families will put far less of a strain on the available resources.
And these third world countries were doing just fine before the europeans went and interfered with them... We really should just leave them alone to make their own way without interference.
Well yes, it should be limited to goods they don't already have sufficient supply of...
The trouble is, the local producers can't fulfill all the demand, and many people cannot afford to buy from the local producers. So foreign handouts come along... Suddenly those people who could afford to buy from the local producers, now take the freebies, and many of those who couldn't afford the local producers still have nothing.
On the other hand, there are very few (if any?) producers of computer hardware in the third world. I think we should send obsolete but still fully functional computers there, while educating locals how to provide service and support for those who don't want to learn in depth about the computers. As it stands, there is no way people in the third world will be able to produce computer hardware, even to a level that would be considered horrendously obsolete by today's standards... But they are perfectly capable of learning how to support these machines and writing software for them. An otherwise outdated computer that goes to be used in the third world saves landfill. But we should do something about those who send junk, that is completely defective machines that aren't of any use whatsoever.
But if you pay the one time fee for windows, you still have to go on the internet and contact third parties for support (i doubt you will be able to speak to any actual developers)... You have to pay again (a lot more) if you want support, and you can buy the same kind of support for most other systems too.
What i always find, when doing a security test against an AD network...
If you root the DC, the network is completely owned... If you root a workstation you can usually get access to the DC from it, hijack a logged in user, crack cached passwords or keylog as someone logs in (and then break something so an admin has to log in). If you get the password hashes, they will usually be Lanman and NTLM... Lanman is laughably weak and trivially cracked, NTLM is better but still much weaker than the encryption used on todays unix systems.
There is also the shear number of services AD requires you to open up through firewalls, a number of incredibly complex services need to be opened up, including ports that provide access to multiple complex services. Unlike the unix philosophy where one service port has a defined function so it's possible to keep a strict eye on what's happening...
I would much rather use a unix system for centralised authentication, and make sure that access to the authentication server itself is not controlled by the centralised auth (ie you cant own a workstation and sniff your way into the master server).
I would also run it on a different OS to the rest of the systems.
Remember a server that controls authentication to every other device on your network has to be the most important device, and should be very closely guarded.
And as someone else just said: "You can integrate any two OSs with minimal pain provided neither of them is made by M$." Very true, everyone else follows the same published standards, MS make their own. So you end up having to reverse engineer and implement something just for MS, and follow a set of published standards to support everything else. If MS weren't so big they would be laughed out of business for not following the same standards everyone else does.
Yes, but with a lot more options than the current configure script offers... Such a large number of settings is impractical for a configure script, you'd overflow the maximum length of a commandline with a program the size of openoffice. That's why a menuconfig style program makes more sense.
It's worth learning unix if only because competent unix admins are in demand, are usually paid more and having more than one skill (and the ability to learn new things is essential because you never know what technology people will be using 10 years from now)...
I can't understand why anyone working in IT wouldn't want to know unix/linux these days.
IBM really ought to open up the protocols it uses (or better yet, make it use existing standard protocols), so that third party clients can be developed... Many services these days are far more useful when you ditch the official client and use a superior third party one.
MS merely rode on the wave of hardware opening up, and used it as a back door to keeping people locked in to software instead. They got away with it because the software cost was low compared to the benefits of having open hardware. If it weren't for microsoft, and we could have had an open os running on open hardware from the start we would be far better off now.
But thinking from the perspective of the company... Is it better to pay you $40/hour, or a competent unix admin $80/hour?
A good unix admin may cost more, but... They can often get the equivalent work done in less time (and therefore more done in the same time/cost). With competent staff and unix the hardware costs will be lower. With unix the software costs will be lower.
The overall cost with unix is likely to be lower so it's better for the company, and the wages are better with unix so it's better for the individual (for now anyway, as use increases the wages will go up, and then drop back down as people flood into the demand)... I would much rather the company i work for spend more on me, and less on microsoft... money going out to microsoft does not benefit me at all.
Sun's support is different, they support the entire stack, hardware, os, apps... It's a lot better that way because you don't get your app vendor blaming your os vendor who blames the hardware vendor, they just get on with fixing the problems.
Counting them, and maintaining a database of how many you use vs how many you buy is also a costly and time consuming process, and often inefficient (ie results in buying more than necessary since some are not reused properly when old hardware is turned off for instance). When considering total costs, don't forget to factor in hidden costs like license compliance.
The level of support included with pirate copies of windows is nothing, the price is 0, but this is illegal. The level of support included with downloaded copies of linux is nothing, the price is 0 but it's not illegal.
The level of support included with retail copies of windows (ie the expensive boxed copies) is typically 1 or 2 phonecalls to help you get it installed, the price is typically 150-300$ depending on version. The level of support included with retail copies of linux is typically 1 or 2 phonecalls for installation too, but you pay $20 or so for a retail ubuntu these days.
The level of support included with OEM copies of windows is none, the OEM may or may not include support in the cost of the whole machine. Linux is generally the same, the manufacturer may or may not include support at their discretion.
If you want extra support (on top of the cost of acquiring the os, but you have to acquire it legally) the price varies depending on the level of support. People often compare the cost of this support to the purchase cost of windows, but you really need to compare the cost of linux support to the cost of windows plus the cost of equivalent support. Support equivalent to what linux vendors provide is only available from microsoft (no third party can offer patches), and is typically very expensive and only available to large customers.
You also want support for more than just the base OS... A basic install of windows is largely useless, and you need to add additional software to it if you want to do anything useful, even if you buy all this additional software from microsoft (additional purchase cost), support for it is also extra on top of the os support, and is likely handled by a different department who will sometimes keep bouncing you back and forth between app and os support. Buck passing like this is likely to be even worse if you use third party apps. Microsoft also don't make hardware, and are likely to blame hardware vendors and their drivers for incompatibilities. A basic install of any linux distro is typically quite useful, and comes with all the apps a typical user might require, supported as part of the os. Although it's still possible for a software-only linux vendor like ubuntu or redhat to blame the hardware, the drivers are typically included in the kernel and thus supported by the os vendor.
Linux and other unixes are often available as fully supported bundles including hardware, where all of your support comes from one place. This is very good from an enterprise perspective, as you have one entity to call for any issues relating to a particular system. Microsoft don't make hardware, and those companies that do make hardware don't have access to their source code to issue fixes to the core os.
What do you get for your $293 worth of "desktop support"? What level of support do you get when you buy vista ultimate? How much does vista cost without support? Where are the other options for buying vista support (proper support from a company with the source code and an ability to actually fix bugs, not just simple installation guide support)?
If you're going to compare the cost of ubuntu support, at least break down exactly what you get for the money, and which parts of the cost are optional (most people never call support).
Also if your staff only know windows then they are likely to be a false economy, it has been well documented that competent unix staff can take care of more systems.. So while each individual staff member may cost more, the amount you save by having less of them can often outweigh the higher wages.
Not to mention other savings they can make, the cost of the software, the cost (in time and money) of license compliance, savings on hardware as less hardware can be used to perform the same work, and older hardware can be used for long before needing replacing (often because older hardware is less power efficient (performance per watt) and thus not economical to keep running, rather than being too slow to handle the load).
Remember unix was around long before windows, and there are plenty of people with 20+ years of unix experience or more...
I have practical experience of this, having worked as both a unix admin and a windows admin, and worked in several places where a big function of my job was to steer the companies towards unix, and have saved several companies a lot of money by migrating various systems to unix (considerably more than the cost of my wages).
My advise to you, is to put in the effort to learn unix... If you become sufficiently competent you will be able to command a higher level of pay, while introducing companies to various ways of saving money they probably didn't realize existed before. Especially in smaller companies this can take you a long way. Once you become used to unix chances are you will prefer it, and use it in preference to windows anyway, and aside from that having multiple skills and the ability to learn new things is always good, because you can never guarantee what technology will be in use years from now and it's not good to be stuck in a fading niche (think of all those people who trained in wordperfect on dos).
Because their programs save files in open formats that can be used without penalty in other apps... Thus, one person's choice of lotus symphony doesn't force you to also use the same app, you are free to use any app implementing the same standard format.
I have no issue what software other people choose to use, so long as their choice doesn't harm my freedom to choose for myself.
What we really need, is for apps like openoffice to be modular... Then we can have the "clone interface" and one or more "better interfaces", so that those of us who are willing to try something new can benefit from it.
Unfortunately people are used to microsoft, regardless of how inefficient their way of doing things is, people want familiarity.
That's why many open source projects tend to copy them, as it's the only way to capture a large market share (if you do things the right way, you can end up with a small but loyal niche).
Really i think openoffice (and other large apps) should have a config program like the menuconfig present in linux, so you can configure it before compiling, and choose to have a huge monolithic app (like it is now), or choose various features as modules (loaded as theyre used) or simply disable them, and also choose between several interface options (or build these as modules too so you can switch between them). Tho i'm sure this would all be a lot of work...
As for you paying serious money for the infobox, why don't you donate to the project, offer a bounty to someone to implement such features or if your really that serious, hire some developers to implement it.
I have many such utilities installed on several linux machines, and make quite heavy use of them... I do wonder how much it would cost to find commercially available replacements for everything i use, and if those replacements would be as good.
Microsoft have their own audio/video formats which they could release without having to pay royalties to third parties, or they could use existing freely distributable formats.
While I agree about the need for insurance, handing a market like this to private companies is a bad thing... You leave companies with the ability to create huge profits from a captive market.
What we really need is non profit insurance, so those of us who don't make claims are not having to subsidise those who do. Despite being a young male with a powerful car, I have not made a single insurance claim (or caused anyone else to do so) despite driving heavily over the last 8 years. And yet i still have to pay ridiculous amounts for the insurance, to generate profits for the insurance companies and subsidise people who drive far more recklessly than me.
Slashdot Mirror
I've had various VIA, ALi and SiS chipsets which had a poor reputation, and yet had very good stability under Linux...
Similarly, Cyrix CPUs had a reputation for causing stability problems, and yet i had several cyrix based machines running for years without incident. One of the cyrix systems was given to me by someone who got sick of it's instability while running windows. His replacement system fared little better.
I'd say it was more down to drivers than hardware, windows systems often end up with a horrible mess of shoddily written drivers from multiple third parties that can often conflict with other, whereas linux *usually* has all the drivers built together in the same kernel. The only times i've had issues with linux were due to out of kernel drivers of questionable quality.
Perhaps, or perhaps the cpu/motherboard will just be thought of as a single component instead of discrete ones... You choose one, and then choose the other parts to go round it (ram etc)
Of those 3 companies, only AMD currently make a complete package that's likely to satisfy gamers...
VIA only produce low power CPUs that don't compete on the gaming front, and Intel only produce low end videocards that would be no use for serious gaming.
I would imagine nVidia's target would be producing videocards for use with Intel motherboards, and by OEMs such as Apple.
Their linux drivers are also better than ATI's for the time being, tho that is rapidly changing.
Not much...
You need to grow it indoors, if you grow it outside where it can be seen expect a visit from the cops.
You need to give it lots of light to grow, which consumes a lot of power and generates heat, makes your electric bill abnormally high (which flags warning signs) and gives your house a large infrared footprint...
If you just grow a small amount i'm sure you could get away with it, the cops are more concerned with people who are producing and distributing large quantities than someone who has a small amount for their own use.
If they are starving, it's because they don't have sufficient resources to sustain their current population. If you let them starve, the population will contract to a sustainable level. If you give them food, the population will increase to even more unsustainable levels meaning you have to keep giving them food or face an even bigger level of starvation.
They really need to stop having so many kids, smaller families will put far less of a strain on the available resources.
And these third world countries were doing just fine before the europeans went and interfered with them... We really should just leave them alone to make their own way without interference.
Well yes, it should be limited to goods they don't already have sufficient supply of...
The trouble is, the local producers can't fulfill all the demand, and many people cannot afford to buy from the local producers. So foreign handouts come along... Suddenly those people who could afford to buy from the local producers, now take the freebies, and many of those who couldn't afford the local producers still have nothing.
On the other hand, there are very few (if any?) producers of computer hardware in the third world. I think we should send obsolete but still fully functional computers there, while educating locals how to provide service and support for those who don't want to learn in depth about the computers. As it stands, there is no way people in the third world will be able to produce computer hardware, even to a level that would be considered horrendously obsolete by today's standards... But they are perfectly capable of learning how to support these machines and writing software for them. An otherwise outdated computer that goes to be used in the third world saves landfill.
But we should do something about those who send junk, that is completely defective machines that aren't of any use whatsoever.
But if you pay the one time fee for windows, you still have to go on the internet and contact third parties for support (i doubt you will be able to speak to any actual developers)...
You have to pay again (a lot more) if you want support, and you can buy the same kind of support for most other systems too.
What i always find, when doing a security test against an AD network...
If you root the DC, the network is completely owned...
If you root a workstation you can usually get access to the DC from it, hijack a logged in user, crack cached passwords or keylog as someone logs in (and then break something so an admin has to log in).
If you get the password hashes, they will usually be Lanman and NTLM... Lanman is laughably weak and trivially cracked, NTLM is better but still much weaker than the encryption used on todays unix systems.
There is also the shear number of services AD requires you to open up through firewalls, a number of incredibly complex services need to be opened up, including ports that provide access to multiple complex services. Unlike the unix philosophy where one service port has a defined function so it's possible to keep a strict eye on what's happening...
I would much rather use a unix system for centralised authentication, and make sure that access to the authentication server itself is not controlled by the centralised auth (ie you cant own a workstation and sniff your way into the master server).
I would also run it on a different OS to the rest of the systems.
Remember a server that controls authentication to every other device on your network has to be the most important device, and should be very closely guarded.
And as someone else just said:
"You can integrate any two OSs with minimal pain provided neither of them is made by M$."
Very true, everyone else follows the same published standards, MS make their own. So you end up having to reverse engineer and implement something just for MS, and follow a set of published standards to support everything else. If MS weren't so big they would be laughed out of business for not following the same standards everyone else does.
Not sure about vbscript, but vba (msoffice macro language) is going away, and it's not the first time they deprecated a macro language...
Yes, but with a lot more options than the current configure script offers...
Such a large number of settings is impractical for a configure script, you'd overflow the maximum length of a commandline with a program the size of openoffice. That's why a menuconfig style program makes more sense.
It's worth learning unix if only because competent unix admins are in demand, are usually paid more and having more than one skill (and the ability to learn new things is essential because you never know what technology people will be using 10 years from now)...
I can't understand why anyone working in IT wouldn't want to know unix/linux these days.
IBM really ought to open up the protocols it uses (or better yet, make it use existing standard protocols), so that third party clients can be developed...
Many services these days are far more useful when you ditch the official client and use a superior third party one.
MS merely rode on the wave of hardware opening up, and used it as a back door to keeping people locked in to software instead. They got away with it because the software cost was low compared to the benefits of having open hardware.
If it weren't for microsoft, and we could have had an open os running on open hardware from the start we would be far better off now.
but very limited, you should be using windows powershell, which is far more powerful but requires you to learn something new.
But thinking from the perspective of the company...
Is it better to pay you $40/hour, or a competent unix admin $80/hour?
A good unix admin may cost more, but...
They can often get the equivalent work done in less time (and therefore more done in the same time/cost).
With competent staff and unix the hardware costs will be lower.
With unix the software costs will be lower.
The overall cost with unix is likely to be lower so it's better for the company, and the wages are better with unix so it's better for the individual (for now anyway, as use increases the wages will go up, and then drop back down as people flood into the demand)...
I would much rather the company i work for spend more on me, and less on microsoft... money going out to microsoft does not benefit me at all.
Sun's support is different, they support the entire stack, hardware, os, apps... It's a lot better that way because you don't get your app vendor blaming your os vendor who blames the hardware vendor, they just get on with fixing the problems.
Counting them, and maintaining a database of how many you use vs how many you buy is also a costly and time consuming process, and often inefficient (ie results in buying more than necessary since some are not reused properly when old hardware is turned off for instance).
When considering total costs, don't forget to factor in hidden costs like license compliance.
The level of support included with pirate copies of windows is nothing, the price is 0, but this is illegal.
The level of support included with downloaded copies of linux is nothing, the price is 0 but it's not illegal.
The level of support included with retail copies of windows (ie the expensive boxed copies) is typically 1 or 2 phonecalls to help you get it installed, the price is typically 150-300$ depending on version.
The level of support included with retail copies of linux is typically 1 or 2 phonecalls for installation too, but you pay $20 or so for a retail ubuntu these days.
The level of support included with OEM copies of windows is none, the OEM may or may not include support in the cost of the whole machine.
Linux is generally the same, the manufacturer may or may not include support at their discretion.
If you want extra support (on top of the cost of acquiring the os, but you have to acquire it legally) the price varies depending on the level of support. People often compare the cost of this support to the purchase cost of windows, but you really need to compare the cost of linux support to the cost of windows plus the cost of equivalent support. Support equivalent to what linux vendors provide is only available from microsoft (no third party can offer patches), and is typically very expensive and only available to large customers.
You also want support for more than just the base OS... A basic install of windows is largely useless, and you need to add additional software to it if you want to do anything useful, even if you buy all this additional software from microsoft (additional purchase cost), support for it is also extra on top of the os support, and is likely handled by a different department who will sometimes keep bouncing you back and forth between app and os support. Buck passing like this is likely to be even worse if you use third party apps. Microsoft also don't make hardware, and are likely to blame hardware vendors and their drivers for incompatibilities.
A basic install of any linux distro is typically quite useful, and comes with all the apps a typical user might require, supported as part of the os. Although it's still possible for a software-only linux vendor like ubuntu or redhat to blame the hardware, the drivers are typically included in the kernel and thus supported by the os vendor.
Linux and other unixes are often available as fully supported bundles including hardware, where all of your support comes from one place. This is very good from an enterprise perspective, as you have one entity to call for any issues relating to a particular system. Microsoft don't make hardware, and those companies that do make hardware don't have access to their source code to issue fixes to the core os.
What do you get for your $293 worth of "desktop support"?
What level of support do you get when you buy vista ultimate?
How much does vista cost without support?
Where are the other options for buying vista support (proper support from a company with the source code and an ability to actually fix bugs, not just simple installation guide support)?
If you're going to compare the cost of ubuntu support, at least break down exactly what you get for the money, and which parts of the cost are optional (most people never call support).
Also if your staff only know windows then they are likely to be a false economy, it has been well documented that competent unix staff can take care of more systems.. So while each individual staff member may cost more, the amount you save by having less of them can often outweigh the higher wages.
Not to mention other savings they can make, the cost of the software, the cost (in time and money) of license compliance, savings on hardware as less hardware can be used to perform the same work, and older hardware can be used for long before needing replacing (often because older hardware is less power efficient (performance per watt) and thus not economical to keep running, rather than being too slow to handle the load).
Remember unix was around long before windows, and there are plenty of people with 20+ years of unix experience or more...
I have practical experience of this, having worked as both a unix admin and a windows admin, and worked in several places where a big function of my job was to steer the companies towards unix, and have saved several companies a lot of money by migrating various systems to unix (considerably more than the cost of my wages).
My advise to you, is to put in the effort to learn unix... If you become sufficiently competent you will be able to command a higher level of pay, while introducing companies to various ways of saving money they probably didn't realize existed before. Especially in smaller companies this can take you a long way. Once you become used to unix chances are you will prefer it, and use it in preference to windows anyway, and aside from that having multiple skills and the ability to learn new things is always good, because you can never guarantee what technology will be in use years from now and it's not good to be stuck in a fading niche (think of all those people who trained in wordperfect on dos).
Because their programs save files in open formats that can be used without penalty in other apps...
Thus, one person's choice of lotus symphony doesn't force you to also use the same app, you are free to use any app implementing the same standard format.
I have no issue what software other people choose to use, so long as their choice doesn't harm my freedom to choose for myself.
What we really need, is for apps like openoffice to be modular...
Then we can have the "clone interface" and one or more "better interfaces", so that those of us who are willing to try something new can benefit from it.
Unfortunately people are used to microsoft, regardless of how inefficient their way of doing things is, people want familiarity.
That's why many open source projects tend to copy them, as it's the only way to capture a large market share (if you do things the right way, you can end up with a small but loyal niche).
Really i think openoffice (and other large apps) should have a config program like the menuconfig present in linux, so you can configure it before compiling, and choose to have a huge monolithic app (like it is now), or choose various features as modules (loaded as theyre used) or simply disable them, and also choose between several interface options (or build these as modules too so you can switch between them). Tho i'm sure this would all be a lot of work...
As for you paying serious money for the infobox, why don't you donate to the project, offer a bounty to someone to implement such features or if your really that serious, hire some developers to implement it.
I have many such utilities installed on several linux machines, and make quite heavy use of them...
I do wonder how much it would cost to find commercially available replacements for everything i use, and if those replacements would be as good.
Why exactly do they need to use MPEG?
Microsoft have their own audio/video formats which they could release without having to pay royalties to third parties, or they could use existing freely distributable formats.
While I agree about the need for insurance, handing a market like this to private companies is a bad thing...
You leave companies with the ability to create huge profits from a captive market.
What we really need is non profit insurance, so those of us who don't make claims are not having to subsidise those who do. Despite being a young male with a powerful car, I have not made a single insurance claim (or caused anyone else to do so) despite driving heavily over the last 8 years. And yet i still have to pay ridiculous amounts for the insurance, to generate profits for the insurance companies and subsidise people who drive far more recklessly than me.