I had much the same experience, but I switched a little earlier than you.
It's a total no-brainer for me. You're talking about a unix workstation that "just works", without ugly translation layers like cygwin, with all the tools delivered right in the box (full terminal environment, dev tools, X11), and where almost every GUI app is fully scriptable, even the 3rd party ones (applescript).
Plus, the intel machines can seamlessly integrate windows apps using vmware fusion. You get the best of all worlds (windows, max and *nix).
Software as a service can be run locally by a company, rather than on the web.
Welcome to the new world, it's the same as the old. Businesses already run on the software as a service model. Go into any large corporation and try to find a user that installs their own software and obtains all that software themselves. The reality is that software is already centrally provided, and that licensing is already moving to a per-term instead of a per-copy model (see the new windows licensing model for example).
Just because it isn't web-based doesn't mean it's not a service. In fact, I don't see current web technologies becoming the cornerstone of software-as-a-service anytime soon, because they're just not well suited to a remotely updated, remotely stored, locally cached, locally executed model. Probably we're going to see something like adobe's AIR become "the new black".
The ODF supporters are trying to block OOXML from becoming a standard ONLY because they've lobbied governments into passing open standards laws that will REQUIRE those governments to use their format if OOXML fails. They don't give a damn if it's the best format or not; they want a monopoly enforced by law.
Nonsense. The ODF supporters want an open format, so that there is no more microsoft lock-in in the office applications market. Some of these are indeed supporting it for commercial reasons (sun and google), but most of the ODF supporters are in it for transparency in government.
Did you catch those facts? A sitting President lied under oath to a court proceeding.
When Clinton was asked whether he had sexual relations with Lewinski, he should have answered "I don't recall." Sure, it would have been a lie, but as has been demonstrated by Bush cabinet officials over and over again, it's one you can get away with because nobody can prove you're not on the edge of dementia.
Microsoft's answer is that it's a nescessary trade-off for good sound performance. If they acknowledged it as a bug there wouldn't be such a bitch-fest going on.
See, but it is necessary. If you need to encrypt all audio while you're shuttling it around the OS to make sure it doesn't get copied by evil, evil pirates, then that obviously creates a noticeable overhead. This is what they mean when they say "by design". A design that's not meant to serve the user is still a design.
People have forgotten that the purpose of an OS (including linux) is letting the user perform tasks. This is a cross-project discipline.
Suppose a user wanted to record some family video, edit it into a movie with some nice graphics and credits thrown in, and then distribute it on the internet on a site of their own (say "simpsonsfamilyvideo.com") so they could tell the address for the site to friends and family to check out. Suppose that this user was your mom. How much explanation do you think she would need to do this? Do you think she would just give up halfway and make her geek son do it?
The reality is that linux doesn't need any more features. It does everything a user needs. What it needs is integration, polish, documentation and task-orientation. Starting from a common task that users do, and then thinking "how can we improve that?" is the only way to move forward from here on out, and doing that means that you'll have to mess with dozens of projects, instead of focussing on just two or three.
you may have heard about security exploits through Java there... I'm going to go with Dangerous Java Flaw Threatens 'Virtually Everything' - especially because the flaw it's describing turns out to be a buffer overflow.
It still makes a lot more sense to have a (small) sandbox that is checked once for security errors, than to have to check every single piece of software developed because it isn't sandboxed. An average Java app has a pretty decent security track record compared to an average native-compiled app.
No, it hasn't. It's chained you to the lowest common denominator of the systems I've already listed. If the Windows API doesn't offer something, you can't do it, even if the Linux and Solaris APIs do.
That has nothing to do with java, that has something to do with cross-platform development. You always have to program to the lowest common denominator in that case.
Oh really? Cause they actually don't work at all. Where is the netbsd/sparc jre? Where's the openbsd/arm jre? Portable doesn't mean "portable to wherever Sun has decided you can use it", it means "can be made to run anywhere". Python's VM and libraries are available on my openbsd/arm and netbsd/sparc machines, why isn't java's?
Probably because those platforms don't have commercial value. If there was a genuine commercial demand for support for those platforms, there would be a jre for them.
People dissatisfied with Vista pre-installed on their laptops don't install Linux; they return the laptops and demand XP.
I suspect that Apple's record sales figures might figure into this also. Vista is the perfect excuse to finally get that mac you've always lusted after.
You make it sound way more complicated than it is.
Reinitializing all hardware to the previous state is no biggy in a well-designed system, since it is fully under your control.
The only unknown is the network, and that is no problem, since there is no difference between pulling your RJ-45 out of its jack and performing a hibernate. In both cases you're counting on the apps to deal with a lost network connection, and network-aware apps really should be able to deal with that.
In short, the "but the devices" argument goes nowhere. What we're seeing here is a matter of budgetting. It's perfectly possible to build the system so it hibernates perfectly, it just costs more money than microsoft and the hardware/driver makers are willing to spend on it.
In addition to the comments from the other respondent, don't forget that Apple provides *everything* - hardware, OS and drivers. It's a much, much easier problem to solve when you control everything involved.
Irrelevant. We're not talking about home-brew PC's here. We're talking about major-brand laptops from dell, hp, sony and so on that you buy in the store with vista pre-installed. Often sleep/hibernate doesn't work right on those either, despite all the hardware and drivers being designed to work right for that configuration. So, either the argument is that the only way for an OS maker to build decent power management is to design their own hardware (aka the "steve jobs" argument), or the argument is that MS/Sony/HP/Dell collectively dropped the ball.
The government allowed what to happen? That only one ISP chose to put the infrastructure in your area for broadband? ` That infrastructure was put there using government subsidies. It is simple too expensive to provide physical cabling to everyone except in dense metropolitan areas to be able to enter into that market as a new business (and what you see in the US is that only dense metropolitan areas get decent competition).
Network cabling is just the same as electricity lines, water mains, gas mains, sewer system, and so on. It doesn't make economic sense to have more than one of them in your street, so it's up to the goverment to ensure that artificial competition is created on the one line that's there.
And suddenly things like downloading videos from iTunes become a whole lot less attractive. Torrent-gobbling nerds aren't the only ones using a lot of bandwidth, and that will become more and more true in the near future.
Ha. Been there also. I used to have an internet provider that charged so much for bandwidth that every time I watched a movie on their rental service it added a full euro in bandwidth cost. Suffice it to say that the movie rental service has never really taken off.
A shame, because I like the idea of renting movies on-line and have been waiting for something that is price-competitive with physical rental for a while.
To prevent runaway bills, allow customers to set their own "caps" and "throttle-down speeds" that would kick in after the cap was reached. If a customer never wanted to pay more than $20, he could set his "monthly cap" at 80% of what $20 would buy, and set the throttle-down rate low enough that he could never use up the remaining 20% even if he was maxing out his connection.
Been there done that. In my country (belgium) they've been doing it this way since the turn of the century. I currently pay 30 euro for 30 gb / month on 6 mbit ADSL (+ 6 euro for the phone line). Every extra gb costs me 0.25 euro, up to a maximum of 100 gb, at which point they rate-limit me (or lower if I choose to not allow extra volume). The rate-limiting puts you on an 8KB/sec connection, which is too slow for any major downloading, but still enough to check your email and do some light browsing.
OS X itself only sells for about $130, whereas the only version of Windows you can get for that price is either an upgrade, or a severely hobbled version that can't even begin to stand up against OS X. The bulk of the cost of any Mac comes in the worth of its hardware; the best available for the price you're paying.
Apple is a software company. Steve Jobs has said so several times. When you buy a mac, you're buying software, and the hardware it runs on. The $130 OS X price tag is for upgrades only, since you are not allowed to run it on any hardware that didn't come with a mac os license originally.
What criteria is used to decide that the local government should get involved in becoming an ISP? Why shouldn't they also get involved in making coffee? There's a lot of people that need coffee in the morning.
Government should be involved in everything that is a natural monopoly, either by regulating it so that it has artifically induced competition (my preferred solution), or by doing it outright. The notion that everything is better if privatized and unregulated is absolutely silly, because the worst places in the world are places with exactly that situation (parts of africa, areas with civil war,...). It's economy 101: if something is easily monopolized, it will be, and private monopolies are ALWAYS bad for the customer.
Fire departments are a bad analogy, they are required by law to maintain a certain level of service because fire spreads.
That's the worst "that's a bad analogy" I've ever heard. It's a perfect analogy because you can mandate that the telco service provided by the government has minimum standards of service.
This speaks legions to me, and it says Apple is not only going to turn the iPhone into a a cool smartphone, but they will also start selling games with it. IT has enough horsepower and screen real estate to take on the PSP..... and the DS, with the multi touch interface.
Newsflash: they intend to follow the same model they followed with the iPod, namely to distribute a few games via the itms, and to not have any 3rd party development. Games are a gimmick on the iPod, and they'll be a gimmick on the iPhone. It's pretty clear that apple has zero intention for opening up the iPhone, and as a result the iPhone will sell well, but it won't be industry-changing.
By that argument there's nothing wrong with patents that can't be fixed by halving their length. Quite a big difference between that and abolishing them altogether.
No, by that argument, why bother?
Can you really say that you have cold hard numbers to back up the basic premises that underly the patent system? I'd love to see them, because I want faith to turn into fact.
Remember that his patent will eventually expire, and when it does he's back to good old-fashioned "beat the competition on price and service because they've got much the same technology" (which is buggery difficult without a unique aspect to your product) unless he can come up with another patentable idea.
Exactly! You point out the basic flaw in the patent system: the innovator only needs to innovate once every 20 years. It rewards laziness! Without patents, under a free market, the innovator is forced to innovate constantly, and only the very brightest survive the market's forces.
If you scrap the whole thing you're left with only the negative effects and none of the positive ones which people are currently enjoying.
Not true. If you scrap it, you remove all the costs involved with the patent system, like ineffective markets, delayed introduction of some categories of products (especially software is delayed strongly under the influence of a patent system), and spurious litigation by patent owners.
You're assuming that the current system has no positive benefits at all which I believe is not the case. I agree the patent system is open to abuse, particulary in the area of software but I also think a lot of people have used the patent system in the way it was intended and reaped benefits from it.
Let me put it like this: what study or source of reliable data do you use as a basis for your idea that the patent system is a net benefit to society?
I am in fact personally in favor of scrapping the patent system, because I've never heard anyone make a sensible case for why it should be kept. Yes, it has benefits, but those benefits come at, to me, a greater cost than they are worth. You don't need a patent system to have innovation. A free market will force people to innovate or fall behind. And for those things that aren't profitable without monopolies: put them in the hands of government. If you're going to have an inefficient mandatory monopoly, let it be one that people can vote out of office rather than one they have zero control over.
I sometimes suffer from communicating with the rest 5% support people. They insist to just follows the stylized procedure, and does not allow me the shortcuts.
Keep pressing 0 on the dial pad to speak to an operator. If nothing else, it'll annoy the shit out of them.
They also offer no support. If you call with a trouble report you'd better pretend you are using a Windows machine when they give you their step by step connection test instructions. If they say "click Start -> Control Panel" and you say "I have neither", the problem is obviously on your end.
This is something I've done more than once. Just pretend you're doing what they ask, and every time they ask you if it's working, answer with a negative. Sooner or later the script moves on to something that's useful to you, and your problem is likely to get resolved.
I'm very doubtful that a requirement to vote will lead to an improvement, at least not for America - maybe it works for Australia. Instead I believe it would simply lead to more misinformed/uninformed voting (akin to drunk driving, except that no one is likely to be directly killed) and even more aggressive marketing of candidates with as little or even less substance.
An uninformed vote is still a valid vote. People's voice counts even if they're stupid.
Besides, what makes people vote is not being informed, it's caring about the result. There are many informed people who don't vote because they believe the system is broken (and they're right). Just like there are many uninformed people who do vote because they care about their specific issue. Optional voting encourages only the extremists to vote.
Slashdot Mirror
I had much the same experience, but I switched a little earlier than you.
It's a total no-brainer for me. You're talking about a unix workstation that "just works", without ugly translation layers like cygwin, with all the tools delivered right in the box (full terminal environment, dev tools, X11), and where almost every GUI app is fully scriptable, even the 3rd party ones (applescript).
Plus, the intel machines can seamlessly integrate windows apps using vmware fusion. You get the best of all worlds (windows, max and *nix).
Software as a service can be run locally by a company, rather than on the web.
Welcome to the new world, it's the same as the old. Businesses already run on the software as a service model. Go into any large corporation and try to find a user that installs their own software and obtains all that software themselves. The reality is that software is already centrally provided, and that licensing is already moving to a per-term instead of a per-copy model (see the new windows licensing model for example).
Just because it isn't web-based doesn't mean it's not a service. In fact, I don't see current web technologies becoming the cornerstone of software-as-a-service anytime soon, because they're just not well suited to a remotely updated, remotely stored, locally cached, locally executed model. Probably we're going to see something like adobe's AIR become "the new black".
The ODF supporters are trying to block OOXML from becoming a standard ONLY because they've lobbied governments into passing open standards laws that will REQUIRE those governments to use their format if OOXML fails. They don't give a damn if it's the best format or not; they want a monopoly enforced by law.
Nonsense. The ODF supporters want an open format, so that there is no more microsoft lock-in in the office applications market. Some of these are indeed supporting it for commercial reasons (sun and google), but most of the ODF supporters are in it for transparency in government.
The case against OOXML on technical grounds has been made. The format is not open. See this link if you want more info: http://ooxmlisdefectivebydesign.blogspot.com/
Did you catch those facts? A sitting President lied under oath to a court proceeding.
When Clinton was asked whether he had sexual relations with Lewinski, he should have answered "I don't recall." Sure, it would have been a lie, but as has been demonstrated by Bush cabinet officials over and over again, it's one you can get away with because nobody can prove you're not on the edge of dementia.
Microsoft's answer is that it's a nescessary trade-off for good sound performance. If they acknowledged it as a bug there wouldn't be such a bitch-fest going on.
See, but it is necessary. If you need to encrypt all audio while you're shuttling it around the OS to make sure it doesn't get copied by evil, evil pirates, then that obviously creates a noticeable overhead. This is what they mean when they say "by design". A design that's not meant to serve the user is still a design.
People have forgotten that the purpose of an OS (including linux) is letting the user perform tasks. This is a cross-project discipline.
Suppose a user wanted to record some family video, edit it into a movie with some nice graphics and credits thrown in, and then distribute it on the internet on a site of their own (say "simpsonsfamilyvideo.com") so they could tell the address for the site to friends and family to check out. Suppose that this user was your mom. How much explanation do you think she would need to do this? Do you think she would just give up halfway and make her geek son do it?
The reality is that linux doesn't need any more features. It does everything a user needs. What it needs is integration, polish, documentation and task-orientation. Starting from a common task that users do, and then thinking "how can we improve that?" is the only way to move forward from here on out, and doing that means that you'll have to mess with dozens of projects, instead of focussing on just two or three.
you may have heard about security exploits through Java there... I'm going to go with Dangerous Java Flaw Threatens 'Virtually Everything' - especially because the flaw it's describing turns out to be a buffer overflow.
It still makes a lot more sense to have a (small) sandbox that is checked once for security errors, than to have to check every single piece of software developed because it isn't sandboxed. An average Java app has a pretty decent security track record compared to an average native-compiled app.
No, it hasn't. It's chained you to the lowest common denominator of the systems I've already listed. If the Windows API doesn't offer something, you can't do it, even if the Linux and Solaris APIs do.
That has nothing to do with java, that has something to do with cross-platform development. You always have to program to the lowest common denominator in that case.
Oh really? Cause they actually don't work at all. Where is the netbsd/sparc jre? Where's the openbsd/arm jre? Portable doesn't mean "portable to wherever Sun has decided you can use it", it means "can be made to run anywhere". Python's VM and libraries are available on my openbsd/arm and netbsd/sparc machines, why isn't java's?
Probably because those platforms don't have commercial value. If there was a genuine commercial demand for support for those platforms, there would be a jre for them.
The irony is that the hardest core of early slashdot users don't have low uid's, because they resisted creating accounts to protect their privacy.
People dissatisfied with Vista pre-installed on their laptops don't install Linux; they return the laptops and demand XP.
I suspect that Apple's record sales figures might figure into this also. Vista is the perfect excuse to finally get that mac you've always lusted after.
You make it sound way more complicated than it is.
Reinitializing all hardware to the previous state is no biggy in a well-designed system, since it is fully under your control.
The only unknown is the network, and that is no problem, since there is no difference between pulling your RJ-45 out of its jack and performing a hibernate. In both cases you're counting on the apps to deal with a lost network connection, and network-aware apps really should be able to deal with that.
In short, the "but the devices" argument goes nowhere. What we're seeing here is a matter of budgetting. It's perfectly possible to build the system so it hibernates perfectly, it just costs more money than microsoft and the hardware/driver makers are willing to spend on it.
In addition to the comments from the other respondent, don't forget that Apple provides *everything* - hardware, OS and drivers. It's a much, much easier problem to solve when you control everything involved.
Irrelevant. We're not talking about home-brew PC's here. We're talking about major-brand laptops from dell, hp, sony and so on that you buy in the store with vista pre-installed. Often sleep/hibernate doesn't work right on those either, despite all the hardware and drivers being designed to work right for that configuration. So, either the argument is that the only way for an OS maker to build decent power management is to design their own hardware (aka the "steve jobs" argument), or the argument is that MS/Sony/HP/Dell collectively dropped the ball.
The government allowed what to happen? That only one ISP chose to put the infrastructure in your area for broadband?
`
That infrastructure was put there using government subsidies. It is simple too expensive to provide physical cabling to everyone except in dense metropolitan areas to be able to enter into that market as a new business (and what you see in the US is that only dense metropolitan areas get decent competition).
Network cabling is just the same as electricity lines, water mains, gas mains, sewer system, and so on. It doesn't make economic sense to have more than one of them in your street, so it's up to the goverment to ensure that artificial competition is created on the one line that's there.
And suddenly things like downloading videos from iTunes become a whole lot less attractive. Torrent-gobbling nerds aren't the only ones using a lot of bandwidth, and that will become more and more true in the near future.
Ha. Been there also. I used to have an internet provider that charged so much for bandwidth that every time I watched a movie on their rental service it added a full euro in bandwidth cost. Suffice it to say that the movie rental service has never really taken off.
A shame, because I like the idea of renting movies on-line and have been waiting for something that is price-competitive with physical rental for a while.
To prevent runaway bills, allow customers to set their own "caps" and "throttle-down speeds" that would kick in after the cap was reached. If a customer never wanted to pay more than $20, he could set his "monthly cap" at 80% of what $20 would buy, and set the throttle-down rate low enough that he could never use up the remaining 20% even if he was maxing out his connection.
Been there done that. In my country (belgium) they've been doing it this way since the turn of the century. I currently pay 30 euro for 30 gb / month on 6 mbit ADSL (+ 6 euro for the phone line). Every extra gb costs me 0.25 euro, up to a maximum of 100 gb, at which point they rate-limit me (or lower if I choose to not allow extra volume). The rate-limiting puts you on an 8KB/sec connection, which is too slow for any major downloading, but still enough to check your email and do some light browsing.
OS X itself only sells for about $130, whereas the only version of Windows you can get for that price is either an upgrade, or a severely hobbled version that can't even begin to stand up against OS X. The bulk of the cost of any Mac comes in the worth of its hardware; the best available for the price you're paying.
Apple is a software company. Steve Jobs has said so several times. When you buy a mac, you're buying software, and the hardware it runs on. The $130 OS X price tag is for upgrades only, since you are not allowed to run it on any hardware that didn't come with a mac os license originally.
What criteria is used to decide that the local government should get involved in becoming an ISP? Why shouldn't they also get involved in making coffee? There's a lot of people that need coffee in the morning.
...). It's economy 101: if something is easily monopolized, it will be, and private monopolies are ALWAYS bad for the customer.
Government should be involved in everything that is a natural monopoly, either by regulating it so that it has artifically induced competition (my preferred solution), or by doing it outright. The notion that everything is better if privatized and unregulated is absolutely silly, because the worst places in the world are places with exactly that situation (parts of africa, areas with civil war,
Fire departments are a bad analogy, they are required by law to maintain a certain level of service because fire spreads.
That's the worst "that's a bad analogy" I've ever heard. It's a perfect analogy because you can mandate that the telco service provided by the government has minimum standards of service.
This speaks legions to me, and it says Apple is not only going to turn the iPhone into a a cool smartphone, but they will also start selling games with it. IT has enough horsepower and screen real estate to take on the PSP..... and the DS, with the multi touch interface.
Newsflash: they intend to follow the same model they followed with the iPod, namely to distribute a few games via the itms, and to not have any 3rd party development. Games are a gimmick on the iPod, and they'll be a gimmick on the iPhone. It's pretty clear that apple has zero intention for opening up the iPhone, and as a result the iPhone will sell well, but it won't be industry-changing.
By that argument there's nothing wrong with patents that can't be fixed by halving their length. Quite a big difference between that and abolishing them altogether.
No, by that argument, why bother?
Can you really say that you have cold hard numbers to back up the basic premises that underly the patent system? I'd love to see them, because I want faith to turn into fact.
Remember that his patent will eventually expire, and when it does he's back to good old-fashioned "beat the competition on price and service because they've got much the same technology" (which is buggery difficult without a unique aspect to your product) unless he can come up with another patentable idea.
Exactly! You point out the basic flaw in the patent system: the innovator only needs to innovate once every 20 years. It rewards laziness! Without patents, under a free market, the innovator is forced to innovate constantly, and only the very brightest survive the market's forces.
If you scrap the whole thing you're left with only the negative effects and none of the positive ones which people are currently enjoying.
Not true. If you scrap it, you remove all the costs involved with the patent system, like ineffective markets, delayed introduction of some categories of products (especially software is delayed strongly under the influence of a patent system), and spurious litigation by patent owners.
You're assuming that the current system has no positive benefits at all which I believe is not the case. I agree the patent system is open to abuse, particulary in the area of software but I also think a lot of people have used the patent system in the way it was intended and reaped benefits from it.
Let me put it like this: what study or source of reliable data do you use as a basis for your idea that the patent system is a net benefit to society?
I am in fact personally in favor of scrapping the patent system, because I've never heard anyone make a sensible case for why it should be kept. Yes, it has benefits, but those benefits come at, to me, a greater cost than they are worth. You don't need a patent system to have innovation. A free market will force people to innovate or fall behind. And for those things that aren't profitable without monopolies: put them in the hands of government. If you're going to have an inefficient mandatory monopoly, let it be one that people can vote out of office rather than one they have zero control over.
I sometimes suffer from communicating with the rest 5% support people.
They insist to just follows the stylized procedure, and does not allow me the shortcuts.
Keep pressing 0 on the dial pad to speak to an operator. If nothing else, it'll annoy the shit out of them.
They also offer no support. If you call with a trouble report you'd better pretend you are using a Windows machine when they give you their step by step connection test instructions. If they say "click Start -> Control Panel" and you say "I have neither", the problem is obviously on your end.
This is something I've done more than once. Just pretend you're doing what they ask, and every time they ask you if it's working, answer with a negative. Sooner or later the script moves on to something that's useful to you, and your problem is likely to get resolved.
I'm very doubtful that a requirement to vote will lead to an improvement, at least not for America - maybe it works for Australia. Instead I believe it would simply lead to more misinformed/uninformed voting (akin to drunk driving, except that no one is likely to be directly killed) and even more aggressive marketing of candidates with as little or even less substance.
An uninformed vote is still a valid vote. People's voice counts even if they're stupid.
Besides, what makes people vote is not being informed, it's caring about the result. There are many informed people who don't vote because they believe the system is broken (and they're right). Just like there are many uninformed people who do vote because they care about their specific issue. Optional voting encourages only the extremists to vote.