The whole way security is treated needs to be changed. Having root and an ordinary user just doesn't offer the level of granularity that users need. As a user I want to be able to do everything on my computer, what's really needed is fine grained access control per program. Of course, that has issues with users having to grant those privileges but you could have profiles. Imagine installing Evolution or something and it pops up and says "This software says it's a mail client, does that sound right to you?" and then what privileges it gets granted will be set by a "mail client" profile already installed on the system.
It won't work because such a system will pretty much require most programs be distributed with their own, tailored profile with "appropriate" permissions. Naturally, instead of actually trying to do it properly, developers will just distribute their software with a profile that lets their application do anything (as will every little flash game, and the like).
You can't secure a system where ignorant end users are allowed to make critical security decisions. Not now, not ever.
If we had secure operating systems, you think all the A/V and other companies would make money?
Of course they would. AV and anti-malware software isn't there to replace OS security, it's there for when the OS security has already been circumvented (typically deliberately by the end user).
No amount of OS security will protect the machine from an end-user deliberately running malicious code.
That's a long time for MS to wait. If the Asus prediction [engadget.com] is anywhere near correct, there'll be a whole lot of people learning that Linux isn't just some command line based geek OS.
If Asus's prediction is correct, that merely highlights how wrong you are.
Something in the region of 220 million laptops were sold in 2007. Even assuming that figure doesn't grow in 2008 (ridiculously conservative), Eee PC sales will represent less than 2% of them. Assuming half of those are running Linux (probably a significant overestimate), and that every single one of those people who don't get an Eee with Windows realises that it runs "Linux", then about 1% of *laptop buyers* will "learn that Linux isn't just some command line based geek OS".
Take out all the assumptions stacked in Linux's favour, and you'll be lucky to see a tenth of a percent of people "learning that Linux isn't just some command line based geek OS" - and probably only a tenth of them actually caring.
I don't think Microsoft have too much to worry about. XP will easily tide them over for the 6-9 months it will take the hardware to be able to run Vista well.
The historical pattern is to both, price and performance, get better with time, as power consuption increases.
No, the historical pattern is for performance at a given price point to increase substantially faster than price points reduce.
For example, in ten years, the price point for a "cheap computer" has halved from about US$1000 to about US$500. In that same time, performance has increased on the order of 15x.
But the current market is clearly different from the main history of computing, since we are talking about mobile devices, and history is focused on non-mobile ones.
No, we're talking about laptops, and they've been around for at least twenty years demonstrating the same pattern we're going to see again. The $400ish price point is going to stay relatively solid, but the amount of computing power at that price point will have doubled in 6-9 months and then doubled again another 12-18 months after that, at which point you *might* see the price point start to reduce a bit.
Microsoft is blaming its disappointing third quarter client software sales on factors that exclude Vista. Meanwhile, though, a variety of evidence indicates that many Windows XP users aren't upgrading, but switching to Mac or Linux.
You'll need to do better than that. A reasonable (but still far from conclusive) start would be sales numbers showing Eee PCs running Linux and Macs making up a meaningful proportion of purchased computers.
Asus (EEE PC), Ubuntu and Apple have taken significant desktop market shares away from Microsoft.
No, they have not.
I challenge you to find even the slightest bit of evidence to demonstrate otherwise.
(Apple might just barely qualify for taking away a small part of Microsoft's desktop market share. The other two wouldn't even qualify as rounding errors.)
Because, to use the Slashdot vernacular, Netcraft confirms it !
Features and/or power increase much, much faster than prices decrease. That has been the continual and repeating pattern for at least the last couple of decades I've been paying attention.
Do you remember when it was a big deal when PCs dropped under $1000?
And this handily demonstrates my point. It was roughly ten years ago that "average" PCs broke the "$1000 barrier". Today, a decade on, that price point has barely halved (if that) while the power of the machines has increased more than an order of magnitude.
The point here is that price points remain fairly static and move over a time period of years, whereas specifications are relatively volatile, and move over a time period of months.
Yes, specs will continue to go up, but I'd say that computer prices seem to be on a steady decline - at least when you look at what's available on the low end.
Prices aren't really declining any faster than they were - if anything the rate is slowing. However, computing power at a given price point continues to advance as rapidly as it ever has, which is why in 6-12 months the Eee PC will still cost about the same, but have at least twice as much "grunt".
1. A beefy machine running Vista selling at the same $350-$550 range that is succeeding today.
It won't be a "beefy machine". It'll be a baseline machine. Just like the Eee and friends are today.
2. The same specs as today but retailing for $200-$400.
Your assumption that the specs will stay the same and the price will drop, rather than the price staying the same and the specs increasing, is highly questionable. The latter is the far more typical historical pattern.
Personally, I can't wait. The Eee PC today isn't really an option as an only machine. With a ~1.4Ghz dual core and 2-4G of RAM, however, it realistically is. Especially with a cheap docking station/monitor stand for when you're sitting at a desk and don't want to squint at the tiny screen.
It already has impacted the OS market...Microsoft has had to promise to keep XP alive to serve the "Netbook" market. They cannot conceed this segment to Linux without a fight and Vista will never run on these little puppies.
This class of machine will be quite capable of running Vista well within 12 months, probably less.
It seems like when you're dealing with price points of one to several hundred dollars, this is a big deal for free software, specifically Linux. When you're talking about adding anywhere from 25% to 100% of the cost of the computer just for the operating system, it paints things in a different light.
Which is why, if necessary, Microsoft will just have a specially-priced version of Windows just for the OEMs selling them.
That, and you'd have to put an older (soon to be non-supported) version of Windows (XP) on the thing. I can't see these running Vista anytime in the near future.
By the end of this year, these machines will baseline with dual-core 1Ghz+ CPUs and 2-4G of RAM. More than enough to run Vista. Which is right around the time I expect they'll _really_ pickup in the market - when they become useful as the only machine most people need to have, rather than a toy that must be supplemented with a "primary" computer.
Should be interesting to see how this impacts the OS playing field...
I'm asking because I'm thinking that Speed Racer is primarily a U.S. childhood memory keepsake.
I don't recall ever seeing much of it in Australia either, although I do remember watching (off the top of my head): Astroboy, Robotech, G-Force/Battle of the Planets, Cities of Gold, Transformers, Dangermouse, Inspector Gadget, TMNT, Ulysses 31.
So either it wasn't shown, wasn't shown much, or wasn't very popular.
Which was kinda my point - that there are advantages to SANs, but performance is generally not a major one (unless you're getting into environments that are far beyond average).
My response was directed at this:
SANs can deliver I/O at speeds local disks can only dream of.
The point being that to get a SAN to deliver "I/O at speeds local disks can only dream of" you need to spend a very, very large amount of money - far more than most companies can afford to and in great disproportion to the benefits they would see.
A smaller brother, both in size and power, to the MacPro, priced between the lowest and highest price iMac would probably be a very popular item they should add to their list.
Some proper business-class laptops (ie: with docking stations) in the 12-13" and 15" classes, would be bigger hits, IMHO.
The repairability of Apple machines seems to be very oddly hit or miss - the original white G5 iMacs are a pleasure to work on; the main components are even held on their own sub-assembly meaning that it's a relatively quick job to separate the computer from the LCD despite them being in the same case. Suddenly they switch to the setup you described above and it's almost more effort than its worth to deal with the things.
Yes, but that probably saved 1/3" in depth, so of _course_ it's worth the servicability tradeoff;).
There are a lot of comments here that are making the false assumption that if it weren't for computers, we would be doing the same things, but with paper. Then, they're concluding that computers save paper.
But this is no different to your assumption in the opposite direction. That is, that we'd be able to work at the same levels without the advantages that computers deliver, and/or that those same advantages could be delivered using less paper (although nothing else but the value of the internet as a research tool should blow that idea out of the water).
We have easily quadrupled the use of paper since we changed over to computers.
Correlation != causation. How much would you have increased its use if you *hadn't* started using computers ? How much lower would productivity be without the volume and rapidity of communications that computers allow ? Have your volume/profits/productivity "quadrupled" since you changed to computers ?
All you need to do is compare the relatively growths of non-verbal communications to paper usage. I think you'll find the former far, far outstrips the latter.
Aside from that, I/O architecture is still changing as we speak. SANs can deliver I/O at speeds local disks can only dream of.
You have to spend a shitload of money on a SAN to get performance even as good as - let alone better than - local disks. Their raw price/performance is massively worse than local disk.
Eight 2.5" 146G, 10k RPM drives in something like a Dell PE2950 will add about US$1600 to the price tag. For the same money you'll get 3 - maybe 4 - (slower) 3.5" 146G, 10k RPM drives into your SAN. To say nothing of the supporting infrastructure cost - FC switches, cables, HBAs, etc.
You need to spend on the order of 3-4x as much money to get close to similar performance out of a SAN than local disk.
There are very good reasons to get a SAN - but performance/$ is rarely one of them.
I guess the difference is that you can get cheaper Vista laptops with lower specs that the cheapest Mac?
The difference is there are a few gaping holes in Apple's hardware lineup. So if you needs happen to fall into one of them, you either need to a) settle for a lesser machine than you want, b) spend a disproportionately larger amount of money to get more than you want, or c) give up because Apple simply doesn't have anything at all in that market space (eg: laptop with a docking station).
Macs are reasonably well priced if what you want is exactly the same as one of their products. If you want to think a little bit outside of their boxes, however, you're screwed.
[...] who wants to be managing something with a GUI under Windows, when you could be SSHing in and changing all the settings.
Me. More accurately, me on behalf of my staff, because a GUI-based system is orders of magnitude more robust than one where they can make typos, or other, similar, involuntary mistakes.
So what you're saying is I can't blame Bill anymore?
Dude, this is Slashdot. If your 108-year-old grandmother died thirty years after being diagnosed with terminal cancer, you could still blame it on "Bill" and get modded "+1 Insightful".
Slashdot Mirror
The whole way security is treated needs to be changed. Having root and an ordinary user just doesn't offer the level of granularity that users need. As a user I want to be able to do everything on my computer, what's really needed is fine grained access control per program. Of course, that has issues with users having to grant those privileges but you could have profiles. Imagine installing Evolution or something and it pops up and says "This software says it's a mail client, does that sound right to you?" and then what privileges it gets granted will be set by a "mail client" profile already installed on the system.
It won't work because such a system will pretty much require most programs be distributed with their own, tailored profile with "appropriate" permissions. Naturally, instead of actually trying to do it properly, developers will just distribute their software with a profile that lets their application do anything (as will every little flash game, and the like).
You can't secure a system where ignorant end users are allowed to make critical security decisions. Not now, not ever.
If we had secure operating systems, you think all the A/V and other companies would make money?
Of course they would. AV and anti-malware software isn't there to replace OS security, it's there for when the OS security has already been circumvented (typically deliberately by the end user).
No amount of OS security will protect the machine from an end-user deliberately running malicious code.
That's a long time for MS to wait. If the Asus prediction [engadget.com] is anywhere near correct, there'll be a whole lot of people learning that Linux isn't just some command line based geek OS.
If Asus's prediction is correct, that merely highlights how wrong you are.
Something in the region of 220 million laptops were sold in 2007. Even assuming that figure doesn't grow in 2008 (ridiculously conservative), Eee PC sales will represent less than 2% of them. Assuming half of those are running Linux (probably a significant overestimate), and that every single one of those people who don't get an Eee with Windows realises that it runs "Linux", then about 1% of *laptop buyers* will "learn that Linux isn't just some command line based geek OS".
Take out all the assumptions stacked in Linux's favour, and you'll be lucky to see a tenth of a percent of people "learning that Linux isn't just some command line based geek OS" - and probably only a tenth of them actually caring.
I don't think Microsoft have too much to worry about. XP will easily tide them over for the 6-9 months it will take the hardware to be able to run Vista well.
The historical pattern is to both, price and performance, get better with time, as power consuption increases.
No, the historical pattern is for performance at a given price point to increase substantially faster than price points reduce.
For example, in ten years, the price point for a "cheap computer" has halved from about US$1000 to about US$500. In that same time, performance has increased on the order of 15x.
But the current market is clearly different from the main history of computing, since we are talking about mobile devices, and history is focused on non-mobile ones.
No, we're talking about laptops, and they've been around for at least twenty years demonstrating the same pattern we're going to see again. The $400ish price point is going to stay relatively solid, but the amount of computing power at that price point will have doubled in 6-9 months and then doubled again another 12-18 months after that, at which point you *might* see the price point start to reduce a bit.
Microsoft is blaming its disappointing third quarter client software sales on factors that exclude Vista. Meanwhile, though, a variety of evidence indicates that many Windows XP users aren't upgrading, but switching to Mac or Linux.
You'll need to do better than that. A reasonable (but still far from conclusive) start would be sales numbers showing Eee PCs running Linux and Macs making up a meaningful proportion of purchased computers.
Asus (EEE PC), Ubuntu and Apple have taken significant desktop market shares away from Microsoft.
No, they have not.
I challenge you to find even the slightest bit of evidence to demonstrate otherwise.
(Apple might just barely qualify for taking away a small part of Microsoft's desktop market share. The other two wouldn't even qualify as rounding errors.)
The version in retail boxes won't install or run on generic hardware.
Depends on your definition of "generic". With appropriate "generic" hardware selection, you can install and run a standard retail copy of OS X.
How do you figure?
Because, to use the Slashdot vernacular, Netcraft confirms it !
Features and/or power increase much, much faster than prices decrease. That has been the continual and repeating pattern for at least the last couple of decades I've been paying attention.
Do you remember when it was a big deal when PCs dropped under $1000?
And this handily demonstrates my point. It was roughly ten years ago that "average" PCs broke the "$1000 barrier". Today, a decade on, that price point has barely halved (if that) while the power of the machines has increased more than an order of magnitude.
The point here is that price points remain fairly static and move over a time period of years, whereas specifications are relatively volatile, and move over a time period of months.
Yes, specs will continue to go up, but I'd say that computer prices seem to be on a steady decline - at least when you look at what's available on the low end.
Prices aren't really declining any faster than they were - if anything the rate is slowing. However, computing power at a given price point continues to advance as rapidly as it ever has, which is why in 6-12 months the Eee PC will still cost about the same, but have at least twice as much "grunt".
1. A beefy machine running Vista selling at the same $350-$550 range that is succeeding today.
It won't be a "beefy machine". It'll be a baseline machine. Just like the Eee and friends are today.
2. The same specs as today but retailing for $200-$400.
Your assumption that the specs will stay the same and the price will drop, rather than the price staying the same and the specs increasing, is highly questionable. The latter is the far more typical historical pattern.
Personally, I can't wait. The Eee PC today isn't really an option as an only machine. With a ~1.4Ghz dual core and 2-4G of RAM, however, it realistically is. Especially with a cheap docking station/monitor stand for when you're sitting at a desk and don't want to squint at the tiny screen.
It already has impacted the OS market...Microsoft has had to promise to keep XP alive to serve the "Netbook" market. They cannot conceed this segment to Linux without a fight and Vista will never run on these little puppies.
This class of machine will be quite capable of running Vista well within 12 months, probably less.
It seems like when you're dealing with price points of one to several hundred dollars, this is a big deal for free software, specifically Linux. When you're talking about adding anywhere from 25% to 100% of the cost of the computer just for the operating system, it paints things in a different light.
Which is why, if necessary, Microsoft will just have a specially-priced version of Windows just for the OEMs selling them.
That, and you'd have to put an older (soon to be non-supported) version of Windows (XP) on the thing. I can't see these running Vista anytime in the near future.
By the end of this year, these machines will baseline with dual-core 1Ghz+ CPUs and 2-4G of RAM. More than enough to run Vista. Which is right around the time I expect they'll _really_ pickup in the market - when they become useful as the only machine most people need to have, rather than a toy that must be supplemented with a "primary" computer.
Should be interesting to see how this impacts the OS playing field...
Very little change.
Why are these machines so expensive?
Because price only scales up with features, not down.
I'm asking because I'm thinking that Speed Racer is primarily a U.S. childhood memory keepsake.
I don't recall ever seeing much of it in Australia either, although I do remember watching (off the top of my head): Astroboy, Robotech, G-Force/Battle of the Planets, Cities of Gold, Transformers, Dangermouse, Inspector Gadget, TMNT, Ulysses 31.
So either it wasn't shown, wasn't shown much, or wasn't very popular.
Not to mention the fact that Anime tend towards trying to make animation as real as possible [...]
Say what ? O.o
My response was directed at this:
SANs can deliver I/O at speeds local disks can only dream of.The point being that to get a SAN to deliver "I/O at speeds local disks can only dream of" you need to spend a very, very large amount of money - far more than most companies can afford to and in great disproportion to the benefits they would see.
A smaller brother, both in size and power, to the MacPro, priced between the lowest and highest price iMac would probably be a very popular item they should add to their list.
Some proper business-class laptops (ie: with docking stations) in the 12-13" and 15" classes, would be bigger hits, IMHO.
The repairability of Apple machines seems to be very oddly hit or miss - the original white G5 iMacs are a pleasure to work on; the main components are even held on their own sub-assembly meaning that it's a relatively quick job to separate the computer from the LCD despite them being in the same case. Suddenly they switch to the setup you described above and it's almost more effort than its worth to deal with the things.
Yes, but that probably saved 1/3" in depth, so of _course_ it's worth the servicability tradeoff ;).
There are a lot of comments here that are making the false assumption that if it weren't for computers, we would be doing the same things, but with paper. Then, they're concluding that computers save paper.
But this is no different to your assumption in the opposite direction. That is, that we'd be able to work at the same levels without the advantages that computers deliver, and/or that those same advantages could be delivered using less paper (although nothing else but the value of the internet as a research tool should blow that idea out of the water).
We have easily quadrupled the use of paper since we changed over to computers.
Correlation != causation. How much would you have increased its use if you *hadn't* started using computers ? How much lower would productivity be without the volume and rapidity of communications that computers allow ? Have your volume/profits/productivity "quadrupled" since you changed to computers ?
All you need to do is compare the relatively growths of non-verbal communications to paper usage. I think you'll find the former far, far outstrips the latter.
DO you have any stats to back this up with?
How many people do you know who print every email they send and receive ?
Aside from that, I/O architecture is still changing as we speak. SANs can deliver I/O at speeds local disks can only dream of.
You have to spend a shitload of money on a SAN to get performance even as good as - let alone better than - local disks. Their raw price/performance is massively worse than local disk.
Eight 2.5" 146G, 10k RPM drives in something like a Dell PE2950 will add about US$1600 to the price tag. For the same money you'll get 3 - maybe 4 - (slower) 3.5" 146G, 10k RPM drives into your SAN. To say nothing of the supporting infrastructure cost - FC switches, cables, HBAs, etc.
You need to spend on the order of 3-4x as much money to get close to similar performance out of a SAN than local disk.
There are very good reasons to get a SAN - but performance/$ is rarely one of them.
I guess the difference is that you can get cheaper Vista laptops with lower specs that the cheapest Mac?
The difference is there are a few gaping holes in Apple's hardware lineup. So if you needs happen to fall into one of them, you either need to a) settle for a lesser machine than you want, b) spend a disproportionately larger amount of money to get more than you want, or c) give up because Apple simply doesn't have anything at all in that market space (eg: laptop with a docking station).
Macs are reasonably well priced if what you want is exactly the same as one of their products. If you want to think a little bit outside of their boxes, however, you're screwed.
Apple did not take BSD and "slap a bubbly gui on top of it". You should consider learning something about software one day.
Excellent point. They actually took NeXTSTEP and slapped a bubbly GUI on top of it.
[...] who wants to be managing something with a GUI under Windows, when you could be SSHing in and changing all the settings.
Me. More accurately, me on behalf of my staff, because a GUI-based system is orders of magnitude more robust than one where they can make typos, or other, similar, involuntary mistakes.
So what you're saying is I can't blame Bill anymore?
Dude, this is Slashdot. If your 108-year-old grandmother died thirty years after being diagnosed with terminal cancer, you could still blame it on "Bill" and get modded "+1 Insightful".
It's called an iMac?
No, the iMac is a different class of machine - an all-in-one. May as well just buy a laptop.