Even if you look at NASA as a pork barrel tool that feed the aerospace industry, it's a lot better to feed them thru NASA than it is to feed them through the military.
In the end, less people get hurt, less people get really pissed of and we end up with better pictures.
There is no customer loyalty when the customer intends to build something by the millions. Even a minute difference, a manufacturing step, a screw that's not required represents a huge economy when you factor in how many units would be built.
You may chose to stick with Dell (or HP or Sun or IBM) for your 5 new servers and ignore a $10 pricing difference. You should certainly not ignore a $10 difference a piece for your next batch of 1,000,000 parts for manufacture. Customer loyalty goes only as far as not to hurt the bottom line and the suicidal have really short careers.
As it is, Intel stood to gain nothing from partnering with OLPC at this stage - it would have to compete in price with AMD on a product that competes with the Classmate instead of being the sole provider of components for the Classmate series. They didn't achieve the dominance they currently enjoy by shooting themselves in the foot frequently.
And no... Warm and fuzzy feelings would not impact their sales to their real clients (Dell, Sun, IBM, HP...)
"however the technologies of different networks differ enough that you can't take a Sprint phone to AT&T."
Not exactly. Most phones are either CDMA or GSM. I always wondered how it happened that in the USA the phone customers get so much more screwed than in any other place I know and seem uninformed about it - it's not incompatible tech - it's that telcos configure the phones they sell not to be able to connect to any other network.
As in most places, you can easily buy an unblocked GSM phone you can use in any network you chose.
"XP isn't a factor for me. But it might be a factor in the availability which is why I cannot find them for sale. That is something that makes the OLPC so attractive."
I used to joke they should build it around a fast multi-core ARM11 with integrated caches and FPU/GPU and make it utterly Windows-proof so there is no risk Microsoft will ever ruin the fun.
In fact, this was a re-hash of half-jokes I made in the beginning of the "Computador para Todos" ("Computers for Everyone", an incentive program Brazilian government engaged in, to make inexpensive computers available that, at one point, generated some interesting hardware prototypes. Now it is a tax-exemption program that works very well.
"Well, it was my understanding that some of the stuff was created specifically for the project"
Yes. Lots of stuff was developed for the XO, but there is nothing precluding those components from hitting the shelves. The biggest hurdle probably is that few of them have Windows drivers;-)
"I seem to be having problems finding the EeePC for under $400"
I think there was a version of the EeePC for $299, with 4 GB of storage instead of 8. The SSD is, certainly, a major factor on unit price and 5 GB seems to be a requirement for it to run Windows XP.
"They don't even need to sell the entire thing. Something as simple as selling developed components or licensing tech that could in effect pay for the distribution of the laptops to the target audience would be both beneficial and attract the type of participation that could make it a reality."
That's what going to happen with the screen http://hardware.slashdot.org/article.pl?sid=08/01/01/1324240. AMD sells the processor, Marvell sells all the wireless components and there are literally dozens of manual chargers for phones. All components of the OLPC are readily available if you want to build, say, a thousand of them and many are available in stores.
You complain nobody is making a cheap $200 notebook. There is the EeePC from ASUS, which does not employ XO's cool display, but has a better keyboard. If sales are strong, they have proved a market exists for cheap ultraportable notebooks.
Have some patience. Someone is bound to pursue that market.
And, BTW, the last real interesting innovation I saw from Microsoft in the desktop metaphor was the Mail and News client that came with IE 3 (IIRC). It sort of promoted e-mail to a document type and employed Explorer extensions to show it properly as e-mail. That was a beautiful way to show it and it's a very clever (if obvious) idea. For Microsoft, it was dangerously subversive too.
I really wish Gnome could do that on the desktop level, without resorting to full-fledged applications like Evolution. E-mail, appointments, contacts really should be treated the same way as files.
I agree Adobe can't complain much, because Flash got where it is now (just like Internet Explorer) thanks to Microsoft bundling it with Windows. Microsoft used its monopoly on Windows to create a smaller one (but still very profitable) for Flash.
But, anyway, Microsoft still has life-and-death power over it. They may decide to stop bundling Flash starting with IE8 and start bundling Silverlight instead. If they implemented a Flash player in Silverlight, that would be a death blow to Flash as we now know it. Microsoft would, in effect, use their monopoly on Windows to destroy the monopoly they gave Flash and to create a new one around Silverlight. They already did that at least once with the Netware client for Windows 3.
Adobe makes a lot of money from Flash development tools. It would have a hard time selling them if "Vista + 1" machines would not be able to use Flash-based sites.
Netscape only thrived because Microsoft looked the other way for too long and failed to recognize the threat new monopolies could represent. Microsoft is fully awake now and will not let anyone threaten its dominance.
You can also do both - you back-up important files from inside the VM and back-up the VM itself from the host environment.
A third option would be to create a filesystem snapshot on a storage server and back the snapshot on the storage. That would make restoring individual files simpler than from a VM snapshot (which seems somewhat scary on VMWare)
Depending on your choice of virtualization (I use OpenVZ) recovering files from a VM snapshot is trivial (as all the VM files are on a file system visible to the host environment.
"Personally I'll stick with my Microsoft keyboard and mouse which work surprisingly well with Windows, Linux, and OSX."
That's something Microsoft does right. When I have to enumerate the best products Microsoft makes, I say, in that order, the Natural keyboard series, their mice and SQL Server (which is a respectable database server, even if it runs on a less than respectable OS).
Those three are good.
As for the rest... Well... They did the Apple II+ BASIC, didn't they? That was cool.
"Hmmm... Ok so tell me how often are you going to be visiting the Microsoft website if you happen to be a Linux and Firefox user?
Probably 0...."
This pretty much isolates them into the ranks of Windows users (who don't need to be converted into Windows users) and Macintosh users (good luck on that, MS). If they successfully prevent any casual Linux user (those who got Linux with their PCs) from accessing their web site, they may be preventing some people to switch to Windows. This is a Good Thing.
Also, this is probably not going to happen. It would be a pretty dumb move, any way I try to see it.
Most probably, if a Silverlight-hostile browser is detected, they will redirect to a simpler HTML page that tells you the wonders of Silverlight are not available to you until you sacrifice your computer in Silverlight's altar.
"Good point. People on/. should stop trying to talk about how great Linux and MacOSX are. I mean, if they were so great they would be dominant already."
If Microsoft's dominance had anything to do with software quality and not with barely legal tactics of coercing OEMs into bundling their and their software only, sabotaging Windows so it would not work properly with DR-DOS, and generally abusing one monopoly to create more monopolies, your comment would have some measure of correctness.
WFWG made obvious (to Novell's disgrace) people didn't want file servers - they wanted to share files and printers. Excel was respectable. Word (first on Mac, then on Windows) was decent. Multiplan and Word were even honest products on DOS and on Macintosh. Windows brought some GUI multitasking for those who couldn't afford to run Unix and X. I did a lot of Applesoft BASIC code during college.
Unfortunately, the real world is not like that. This Microsoft is not the same company it was on the 70s, 80s and early 90s.
I have no problem with religions being taught in schools, but only as far as they are no treated as science and only as long the curriculum provides an unbiased view of as many as possible different religions and compares their differences, similarities and explains why humans seem to feel the urge to try to understand the universe around them by resorting to supersti^H^H^H^H^H^H^H^Hreligious explanations.
If the OLPC is ever to reach the US$ 100 target price (even if we give it the adjustment for a shrinking dollar) it is via production volume of its key parts. Making them available to other companies via a for-profit seems to be the best way to do it.
It was always pretty obvious to me that, even if the XO itself does not bring a huge change, its technologies and its "less is enough" approach are bound to make a massive change to a very monotonous market.
"the idea of a virtual file system that would replace the file/folder metaphor with something resembling the filing system of email clients, with virtual folders, tags, etc."
This does not belong into the OS. It belongs in the application/desktop realm.
"The entire desktop metaphor should also be ditched in favor of something else"
This, again, has nothing to do with the OS.
"for example, why won't the OS auto-save each document I'm working on every 1-5 minutes so I can recover from mistakenly overwriting a file or saving it when I intended to discard changes?"
Another application feature that does not belong into the OS.
"Microsoft tried to do some of it with WinFS and failed"
Microsoft tried a lot of things, many of them flawed ideas. WinFS is one of them. Only applications (or a desktop infrastructure relying on them) would be able to extract meaningful metadata and relationships from the files. The catch is that your data could be accessed only by the applications that understand them and metadada being created by different computers on different OSs could make removable media a nightmare.
"OSX now has "time machine" to recover files but they could go further."
Something on the lines of file versioning, VMS-style, or ZFS, perhaps. Apple also has its fair share of bad ideas (and some brilliant ones - the stationery thing on Lisa or the whole Newton "soup" idea is should be tried again), like bundling the lines between GUI and OS.
If I had to pick one reason why computing is in its current boring state, I would point to the x86 PC. I remember the innovative architectures I saw in the 70s and 80s, ranging from multi-Z-80s with math coprocessors or Lisp machines and lots of different mainframe structures. I would love to see a desktop computer based on the Cell, a Niagara 2 or a Cavium Octeon-style with, say, two vector units instead of an integer core. I would love to see hardware assisted garbage collection like Lisp machines had in the 80s. The PC industry gave us cheap PCs (and the laptop I am writing this on) but its compatibility shackles weighted progress down for decades. The current panorama is completely different from the thriving diverse environment we had decades ago. It would be stupid to blame it on FOSS.
Industries rise and fall to the winds of technological advances. Textiles, horse-pulled carriages, ice for refrigeration. All intangibles, like software, books, music and movies are next. No surprise at all.
Slashdot Mirror
Even if you look at NASA as a pork barrel tool that feed the aerospace industry, it's a lot better to feed them thru NASA than it is to feed them through the military.
In the end, less people get hurt, less people get really pissed of and we end up with better pictures.
There is no customer loyalty when the customer intends to build something by the millions. Even a minute difference, a manufacturing step, a screw that's not required represents a huge economy when you factor in how many units would be built.
You may chose to stick with Dell (or HP or Sun or IBM) for your 5 new servers and ignore a $10 pricing difference. You should certainly not ignore a $10 difference a piece for your next batch of 1,000,000 parts for manufacture. Customer loyalty goes only as far as not to hurt the bottom line and the suicidal have really short careers.
As it is, Intel stood to gain nothing from partnering with OLPC at this stage - it would have to compete in price with AMD on a product that competes with the Classmate instead of being the sole provider of components for the Classmate series. They didn't achieve the dominance they currently enjoy by shooting themselves in the foot frequently.
And no... Warm and fuzzy feelings would not impact their sales to their real clients (Dell, Sun, IBM, HP...)
"however the technologies of different networks differ enough that you can't take a Sprint phone to AT&T."
Not exactly. Most phones are either CDMA or GSM. I always wondered how it happened that in the USA the phone customers get so much more screwed than in any other place I know and seem uninformed about it - it's not incompatible tech - it's that telcos configure the phones they sell not to be able to connect to any other network.
As in most places, you can easily buy an unblocked GSM phone you can use in any network you chose.
In a perfect free market, there is no such thing as a loyal customer.
"XP isn't a factor for me. But it might be a factor in the availability which is why I cannot find them for sale. That is something that makes the OLPC so attractive."
I used to joke they should build it around a fast multi-core ARM11 with integrated caches and FPU/GPU and make it utterly Windows-proof so there is no risk Microsoft will ever ruin the fun.
In fact, this was a re-hash of half-jokes I made in the beginning of the "Computador para Todos" ("Computers for Everyone", an incentive program Brazilian government engaged in, to make inexpensive computers available that, at one point, generated some interesting hardware prototypes. Now it is a tax-exemption program that works very well.
That were, of course, absolutely Windows-proof.
"Well, it was my understanding that some of the stuff was created specifically for the project"
;-)
Yes. Lots of stuff was developed for the XO, but there is nothing precluding those components from hitting the shelves. The biggest hurdle probably is that few of them have Windows drivers
"I seem to be having problems finding the EeePC for under $400"
I think there was a version of the EeePC for $299, with 4 GB of storage instead of 8. The SSD is, certainly, a major factor on unit price and 5 GB seems to be a requirement for it to run Windows XP.
"They don't even need to sell the entire thing. Something as simple as selling developed components or licensing tech that could in effect pay for the distribution of the laptops to the target audience would be both beneficial and attract the type of participation that could make it a reality."
That's what going to happen with the screen http://hardware.slashdot.org/article.pl?sid=08/01/01/1324240. AMD sells the processor, Marvell sells all the wireless components and there are literally dozens of manual chargers for phones. All components of the OLPC are readily available if you want to build, say, a thousand of them and many are available in stores.
You complain nobody is making a cheap $200 notebook. There is the EeePC from ASUS, which does not employ XO's cool display, but has a better keyboard. If sales are strong, they have proved a market exists for cheap ultraportable notebooks.
Have some patience. Someone is bound to pursue that market.
And, BTW, the last real interesting innovation I saw from Microsoft in the desktop metaphor was the Mail and News client that came with IE 3 (IIRC). It sort of promoted e-mail to a document type and employed Explorer extensions to show it properly as e-mail. That was a beautiful way to show it and it's a very clever (if obvious) idea. For Microsoft, it was dangerously subversive too.
I really wish Gnome could do that on the desktop level, without resorting to full-fledged applications like Evolution. E-mail, appointments, contacts really should be treated the same way as files.
I agree Adobe can't complain much, because Flash got where it is now (just like Internet Explorer) thanks to Microsoft bundling it with Windows. Microsoft used its monopoly on Windows to create a smaller one (but still very profitable) for Flash.
But, anyway, Microsoft still has life-and-death power over it. They may decide to stop bundling Flash starting with IE8 and start bundling Silverlight instead. If they implemented a Flash player in Silverlight, that would be a death blow to Flash as we now know it. Microsoft would, in effect, use their monopoly on Windows to destroy the monopoly they gave Flash and to create a new one around Silverlight. They already did that at least once with the Netware client for Windows 3.
Adobe makes a lot of money from Flash development tools. It would have a hard time selling them if "Vista + 1" machines would not be able to use Flash-based sites.
Netscape only thrived because Microsoft looked the other way for too long and failed to recognize the threat new monopolies could represent. Microsoft is fully awake now and will not let anyone threaten its dominance.
You can also do both - you back-up important files from inside the VM and back-up the VM itself from the host environment.
A third option would be to create a filesystem snapshot on a storage server and back the snapshot on the storage. That would make restoring individual files simpler than from a VM snapshot (which seems somewhat scary on VMWare)
Depending on your choice of virtualization (I use OpenVZ) recovering files from a VM snapshot is trivial (as all the VM files are on a file system visible to the host environment.
The simplest solution is to run back-up software on the VM itself.
Another very simple way is taking a snapshot of the running VM and backint that up.
A properly configured router and a proxy server are all anyone needs for this.
Why bother installing spyware to track web usage if you control the network?
Let's just call it Microsoft Blackbird 2.0.
It feels like 1995 all over again.
"Personally I'll stick with my Microsoft keyboard and mouse which work surprisingly well with Windows, Linux, and OSX."
That's something Microsoft does right. When I have to enumerate the best products Microsoft makes, I say, in that order, the Natural keyboard series, their mice and SQL Server (which is a respectable database server, even if it runs on a less than respectable OS).
Those three are good.
As for the rest... Well... They did the Apple II+ BASIC, didn't they? That was cool.
"Hmmm... Ok so tell me how often are you going to be visiting the Microsoft website if you happen to be a Linux and Firefox user?
;-)
Probably 0...."
This pretty much isolates them into the ranks of Windows users (who don't need to be converted into Windows users) and Macintosh users (good luck on that, MS). If they successfully prevent any casual Linux user (those who got Linux with their PCs) from accessing their web site, they may be preventing some people to switch to Windows. This is a Good Thing.
Also, this is probably not going to happen. It would be a pretty dumb move, any way I try to see it.
Most probably, if a Silverlight-hostile browser is detected, they will redirect to a simpler HTML page that tells you the wonders of Silverlight are not available to you until you sacrifice your computer in Silverlight's altar.
That is more like Microsoft.
"Good point. People on /. should stop trying to talk about how great Linux and MacOSX are. I mean, if they were so great they would be dominant already."
If Microsoft's dominance had anything to do with software quality and not with barely legal tactics of coercing OEMs into bundling their and their software only, sabotaging Windows so it would not work properly with DR-DOS, and generally abusing one monopoly to create more monopolies, your comment would have some measure of correctness.
WFWG made obvious (to Novell's disgrace) people didn't want file servers - they wanted to share files and printers. Excel was respectable. Word (first on Mac, then on Windows) was decent. Multiplan and Word were even honest products on DOS and on Macintosh. Windows brought some GUI multitasking for those who couldn't afford to run Unix and X. I did a lot of Applesoft BASIC code during college.
Unfortunately, the real world is not like that. This Microsoft is not the same company it was on the 70s, 80s and early 90s.
Except than when you have a monopoly, it's illegal to do so.
Thanks. I thoroughly enjoy being called subversive. :-)
I have no problem with religions being taught in schools, but only as far as they are no treated as science and only as long the curriculum provides an unbiased view of as many as possible different religions and compares their differences, similarities and explains why humans seem to feel the urge to try to understand the universe around them by resorting to supersti^H^H^H^H^H^H^H^Hreligious explanations.
For ASCII-happiness, I usually write 3.5 and 5.25, but that's just me.
It's 5¼, not 5½.
The world would be a better place if people were given proper scientific education.
You know... That where you observe facts, formulate hypotheses and try to invalidate them through experiments.
If the OLPC is ever to reach the US$ 100 target price (even if we give it the adjustment for a shrinking dollar) it is via production volume of its key parts. Making them available to other companies via a for-profit seems to be the best way to do it.
;-)
It was always pretty obvious to me that, even if the XO itself does not bring a huge change, its technologies and its "less is enough" approach are bound to make a massive change to a very monotonous market.
Let's hope it's the next Apple II
"the idea of a virtual file system that would replace the file/folder metaphor with something resembling the filing system of email clients, with virtual folders, tags, etc."
This does not belong into the OS. It belongs in the application/desktop realm.
"The entire desktop metaphor should also be ditched in favor of something else"
This, again, has nothing to do with the OS.
"for example, why won't the OS auto-save each document I'm working on every 1-5 minutes so I can recover from mistakenly overwriting a file or saving it when I intended to discard changes?"
Another application feature that does not belong into the OS.
"Microsoft tried to do some of it with WinFS and failed"
Microsoft tried a lot of things, many of them flawed ideas. WinFS is one of them. Only applications (or a desktop infrastructure relying on them) would be able to extract meaningful metadata and relationships from the files. The catch is that your data could be accessed only by the applications that understand them and metadada being created by different computers on different OSs could make removable media a nightmare.
"OSX now has "time machine" to recover files but they could go further."
Something on the lines of file versioning, VMS-style, or ZFS, perhaps. Apple also has its fair share of bad ideas (and some brilliant ones - the stationery thing on Lisa or the whole Newton "soup" idea is should be tried again), like bundling the lines between GUI and OS.
If I had to pick one reason why computing is in its current boring state, I would point to the x86 PC. I remember the innovative architectures I saw in the 70s and 80s, ranging from multi-Z-80s with math coprocessors or Lisp machines and lots of different mainframe structures. I would love to see a desktop computer based on the Cell, a Niagara 2 or a Cavium Octeon-style with, say, two vector units instead of an integer core. I would love to see hardware assisted garbage collection like Lisp machines had in the 80s. The PC industry gave us cheap PCs (and the laptop I am writing this on) but its compatibility shackles weighted progress down for decades. The current panorama is completely different from the thriving diverse environment we had decades ago. It would be stupid to blame it on FOSS.
This is business as usual.
Industries rise and fall to the winds of technological advances. Textiles, horse-pulled carriages, ice for refrigeration. All intangibles, like software, books, music and movies are next. No surprise at all.