The article on Wikipedia is actually pretty good. There may be some transitory or faint effects caused by subliminal messages. Advertisers have been trying to capitalize on this possibility for 49 years.
If subliminal messages had any significant effect we would know about it. They've been trying for years.
There have been interesting claims at subliminal messages in popular music. KAKE-TV in Wichita, Kansas, used a subliminal message to attempt catch the BTK killer, but it had no perceptible effect.
Okay, this is one of the most informative posts ever. People are thinking this is Williams, the original guy who built the box (even though the thread credits someone else).
I don't see how that post could be modded overrated. If I get modded troll and otherwise ignored...
...Here, here is our best hope for Microsoft. They are combining the strengths of Windows CE from their impressive mobile line, Windows ME, and of course, the reliability of the Windows NT kernel...
We'll keep plugging away on application compatibility
Isn't it touching how loyal the Vista Dev Team is to Duke Nukem Forever? Real Soon Now, Vista will be ready for release! I mean, they even have a Release Candidate out!
Quick, name one player that uses ICT? Anyone? That's because the studios agreed to not use it
You misspelled "disk." The player is required to support ICT as a part of the AACS spec. If the studios want to release a few unimportant disks with ICT turned off to sucker us in, they might find we will not buy in.
but when the source is 24 fps
Cable and broadcast HDTV already support 60fps. I would think you would understand, since you own a 1080p HDTV. If you already own a player, fine. I'm going to wait for one that can decode 1080p60. Your computer can already do that.
Yes, currently they do not. But the PC does, so it's only a matter of time before the players will too. Do you want to replace your $3000 HDTV at that point?
Not in HD, where PC playback is being held back by the DRM morass.
Sure, but there is content that is not DRM'd. To shoot 1080p60 requires an expensive camera, but I was thinking of CGI, where impossibly fast action and detail are commonplace.
Still, your point is valid. The technology isn't there yet. For me, that means I will wait until it's ready, because I think 1080p60 will be worth it.
Dont forget there was also the early promise that the media would eventually be cheaper (even though it never happened)
That will happen in just a few years when these HD DVD players start gaining popularity. Then hopefully you will be able to get new release DVD's for $5. Keep your fingers crossed, and don't tell the media execs you're waiting for it, or it will never happen...
Due to the Image Constraint Token (ICT) your new HD player will not play your movies at full resolution. (Because the HDTVs sold to date mostly have component inputs, and no HDMI input)
There are only a few titles in each format, and no guarantee that all titles will become available in one format (until the format war is over)
The AACS DRM offers features like remotely blacklisting your player, which will immediately brick it when you play that new movie. Why would I buy this? No, I'm not a pirate; this feature is not a feature at all.
Demos at the store have been disappointing at best. The improvement over DVD is pretty slim.
I'd rather wait for the price to drop on 1080p players. I know that a 60Hz 1080i can play a 24fps 1080p movie. But what if I want to watch a 60fps 1080p movie? See point above about your computer being better than your DVD player for this.
This article points out that your computer will probably out-perform any DVD player you can buy
BD-R and HD-DVD-R are available but still pretty expensive. This might not seem like a factor at first, but remember that the big pirating outfits are not using recordable media. The early adopters will. Case in point: I work with an independent movie studio and they want to show their previews in HD when they travel. What they do right now is bring a nice powerful laptop with the movie on the hard disk. How is a player going to compete with that?
I think people on slashdot understand the technology and its usefulness.
Now instead of the book being checked out, it will never be checked out but all the laptops will have something carved in the back, coke spilled inside it, and then kicked around under the desk.
All at the expense of the school district. There are better ways to use that money, like upgrading the computer lab.
What? Red Alert wasn't the most important thing to learn at college?
Seriously, give the parents the money that would have been spent on the laptop. If they think their kid needs a laptop, they'll buy it and the kid will be labelled a geek.
You cite iTunes taking 35 cents on each $0.99 sale. That information seems outdated.
Here's one website that has more information. I think there is at least one other site to check, cdbaby.com, but I'm not going to do that right here. This is the payment plan for independent artists, so I imagine the "labels" could try and negotiate a better deal, but this deal is pretty sweet:
iTunes: $0.70 per song, for 10+ songs (AKA an "album") payment is only 10 x $0.70 = $7.00 Rhapsody: $0.65 per song (subscriber download), $0.70 per song (non subscriber download), for 10+ songs album payment is 10x Napster: $0.65 per song, for 10+ songs album payment is 10x MusicNet: there is some undisclosed "independent record label" pay rate. See the FAQ eMusic: the "subscription pay rate" model. They pay a percentage of their profits. Sony Connect: the "subscription pay rate" model. They pay a percentage of their profits.
So, the best deals offered by tunecore.com, at 70.707070% of the purchase price, are very, very good for the artist. Any artist only making around 30% is simply losing money on their next album.
Wow. I looked it up on the OED and didn't see the information you're talking about. I suppose I could keep searching... care to share a source? I'd like to learn the etymology of "bug."
Would breathing be innate and yet simple? There is a breathing reflex, and yet breathing must be learned, so it's not innate in the sence of automatic or instinctive. It's innate in the sense that it is learned by everyone.
Or how about focusing each eye...the algorithms to determine optimal focus (without infrared ranging) are simple. Now, aiming both eyes at the same point to produce a stereo image is more difficult.
Okay, I agree that BSD 4.3 was great. That much the article is clear about.
But what I want to know is which ideology will win out in the end. The GPLv3 just hilights the question. The BSD license and GPL have been around for a while now, and TiVo has got Richard Stallman on YAC (Y. A. Crusade). Some say DRM will be the end of the GPL, making it a shadow of the BSD license. Others say DRM will allow companies to steal BSD code without a backward glance.
Anybody know the future? I'm going to guess that the GPL will last the longest, because it is making the most noise. (The squeaky wheel and all that.)
He just copies in his version of the software. As long as he does this after the initial verification sweep, it won't be detected. The key that signs the software much remain secret, or it doesn't really help security all that much.
I realize ROM chips are slower than RAM, but since we're talking about hardware where security matters so much, they should execute directly from the ROM.
You are right about exploits though. For instance, if the USB driver has an exploit, he sticks in a special USB key, takes over the system, and executes new code from RAM. No amount of DRM will ever protect against a completely new exploit.
But the GPLv3 is not the problem. Whether the binary is signed or not doesn't matter if the hypervisor, kernel, and/or TPM chip can be bypassed. That's actually a great argument for releasing the keys and the source. Release the keys and the source so that I can compile the entire ROM image using a toolchain of my choice and I can verify it comes out bit-for-bit just like the one you gave me. Otherwise, how can I, the customer, trust that you haven't put in a back door.
You must also sell a non-drm version of the hardware or a version with a published secret
It sounds like a good idea. However, they are not the same thing to begin with, right? I guess I wouldn't trust the company not to use separate source trees to build the secure and open versions. If a company can make it happen as a viable business model, then that means it's a good idea, my reservations notwithstanding.
The problem with the GPLv3 is that it is expected to be widely used, and many people who wouldn't have a problem with the above uses of DRM would release their code under the latest GPL without really thinking about it.
Here's the way to release GPLv3 (current draft) hardware that is tamper-resistant. I'm ignoring the obvious attacks, like physically pulling the ROM chip out and sticking in another one. Sticking just to software attacks, and how the GPLv3 doesn't stop a hypothetical election official from protecting his voting machine.
You build the system with DRM to verify that all the software is signed properly. You distribute the key with the source code. So, anyone who wants to tamper with your machine can put in another ROM that is signed.
*BUT* The election official probes the ROM with a ROM reader, with the machine off. He gets the contents of the ROM before any election, and reads them into a different computer. This means the ROM must be properly isolated electrically when the power is off. Design that into the board.
Now, he computes the MD5 sum of the ROM on his other computer, and checks that by hand with the MD5 sum printed on the contract he signed with the manufacturer.
The manufacturer's contract with the election official includes the MD5 sum of the software they both have agreed to use.
So why bother signing it at all in the first place? To prevent "accidental" inclusions of unsigned software by the manufacturer. The election official has studied and accepted *ONE* form of the software on the machine. He only has to find that one MD5 sum, and he's satisfied.
Note how it is the final owner (often, that's me, like when it's my computer, and not a voting machine) that must be responsible for making the "trustworthy" call, to trust or not to trust. I won't let some chip or some chip manufacturer make that decision for me.
> 1. locked digital media This is where the GPLv3 works
Sort of. If someone wants to include a DRM'd player on an otherwise free system (including GPLv3), there's no problem.
Agreed. It appears this will be the first major battlefield for TPM. The media center PC (Mac Mini, XBox) in the future will have a TPM chip to make DRM unbreakable. Microsoft is already moving in this direction. GPLv3 will prevent the TiVo scenario, keeping FOSS code out of locked TPM software stacks.
But will that stop TPM in the media center PC? I doubt it. The battle will be won by GPLv3 because of the security holes in any closed-source system. HD-DVD and BluRay will be cracked. Windows will have its security holes. And so there will always be a percentage of the market who need systems with GPLv3 software enough to hold the rest of the industry at bay.
> 2. locked FOSS-using devices (the Tivo scenario)
> I think the FSF are overstepping their bounds. They're telling hardware developers that they have to make THEIR OWN product convertable to any use their customers see fit.
The FSF is simply telling hardware developers that they don't have any special additional rights to control the software (not their software, remember) simply because they are working in hardware rather than software.
Everyone is potentially a developer.
Agreed. This is the second battle, when we are years into the media center PC, the TPM + DRM lockdown, and the DMCA will still be there. The businesses who have a decided to make their money by manipulating customers' choices (this is not advertising) will continue to try to push for a monopoly (100%) TPM. This means they will try to drive FOSS out of computing.
Microsoft is a good example of how this happens after the first lockdown. Microsoft first achieved OS domination and vendor lock-in. Then they started leveraging their marketshare to drive the competitors out of business. It will be the same way in this battle.
They have no interest in anyone being able to develop. But GPLv3 preserves the ability to develop any kind of new invention (software or hardware) using equipment you have purchased. It's all a part of fair use, and the GPL has been about distributing source code along with binaries so that fair use covers the source code. The other alternative is reverse engineering. I'd say GPL is just the more efficient method of the two, since both are possible, even with the most elaborate TPM schemes. It's an arms race, and a pointless one, I believe. Especially pointless now that the FOSS community is so strong and capable.
> 3. locked general-purpose computers
> The net result will simply be that hardware developers will stop considering the use of FOSS
There will always be hardware vendors who sell closed-source products. The FOSS projects have never been threatened by closed-source products. Would emacs disappear even if everone switched to vi? No, because old source code always ends up archived somewhere. Old binaries are sometimes archived too, but since they're old, they usually don't work. FOSS is not threatened by the success of other projects.
It is the closed-source shops that are threatened by open source. They will do whatever they can to stop open, community-based development, but it won't do any good in the end.
I think stage three of this battle will not be locked general-purpose computers. I do think that we will see them, but I think that will be part of stage one, above. And I think the GPLv3 (and even the GPLv2) will be around to keep stage one from complete penetration. FOSS will not die in stage one.
Stage three will take a long time to implement, but the ultimate goal is to reduce the internet back to cable TV, what we had before 1990. Efforts toward a tiered internet may not succeed in the near future. But these large companies are not going away, and they will be lobbying your governments very pe
This may not catch every woman in FOSS, but it provides them with a community. And if this is run by men, it is probably doomed. Then again, anonymity is something that women might value even more than men, for any reasons you can think of. And you're spot-on about FOSS progress coming from the obsessive streak in many men.
There are some intangible benefits to contributing to FOSS which might attract some women, like developing a better resume or making professional connections. However, I don't think women will contribute under much of the rationale that men do: scratch an itch, bragging rights, altruism, or even stick-it-to-the-M$.
These programs won't have a major effect on the percentage of women contributing to FOSS. (Is there even a good way of measuring that?) If men wanted to attract women to contribute, they would advertise. There are a lot of businesswomen in marketing. QED
I don't think that's quite true, a lot of the goals at national labs are very blue sky. However, there's one bigger problem: there's so much bureaucracy
I agree with you. The labs I have worked with have all had some very open-ended goals. The bureaucracy kept them from actually pursuing the stated goals, and the pressure also kept them from doing side-projects.
The current business climate rewards IT innovation more than Mechanical Engineering, but I do know a few successful ME's who found research positions--but it's an uphill battle everywhere.
Parent is right. See this story about "Labs of yester-years." I think the general consensus of the replies was that big corporate R&D is no longer blue-sky, and those who want to pursue such open-ended projects balance University research and small business.
Personally I can add a my two cents working in Defense Labs and National Labs: the political forces are too strong for blue-sky research to happen there. But if they happen to be already involved in what you like doing, then you will fit. I'm guessing you want to stick with what the parent post suggested. Good luck!
I see some pretty good ideas there. After Leopard is released, Vista will announce new features for Aero Glass that are either from Leopard or from these ideas. (Did you really think they would ship before WWDC?)
Slashdot Mirror
The article on Wikipedia is actually pretty good. There may be some transitory or faint effects caused by subliminal messages. Advertisers have been trying to capitalize on this possibility for 49 years.
If subliminal messages had any significant effect we would know about it. They've been trying for years.
There have been interesting claims at subliminal messages in popular music. KAKE-TV in Wichita, Kansas, used a subliminal message to attempt catch the BTK killer, but it had no perceptible effect.
I'd say subliminal messages don't work.
Okay, this is one of the most informative posts ever. People are thinking this is Williams, the original guy who built the box (even though the thread credits someone else).
I don't see how that post could be modded overrated. If I get modded troll and otherwise ignored...
...Here, here is our best hope for Microsoft. They are combining the strengths of Windows CE from their impressive mobile line, Windows ME, and of course, the reliability of the Windows NT kernel...
GIF Product Brochure
From the summary:
We'll keep plugging away on application compatibility
Isn't it touching how loyal the Vista Dev Team is to Duke Nukem Forever? Real Soon Now, Vista will be ready for release! I mean, they even have a Release Candidate out!
Quick, name one player that uses ICT? Anyone? That's because the studios agreed to not use it
You misspelled "disk." The player is required to support ICT as a part of the AACS spec. If the studios want to release a few unimportant disks with ICT turned off to sucker us in, they might find we will not buy in.
but when the source is 24 fps
Cable and broadcast HDTV already support 60fps. I would think you would understand, since you own a 1080p HDTV. If you already own a player, fine. I'm going to wait for one that can decode 1080p60. Your computer can already do that.
Blu-ray and HD DVD do not allow 1080p60
Yes, currently they do not. But the PC does, so it's only a matter of time before the players will too. Do you want to replace your $3000 HDTV at that point?
Not in HD, where PC playback is being held back by the DRM morass.
Sure, but there is content that is not DRM'd. To shoot 1080p60 requires an expensive camera, but I was thinking of CGI, where impossibly fast action and detail are commonplace.
Still, your point is valid. The technology isn't there yet. For me, that means I will wait until it's ready, because I think 1080p60 will be worth it.
Hey, don't tell anyone, but...
Dont forget there was also the early promise that the media would eventually be cheaper (even though it never happened)
That will happen in just a few years when these HD DVD players start gaining popularity. Then hopefully you will be able to get new release DVD's for $5. Keep your fingers crossed, and don't tell the media execs you're waiting for it, or it will never happen...
But every kid will have a laptop! Think of the children! (Without any administrator, think of what that will mushroom into...)
I agree with you--the money would be better spent building the labs and network that everyone uses. Or maybe paying the teachers...
I think people on slashdot understand the technology and its usefulness.
Now instead of the book being checked out, it will never be checked out but all the laptops will have something carved in the back, coke spilled inside it, and then kicked around under the desk.
All at the expense of the school district. There are better ways to use that money, like upgrading the computer lab.
What? Red Alert wasn't the most important thing to learn at college?
Seriously, give the parents the money that would have been spent on the laptop. If they think their kid needs a laptop, they'll buy it and the kid will be labelled a geek.
Okay, in a forum about spam, you post the email address of a guy who doesn't know much about computers.
Was that so he would get flooded with spam and give up the account? Or is that "email addresses have been changed to protect the innocent"?
You're right about the normal users though. Hard to deal with.
You cite iTunes taking 35 cents on each $0.99 sale. That information seems outdated.
Here's one website that has more information. I think there is at least one other site to check, cdbaby.com, but I'm not going to do that right here. This is the payment plan for independent artists, so I imagine the "labels" could try and negotiate a better deal, but this deal is pretty sweet:
iTunes: $0.70 per song, for 10+ songs (AKA an "album") payment is only 10 x $0.70 = $7.00
Rhapsody: $0.65 per song (subscriber download), $0.70 per song (non subscriber download), for 10+ songs album payment is 10x
Napster: $0.65 per song, for 10+ songs album payment is 10x
MusicNet: there is some undisclosed "independent record label" pay rate. See the FAQ
eMusic: the "subscription pay rate" model. They pay a percentage of their profits.
Sony Connect: the "subscription pay rate" model. They pay a percentage of their profits.
So, the best deals offered by tunecore.com, at 70.707070% of the purchase price, are very, very good for the artist. Any artist only making around 30% is simply losing money on their next album.
it goes all the way back to Shakespeare
Wow. I looked it up on the OED and didn't see the information you're talking about. I suppose I could keep searching... care to share a source? I'd like to learn the etymology of "bug."
Thanks!
Would breathing be innate and yet simple? There is a breathing reflex, and yet breathing must be learned, so it's not innate in the sence of automatic or instinctive. It's innate in the sense that it is learned by everyone.
Or how about focusing each eye...the algorithms to determine optimal focus (without infrared ranging) are simple. Now, aiming both eyes at the same point to produce a stereo image is more difficult.
Okay, I agree that BSD 4.3 was great. That much the article is clear about.
But what I want to know is which ideology will win out in the end. The GPLv3 just hilights the question. The BSD license and GPL have been around for a while now, and TiVo has got Richard Stallman on YAC (Y. A. Crusade). Some say DRM will be the end of the GPL, making it a shadow of the BSD license. Others say DRM will allow companies to steal BSD code without a backward glance.
Anybody know the future? I'm going to guess that the GPL will last the longest, because it is making the most noise. (The squeaky wheel and all that.)
I feel old.
Can't say whether you're old or not, but I'm 27 and cut my teeth on the Atari 800. My favorite was the "READY" prompt. It felt so powerful.
Of course, I didn't understand all that hex in the DATA statements. I just typed them in.
Later in life I actually met one of the engineers who worked at Atari. That company had an interesting life!
He just copies in his version of the software. As long as he does this after the initial verification sweep, it won't be detected. The key that signs the software much remain secret, or it doesn't really help security all that much.
I realize ROM chips are slower than RAM, but since we're talking about hardware where security matters so much, they should execute directly from the ROM.
You are right about exploits though. For instance, if the USB driver has an exploit, he sticks in a special USB key, takes over the system, and executes new code from RAM. No amount of DRM will ever protect against a completely new exploit.
But the GPLv3 is not the problem. Whether the binary is signed or not doesn't matter if the hypervisor, kernel, and/or TPM chip can be bypassed. That's actually a great argument for releasing the keys and the source. Release the keys and the source so that I can compile the entire ROM image using a toolchain of my choice and I can verify it comes out bit-for-bit just like the one you gave me. Otherwise, how can I, the customer, trust that you haven't put in a back door.
You must also sell a non-drm version of the hardware or a version with a published secret
It sounds like a good idea. However, they are not the same thing to begin with, right? I guess I wouldn't trust the company not to use separate source trees to build the secure and open versions. If a company can make it happen as a viable business model, then that means it's a good idea, my reservations notwithstanding.
The problem with the GPLv3 is that it is expected to be widely used, and many people who wouldn't have a problem with the above uses of DRM would release their code under the latest GPL without really thinking about it.
Here's the way to release GPLv3 (current draft) hardware that is tamper-resistant. I'm ignoring the obvious attacks, like physically pulling the ROM chip out and sticking in another one. Sticking just to software attacks, and how the GPLv3 doesn't stop a hypothetical election official from protecting his voting machine.
You build the system with DRM to verify that all the software is signed properly. You distribute the key with the source code. So, anyone who wants to tamper with your machine can put in another ROM that is signed.
*BUT* The election official probes the ROM with a ROM reader, with the machine off. He gets the contents of the ROM before any election, and reads them into a different computer. This means the ROM must be properly isolated electrically when the power is off. Design that into the board.
Now, he computes the MD5 sum of the ROM on his other computer, and checks that by hand with the MD5 sum printed on the contract he signed with the manufacturer.
The manufacturer's contract with the election official includes the MD5 sum of the software they both have agreed to use.
So why bother signing it at all in the first place? To prevent "accidental" inclusions of unsigned software by the manufacturer. The election official has studied and accepted *ONE* form of the software on the machine. He only has to find that one MD5 sum, and he's satisfied.
Note how it is the final owner (often, that's me, like when it's my computer, and not a voting machine) that must be responsible for making the "trustworthy" call, to trust or not to trust. I won't let some chip or some chip manufacturer make that decision for me.
Just my 2 cents.
Look, I know I haven't taken the time to research this properly, but I thought you'd like to read this:
... (skip ahead, I'm in a hurry) ...
Xerox Parc: The GUI, +1 Brilliant
Apple II: The Usable GUI on a home computer, +1 Informative
Apple II: Hierarchical File System, +1 Interesting
Apple II: 3.5" Floppy, +1 My Favorite
MS-DOS: Directories, -1 Redundant
Macintosh: QuickTime, +1 Interesting
Macintosh: 44khz 16-bit sound, +1 Funny
Microsoft: Windows, -1 Offtopic
Microsoft: MPC standard (attempt at multimedia), -1 Overrated
Macintosh: SCSI, +1 Fast
Macintosh: 68030 multitasking, +1 Useful
Microsoft: Windows 3.1, -1 Redundant
Macintosh: Apple Menu, +1 Informative
Microsoft: Windows 95 Start Menu, -1 Redundant
Microsoft: Windows 95 Recycle Bin, -1 Offtopic
Macintosh: PowerPC, changing processor architectures, +1 Gutsy
Microsoft: Windows NT Alpha, -1 Unsupported
Macintosh: OS X, +1 Drool
Microsoft: Windows 2000, -1 Bugfix
Macintosh: BSD utilities included, and the OpenDarwin project, +1 Insightful
Microsoft: TCP/IP stack, -1 Stolen
Macintosh: Spotlight, +1 Useful
Microsoft: Windows Vista, -1 Nothing To See Here, Move Along
Okay, and the preliminary scores are:
Xerox Parc: +1
Apple: +12
Microsoft: -10
And for the record, I don't own a Mac. (*shakes wallet, hears two nickels rub together*)
Does somebody want to reply to this with a more comprehensive and accurate list? I've gotta go watch "The Pirates of Silicon Valley."
> 1. locked digital media This is where the GPLv3 works
Sort of. If someone wants to include a DRM'd player on an otherwise free system (including GPLv3), there's no problem.
Agreed. It appears this will be the first major battlefield for TPM. The media center PC (Mac Mini, XBox) in the future will have a TPM chip to make DRM unbreakable. Microsoft is already moving in this direction. GPLv3 will prevent the TiVo scenario, keeping FOSS code out of locked TPM software stacks.
But will that stop TPM in the media center PC? I doubt it. The battle will be won by GPLv3 because of the security holes in any closed-source system. HD-DVD and BluRay will be cracked. Windows will have its security holes. And so there will always be a percentage of the market who need systems with GPLv3 software enough to hold the rest of the industry at bay.
> 2. locked FOSS-using devices (the Tivo scenario)
> I think the FSF are overstepping their bounds. They're telling hardware developers that they have to make THEIR OWN product convertable to any use their customers see fit.
The FSF is simply telling hardware developers that they don't have any special additional rights to control the software (not their software, remember) simply because they are working in hardware rather than software.
Everyone is potentially a developer.
Agreed. This is the second battle, when we are years into the media center PC, the TPM + DRM lockdown, and the DMCA will still be there. The businesses who have a decided to make their money by manipulating customers' choices (this is not advertising) will continue to try to push for a monopoly (100%) TPM. This means they will try to drive FOSS out of computing.
Microsoft is a good example of how this happens after the first lockdown. Microsoft first achieved OS domination and vendor lock-in. Then they started leveraging their marketshare to drive the competitors out of business. It will be the same way in this battle.
They have no interest in anyone being able to develop. But GPLv3 preserves the ability to develop any kind of new invention (software or hardware) using equipment you have purchased. It's all a part of fair use, and the GPL has been about distributing source code along with binaries so that fair use covers the source code. The other alternative is reverse engineering. I'd say GPL is just the more efficient method of the two, since both are possible, even with the most elaborate TPM schemes. It's an arms race, and a pointless one, I believe. Especially pointless now that the FOSS community is so strong and capable.
> 3. locked general-purpose computers
> The net result will simply be that hardware developers will stop considering the use of FOSS
There will always be hardware vendors who sell closed-source products. The FOSS projects have never been threatened by closed-source products. Would emacs disappear even if everone switched to vi? No, because old source code always ends up archived somewhere. Old binaries are sometimes archived too, but since they're old, they usually don't work. FOSS is not threatened by the success of other projects.
It is the closed-source shops that are threatened by open source. They will do whatever they can to stop open, community-based development, but it won't do any good in the end.
I think stage three of this battle will not be locked general-purpose computers. I do think that we will see them, but I think that will be part of stage one, above. And I think the GPLv3 (and even the GPLv2) will be around to keep stage one from complete penetration. FOSS will not die in stage one.
Stage three will take a long time to implement, but the ultimate goal is to reduce the internet back to cable TV, what we had before 1990. Efforts toward a tiered internet may not succeed in the near future. But these large companies are not going away, and they will be lobbying your governments very pe
This may not catch every woman in FOSS, but it provides them with a community. And if this is run by men, it is probably doomed. Then again, anonymity is something that women might value even more than men, for any reasons you can think of. And you're spot-on about FOSS progress coming from the obsessive streak in many men.
There are some intangible benefits to contributing to FOSS which might attract some women, like developing a better resume or making professional connections. However, I don't think women will contribute under much of the rationale that men do: scratch an itch, bragging rights, altruism, or even stick-it-to-the-M$.
These programs won't have a major effect on the percentage of women contributing to FOSS. (Is there even a good way of measuring that?) If men wanted to attract women to contribute, they would advertise. There are a lot of businesswomen in marketing. QED
I agree with you. The labs I have worked with have all had some very open-ended goals. The bureaucracy kept them from actually pursuing the stated goals, and the pressure also kept them from doing side-projects.
The current business climate rewards IT innovation more than Mechanical Engineering, but I do know a few successful ME's who found research positions--but it's an uphill battle everywhere.
Personally I can add a my two cents working in Defense Labs and National Labs: the political forces are too strong for blue-sky research to happen there. But if they happen to be already involved in what you like doing, then you will fit. I'm guessing you want to stick with what the parent post suggested. Good luck!
I see some pretty good ideas there. After Leopard is released, Vista will announce new features for Aero Glass that are either from Leopard or from these ideas. (Did you really think they would ship before WWDC?)