More seriously, I wonder if subways currently store some of that kinetic energy by putting the passenger platforms at a slightly higher elevation (not as deep in the ground) in comparison to the other portions of the track. If I have my math right, the kinetic energy of moving at 30 meters per second (
~67 miles/hour) is approximately the potential energy of an elevation of 45 meters in 1 Earth gravity (0.5mv^2 = mgh --> 0.5v^2 = gh --> h=0.5v^2/g --> h = 0.5(30m/sec)^2/(10m/sec^2) = 45 m/sec). I imagine that that would be much too rollercoastery for a local train, and you wouldn't want to have the train fly off the track so easily for arriving a little too fast, but it wouldn't surprise me if a dip of a meter or two is engineered into subway lines for a bit of energy savings.
They need to weight individual MPs votes by their local approval rating.
Interesting, although the devil is in the details of how approval ratings would be measured.
You might want to consider what other improvements are politically and technically practical.
Australia already has a relatively sophisticated vote counting system (instant runoff voting, which is not my favorite, but still pretty good in my opinion). So, perhaps the most universally understandable improvement in responsiveness would come from having more frequent elections. Lowering the stakes in elections and reducing the time that a losing politician needs to wait to run for reelection would encourage them to be slightly more honest in stating their opinions and would give them more accurate feedback about what the public really thinks.
Newsgroup feeds use up about 1.5TB a day. Do the math for 14 days retention. It's a heck of a committment.[sic]
1TB 7200RPM hard disks are currently priced at US$169 on pricewatch.com, with free ground shipping.
1.5TB x $169/TB = $253.50/day of retention.
x 14 days = $3549 per fortnight of retention.
In comparison, Verizon's sales for 2007 were $94.72 billion, that is, over 26 million times as much. Even if you multiply usenet retention costs by a factor of 10 or 100 to account for multiple copies, other hardware, etc., it is nowhere near even the roundoff accuracy of the financial statements in an shareholders' annual report of a company of that size, not really "a heck of a commitment."
I believe that the C3 processor made by VIA and probably other processors in that family allow some of the cache to be configured as SRAM and mapped into physical memory. So, you could store the key in SRAM, which I believe really will lose its data upon power loss, although you may also want to take countermeasures such as those used in loop-AES to avoid detection by physical changes to the chip if key is store for a long time in the same place.
Newever VIA processors also have some hardware AES support available under Linux, which they call "Padlock. So, if they still retain the SRAM feature, then, at that would make a pretty good choice for the little fanless mini-ITX Linux box that receives your email.
While identifying and using an area overwritten by your BIOS is good idea, I think it would not be that hard to work around.
Modern computers store their BIOS in "NOR" flash, a type of flash that can be read and written just like memory, only much more slowly (but costs more per bit than the "NAND" flash used in USB sticks). When the computer boots, it actually starts executing straight from the flash. It should be possible to load a custom BIOS that does not write to RAM at all, only storing its data in the CPU registers and in flash. You would have to be very careful, but it could be done with commodity hardware without physical modification, and in such a way that you could provide a mechanism to restore the flash back to its normal operation later.
Also, there is a more trivial solution for the type of computer that has more than one memory socket and where the memory is not interleaved, as is typical for desktop computers but not for notebooks. Just put the memory to be analyzed in the socket that is not overwritten by BIOS.
I suspect that a better solution here would be to store the key in SRAM. I believe that Via C3 chips and probably their relatives have a configuration option where some of the cache RAM can be used instead regular SRAM, mapped into a particular memory page.
I haven't run Compiere or ADempiere (their capitalization), but my understanding is that ADempiere is a major fork of Compiere that supports PostgreSQL and touts itself
as more "community" oriented. Links:
sourceforge,
dot com,
wikipedia,
another wiki.
Google claims to have found a 1.36 million hits for "ADepmiere", which seem incredible to me, as in "not credible." However, a slightly more credible indicator of commercial support in my opinion is that I see different google advertisements keyed on this word.
t says the *license* under which you distribute it must make it available to third parties. The GPL does not require you to *distribute* the source code to anyone except those who receive the product in executable form.
Lehk228 quoted the wrong part of the GPL, but he was correct. Section 3b requires the offer of source code to extend to third parites unless you included source code in the first place (3a) or shipped it noncommercially and included a pointer to source code (3c):
3. You may copy and distribute the Program (or a work based on it,
under Section 2) in object code or executable form under the terms of
Sections 1 and 2 above provided that you also do one of the following:
a) Accompany it with the complete corresponding machine-readable
source code, which must be distributed under the terms of Sections
1 and 2 above on a medium customarily used for software interchange; or,
b) Accompany it with a written offer, valid for at least three
years, to give any third party, for a charge no more than your
cost of physically performing source distribution, a complete
machine-readable copy of the corresponding source code, to be
distributed under the terms of Sections 1 and 2 above on a medium
customarily used for software interchange; or,
c) Accompany it with the information you received as to the offer
to distribute corresponding source code. (This alternative is
allowed only for noncommercial distribution and only if you
received the program in object code or executable form with such
an offer, in accord with Subsection b above.)
I am not a lawyer. Please do not use this as legal advice.
Yes, as I stated in my reply (to which you replied), I understand that the ehci would be needed for USB-2.0, and either ohci or uhci would be needed for USB-1.x.
You're right. I didn't read your post closely enough before delivering an unnecessary lecuture about USB2, which I hope helped enlighten some other reader, because otherwise I just wasted a lot of my time. I apologize for wasting yours.
I see from the original poster's follow-up that it looks like he has a dedicated ohci usb-1.x controller in addition to the uhci controller(s) that his ehci uses. So it looks like he should have all three drivers loaded, so that is unlikely to be the source of his problems.
I believe that ehci silicon (and therefore the ehci driver) only handles communication with 480mbps USB 2.0 devices. When you plug a USB 1.x device into a USB 2.0 "root hub" port (i.e., a USB port in your computer), the ehci root hub notices the lack of the distinctive USB2 "chirp" signal, disconnects itself, and connects a built-in USB 1.x controller to the line. The USB 1.x controller is a completely separate device, with its own PCI information, its own USB device numbering, etc. If you do an "lspci" on a USB2 computer, you will see lines for several USB controller devices. Typically, there will be one USB2 controller and a bunch of USB1.x controllers, so that not all of the USB1 devices need to share just one USB 1.x bus.
If you do a "mount -t usbdevfs foo/proc/bus/usb" and then do an ls on/proc/bus/usb, you will also see the separate USB controllers.
I belive that is why you will have device drivers for both ehci and uhci loaded. I believe that the uhci driver will talk to USB 1.x device that you plug directly into your computer or through a USB-1.x-only external hub. The ehci driver will talk to any USB 2.x device that you plug in directly or any devices (including USB 1.x devices) that you plug in through a USB2 hub.
USB2 external hubs, by the way, must include USB2->USB1 translators on every port, so a USB2 external hub is enormously more complex than a USB1 external hub. I suppose it would be possible for someone to built a USB2 root hub controller using the same scheme, but I do not know if anyone does.
It's also worth noting that the distinction between 4mbps and 12mbps USB 1.x devices
is handled in a different, simpler, way, but one which would have been a big bandwidth drain on 480mbps devices.
I believe that as the elevator car would climb the cable, it would have the effect of tugging the cable west slightly, giving the cable a slight westward kink where the car is.
Centripetal acceleration would work to straighten out the cable again, pulling the car east and rest of the cable west slightly. The bottom of the cable would be pushed toward the west slightly, which would be resisted by the attachment of the base to the ground, the net result being that the base would be providing an eastward acceleration to the cable, infinitesimally slowing the rotation of Earth.
The top of the cable would drift east, but centripetal acceleration would pull it straight too, again resulting in a westward tug on bottom of the cable.
Of course, there is no way for the cable to pull westward if it is going perfectly straight up, so
I would expect the cable to gradually lean westward until its pull from the base had enough of a westward vector to compensate. I think the angle would be very slight. Here is a rough calculation.
Let's say the elevator car will climb 22,300 miles to geosync orbit at 22.3 miles/hour, which will take 1,000 hours (more than a month). During this time, the car's horizontal velocity will increase to 8,000 meters/second. So, every hour, the car needs to increase its eastward velocity by 8 meters/second, just to remain "stationary" from the standpoint of an observer on the Earth's (rotating) surface. 8 meters/sec/hour of acceleration is.002 meter/sec/sec acceleration. Let's say the elevator car weighs 1000 kg. The eastward force needed would be 1000kg *.002 meters/sec^2 = 2 kg/m/sec. (2 Newtons), about the force that gravity at the earth's surface exerts on a 200 gram weight.
Let's say the cable is angled at a 1:100 grade (that is, about 1 degree from vertical), so that going 100 meters up the cable also takes you 1 meter west. Pulling the cable from the bottom with a force of 200.001 Newtons would give it a downward force of 200 Newtons (which would be counter balanced by the counterweight of the space tether being slightly farther out) and the eastward force of the required 2 Newtons. 200.001 Newtons is approximately the force of gravity on a 20 kilogram weight.
So, even with just a 1 degree westward slant, I think the force needed to provide the necessary horizontal acceleration would be very small.
However, please evaluate this estimate with caution. I'm just doing it on the fly and I could have made a substantial error somewhere.
Yes, simplicity is good, but only in the context of the whole system. Here, [GoboLinux is] just shifting complexity from the per-package scripts to the overall Compile package itself -- creating a large, central, monolithic service.
But, taking a page from the BSD ports collection, you could have modules distributed separately from the "monolithic service" that define common packaging conventions. There could be a description of how to build "standard" autoconf-generated packages, which might be referenced by the more specialized description of how to build gnome packages, and so on.
This way, you would have the advantage of not having so many duplicated descriptions to maintain and you would also avoid having a monolithic kludge.
But that doesn't show that patent incentive caused or even contributed to the invention of the transistor.
Evidence of patent incentive contributing to the invention of the transistor might be memos from Bell Labs executives saying that they would kill the project were it not for the potential of patent royalties. (I'm not saying that that is the only form of evidence that you could find. I just want to provide an example.)
Although I have not tracked the claims down, I've heard it said that Bell Labs's funding was largely the result of the phone company's profit being legally limited. So, if the phone company thought they were going to make too much money in a given quarter, they'd dump it into Bell Labs, which worked on things that were of potential benefit to the phone company, rather than things that were useless to the phone company but could provide substantial patent royalty revenue.
Even if you show that the patent incentive caused or substantially contributed to the invention of the transistor, that still does not show that patents were worthwhile in this case. To do that, you have to estimate when someone else would have invented the transistor in the absence of patent incentives or with reduced patent incentives and show that the net benefit to society was probably greater with patent incentives of that magnitude.
To go from the question of whether patents were worthwhile in this positive example to the question of what the patent law should be, you have then measure the benefits in the positive cases (if you do show that they are examples of net benefit) against the costs of the negative cases (things that are patented that would have been invented anyway, patenting of non-inventions, litigation costs, etc.).
Obviously, this is a big empirical question. When I try to do the "math", I generally find that there are almost no examples of inventions where the patent incentive appears to have brought the invention about substantially faster or better than I think would have occurred without patents, and I see lots of examples of costly negative effects of patents.
Given that patnets are a deprivation of liberty, I think the benefit of the doubt should be made in favor of having less patent restriction. Adding that factor in, my belief at this time is that patents don't seem to be worth it for the public, in practice.
The dark side of Dothan also reared its head, briefly. When asked why Intel was introducing a new naming scheme for Centrino, Chandrasekhar replied that the numbers represented more of
"goodness measure" and reflected features that were not necessarily "performance enhancing", such as Le Grande. Le Grande is Intel's contribution to TCPA-compliant lock-down computing,and allows large media companies to impair the user's ability to exchange media files, such as their favorite songs. So you can see why Le Grande isn't "performance enhancing", and quite the reverse. [...]
Slowing TCPA adoption is enough of an benefit to me to prefer a TCPA-free processor even if it costs $50 more for the same performance. I just hope I'll have that option for a while, as Intel is not the only company promoting TCPA.
This is a free software license but it is incompatible with the GPL.
The IBM Public License is incompatible with the GPL because it has
various specific requirements that are not in the GPL.
For example, it requires certain patent licenses be given that the
GPL does not require. (We don't think those patent license
requirements are inherently a bad idea, but nonetheless they are
incompatible with the GNU GPL.)
By the way, that same FSF page says the same about IBM's other open source license,
the
Common Public License, used in
Eclipse.
I hope that a future version of the GPL will allow LGPL-style linking with code covered by GPL-incompatible "free software" licenses in cases where no proprietary software is linked in or perhaps that a future version of the GPL will address software patent problems in a way that is compatible with the IBM Public License and the Apache 2 License.
In the meantime, to me, the advantages of Postfix are less than the disadvantages of the obstacles to future
code recycling.
If the incompatabilities could be fixed by a change to either the GPL or the IPL, and if enough of the hype about postfix proved to be real, I would probably switch to postfix.
I know I am a statistically insignificant sample, but there are others, mostly developers, who care about being able to copy code between free software projects in general and GPL compatability in particular. So, Postfix gain that much more usage and support if and when this incompatability is resolved, such as, for the short term, by adding some exceptions for the GPL, as the CUPS project did.
It WHOOOOOSHes for about 3 seconds when you power it on, as the hardware is initialized. Then it goes totally and completely
silent as linux boots, and stays silent during heavy use.
"Silent" would mean no noise. A PC can be silent when the hard disk is spun down and all of its fans are stopped and nothing else in it is making any noise. You probably just meant "quiet."
I remember being in a terminal room at usenix many years ago which was full of Network Computing Devices X terminals, which were silent (diskless, fanless, no other parts moving or otherwise making sound). It was like being in a library. By far, the loudest remaining sound was the keystrokes from everyone typing.
I like silent (as opposed to quiet), PC's, although I've only
used them in practice as thin clients.
When I see a vendor using the term "silent" to refer to a PC that is merely quiet, I consider that vendor to be a liar, and I generally can't trust them enough to be willing to do buy from them.
I also buy a lot of quiet hardware, like big slow fans, aluminum cases so I can disconnect more fans and so on, but I try never to buy from vendors that call these things "silent."
If I recall correctly, the "ordinary skill in the art" criterion of United States Code, Title 35 (Patents), Part II, Chapter 10,
Section 103
was enacted by congress in the Patent Act of 1952 to reduce the higher common law standards of patentability, such as in Hotckiss v. Greenwood (1851).
Personally, I doubt that any standard of patentability is high enough, duration short enough and scope narrow enough to make a patent system worth more than the value of the competition and incremental development that it stifles. Perhaps my views are skewed by being based so much on my seeing software development "before" and "after" software patents, but I think
that with the advent of computer aided design and faster communications, the development of almost everything that is currently patentable is becoming more like software development or actually becoming software development.
Do you also believe your local library should stock Playboy on the shelves with Popular Science?
Assuming that they're "on the same shelves" because it's an alphabetical filing system or some similar reason, I would have no problem with that.
If you want to argue for restricting people's freedoms or access to information, including children's, then the burden of proof is upon you to justify it. You didn't include any reasons that I could identify in your posting.
It appears that the NSA have licensed this math in such a manner that they are free to sublicense it however they see fit.
This agreement will give the NSA a nonexclusive, worldwide license with the right to grant sublicenses of MQV-based ECC covered by many of Certicom's US patents and applications and corresponding foreign rights in a
limited field of use.
The "field of use" is not specified in any of the links provided by the slashdot article (and is probably confidential), nor are the parameters of the sublicensing, such as how much, if any, royalties NSA has to pass upstream. It is also worth noting that the users within "the limited field of use" are further slighty restricted, although I don't know enough math to understand how important the restriction is:
The field of use is restricted to implementations of ECC that are over GF(p), where p is a prime greater than 2**256. [exponent notation changed to get through slashdot filters]
If there is ever a court case where this becomes an important point, I would expect that such 'copying' that is required for normal usage of the material falls
squarely under fair use.
"The law also supports the conclusion that Peak's loading of copyrighted software into RAM creates a "copy" of that software in violation of the Copyright Act." MAI Systems Corp. v. Peak Computer, US 9th Circuit federal appeals court.
By the way, I believe that the Digital Millenium Copyright Act slightly narrowed this ruling by putting a special case provision into title 17 United State Code section 117 that only applied to software that was necessary for booting when hardware maintenance was being done by a third party provider.
I am not a lawyer. Please do not use this as legal advice.
Although we discussed in another response that this could be covered by the transfer payments that I originally suggested, I want to point out that that I don't know if your claim is true anyhow, at least when you average in the additional unemployed people earning $0/hour due to minimum wage. I would be interested in hearing a basis for this prediction and seeing what that basis predicts if we were to raise the minimum wage to, say, $500/hour.
I don't think theres a problem finding jobs at the bottom right now, McDonalds seems to be perpetually hiring.
If that were true, then we should rescale welfare to assume that everyone at least works at McDonalds. I think that it's more likely that McDonalds franchises, like many businesses, find it economically worthwhile to give that impression so that many people will interview, and they can keep their turnover costs low. It probably is true that someone educated enough to read slashdot who wants to work at McDonald's will get hired immediately, but I don't think that's true for everyone who you describe as being "at the bottom."
Slashdot Mirror
More seriously, I wonder if subways currently store some of that kinetic energy by putting the passenger platforms at a slightly higher elevation (not as deep in the ground) in comparison to the other portions of the track. If I have my math right, the kinetic energy of moving at 30 meters per second ( ~67 miles/hour) is approximately the potential energy of an elevation of 45 meters in 1 Earth gravity (0.5mv^2 = mgh --> 0.5v^2 = gh --> h=0.5v^2/g --> h = 0.5(30m/sec)^2/(10m/sec^2) = 45 m/sec). I imagine that that would be much too rollercoastery for a local train, and you wouldn't want to have the train fly off the track so easily for arriving a little too fast, but it wouldn't surprise me if a dip of a meter or two is engineered into subway lines for a bit of energy savings.
They need to weight individual MPs votes by their local approval rating.
Interesting, although the devil is in the details of how approval ratings would be measured.
You might want to consider what other improvements are politically and technically practical. Australia already has a relatively sophisticated vote counting system (instant runoff voting, which is not my favorite, but still pretty good in my opinion). So, perhaps the most universally understandable improvement in responsiveness would come from having more frequent elections. Lowering the stakes in elections and reducing the time that a losing politician needs to wait to run for reelection would encourage them to be slightly more honest in stating their opinions and would give them more accurate feedback about what the public really thinks.
Newsgroup feeds use up about 1.5TB a day. Do the math for 14 days retention. It's a heck of a committment.[sic]
1TB 7200RPM hard disks are currently priced at US$169 on pricewatch.com, with free ground shipping.
1.5TB x $169/TB = $253.50/day of retention.
x 14 days = $3549 per fortnight of retention.
In comparison, Verizon's sales for 2007 were $94.72 billion, that is, over 26 million times as much. Even if you multiply usenet retention costs by a factor of 10 or 100 to account for multiple copies, other hardware, etc., it is nowhere near even the roundoff accuracy of the financial statements in an shareholders' annual report of a company of that size, not really "a heck of a commitment."
If the estimated data point is not between known data points, then this process is extrapolation, not interpolation.
I believe that the C3 processor made by VIA and probably other processors in that family allow some of the cache to be configured as SRAM and mapped into physical memory. So, you could store the key in SRAM, which I believe really will lose its data upon power loss, although you may also want to take countermeasures such as those used in loop-AES to avoid detection by physical changes to the chip if key is store for a long time in the same place.
Newever VIA processors also have some hardware AES support available under Linux, which they call "Padlock. So, if they still retain the SRAM feature, then, at that would make a pretty good choice for the little fanless mini-ITX Linux box that receives your email.
While identifying and using an area overwritten by your BIOS is good idea, I think it would not be that hard to work around.
Modern computers store their BIOS in "NOR" flash, a type of flash that can be read and written just like memory, only much more slowly (but costs more per bit than the "NAND" flash used in USB sticks). When the computer boots, it actually starts executing straight from the flash. It should be possible to load a custom BIOS that does not write to RAM at all, only storing its data in the CPU registers and in flash. You would have to be very careful, but it could be done with commodity hardware without physical modification, and in such a way that you could provide a mechanism to restore the flash back to its normal operation later.
Also, there is a more trivial solution for the type of computer that has more than one memory socket and where the memory is not interleaved, as is typical for desktop computers but not for notebooks. Just put the memory to be analyzed in the socket that is not overwritten by BIOS.
I suspect that a better solution here would be to store the key in SRAM. I believe that Via C3 chips and probably their relatives have a configuration option where some of the cache RAM can be used instead regular SRAM, mapped into a particular memory page.
Wikipedia claims "Blu-ray discs may be encoded with a region code" and "There is no Region Coding in the existing HD DVD specification", although these web pages indicate that both formats usually carry digital restrictions management and there are only three blu-ray regions currently defined.
Google claims to have found a 1.36 million hits for "ADepmiere", which seem incredible to me, as in "not credible." However, a slightly more credible indicator of commercial support in my opinion is that I see different google advertisements keyed on this word.
I am not a lawyer. Please do not use this as legal advice.
You're right. I didn't read your post closely enough before delivering an unnecessary lecuture about USB2, which I hope helped enlighten some other reader, because otherwise I just wasted a lot of my time. I apologize for wasting yours.
I see from the original poster's follow-up that it looks like he has a dedicated ohci usb-1.x controller in addition to the uhci controller(s) that his ehci uses. So it looks like he should have all three drivers loaded, so that is unlikely to be the source of his problems.
If you do a "mount -t usbdevfs foo /proc/bus/usb" and then do an ls on /proc/bus/usb, you will also see the separate USB controllers.
I belive that is why you will have device drivers for both ehci and uhci loaded. I believe that the uhci driver will talk to USB 1.x device that you plug directly into your computer or through a USB-1.x-only external hub. The ehci driver will talk to any USB 2.x device that you plug in directly or any devices (including USB 1.x devices) that you plug in through a USB2 hub.
USB2 external hubs, by the way, must include USB2->USB1 translators on every port, so a USB2 external hub is enormously more complex than a USB1 external hub. I suppose it would be possible for someone to built a USB2 root hub controller using the same scheme, but I do not know if anyone does.
It's also worth noting that the distinction between 4mbps and 12mbps USB 1.x devices is handled in a different, simpler, way, but one which would have been a big bandwidth drain on 480mbps devices.
Centripetal acceleration would work to straighten out the cable again, pulling the car east and rest of the cable west slightly. The bottom of the cable would be pushed toward the west slightly, which would be resisted by the attachment of the base to the ground, the net result being that the base would be providing an eastward acceleration to the cable, infinitesimally slowing the rotation of Earth.
The top of the cable would drift east, but centripetal acceleration would pull it straight too, again resulting in a westward tug on bottom of the cable.
Of course, there is no way for the cable to pull westward if it is going perfectly straight up, so I would expect the cable to gradually lean westward until its pull from the base had enough of a westward vector to compensate. I think the angle would be very slight. Here is a rough calculation.
Let's say the elevator car will climb 22,300 miles to geosync orbit at 22.3 miles/hour, which will take 1,000 hours (more than a month). During this time, the car's horizontal velocity will increase to 8,000 meters/second. So, every hour, the car needs to increase its eastward velocity by 8 meters/second, just to remain "stationary" from the standpoint of an observer on the Earth's (rotating) surface. 8 meters/sec/hour of acceleration is .002 meter/sec/sec acceleration. Let's say the elevator car weighs 1000 kg. The eastward force needed would be 1000kg * .002 meters/sec^2 = 2 kg/m/sec. (2 Newtons), about the force that gravity at the earth's surface exerts on a 200 gram weight.
Let's say the cable is angled at a 1:100 grade (that is, about 1 degree from vertical), so that going 100 meters up the cable also takes you 1 meter west. Pulling the cable from the bottom with a force of 200.001 Newtons would give it a downward force of 200 Newtons (which would be counter balanced by the counterweight of the space tether being slightly farther out) and the eastward force of the required 2 Newtons. 200.001 Newtons is approximately the force of gravity on a 20 kilogram weight.
So, even with just a 1 degree westward slant, I think the force needed to provide the necessary horizontal acceleration would be very small.
However, please evaluate this estimate with caution. I'm just doing it on the fly and I could have made a substantial error somewhere.
But, taking a page from the BSD ports collection, you could have modules distributed separately from the "monolithic service" that define common packaging conventions. There could be a description of how to build "standard" autoconf-generated packages, which might be referenced by the more specialized description of how to build gnome packages, and so on.
This way, you would have the advantage of not having so many duplicated descriptions to maintain and you would also avoid having a monolithic kludge.
But that doesn't show that patent incentive caused or even contributed to the invention of the transistor.
Evidence of patent incentive contributing to the invention of the transistor might be memos from Bell Labs executives saying that they would kill the project were it not for the potential of patent royalties. (I'm not saying that that is the only form of evidence that you could find. I just want to provide an example.)
Although I have not tracked the claims down, I've heard it said that Bell Labs's funding was largely the result of the phone company's profit being legally limited. So, if the phone company thought they were going to make too much money in a given quarter, they'd dump it into Bell Labs, which worked on things that were of potential benefit to the phone company, rather than things that were useless to the phone company but could provide substantial patent royalty revenue.
Even if you show that the patent incentive caused or substantially contributed to the invention of the transistor, that still does not show that patents were worthwhile in this case. To do that, you have to estimate when someone else would have invented the transistor in the absence of patent incentives or with reduced patent incentives and show that the net benefit to society was probably greater with patent incentives of that magnitude.
To go from the question of whether patents were worthwhile in this positive example to the question of what the patent law should be, you have then measure the benefits in the positive cases (if you do show that they are examples of net benefit) against the costs of the negative cases (things that are patented that would have been invented anyway, patenting of non-inventions, litigation costs, etc.).
Obviously, this is a big empirical question. When I try to do the "math", I generally find that there are almost no examples of inventions where the patent incentive appears to have brought the invention about substantially faster or better than I think would have occurred without patents, and I see lots of examples of costly negative effects of patents.
Given that patnets are a deprivation of liberty, I think the benefit of the doubt should be made in favor of having less patent restriction. Adding that factor in, my belief at this time is that patents don't seem to be worth it for the public, in practice.
Here is a link about TCPA as a threat to free software.
Slowing TCPA adoption is enough of an benefit to me to prefer a TCPA-free processor even if it costs $50 more for the same performance. I just hope I'll have that option for a while, as Intel is not the only company promoting TCPA.
By the way, that same FSF page says the same about IBM's other open source license, the Common Public License, used in Eclipse.
I hope that a future version of the GPL will allow LGPL-style linking with code covered by GPL-incompatible "free software" licenses in cases where no proprietary software is linked in or perhaps that a future version of the GPL will address software patent problems in a way that is compatible with the IBM Public License and the Apache 2 License.
In the meantime, to me, the advantages of Postfix are less than the disadvantages of the obstacles to future code recycling. If the incompatabilities could be fixed by a change to either the GPL or the IPL, and if enough of the hype about postfix proved to be real, I would probably switch to postfix.
I know I am a statistically insignificant sample, but there are others, mostly developers, who care about being able to copy code between free software projects in general and GPL compatability in particular. So, Postfix gain that much more usage and support if and when this incompatability is resolved, such as, for the short term, by adding some exceptions for the GPL, as the CUPS project did.
Thank you for your clarification. It helps me adjust my interpretation of your engineering claims.
Do you mean you simulated your design on a computer or that you physically constructed this airplane and it really flew?
"Silent" would mean no noise. A PC can be silent when the hard disk is spun down and all of its fans are stopped and nothing else in it is making any noise. You probably just meant "quiet."
I remember being in a terminal room at usenix many years ago which was full of Network Computing Devices X terminals, which were silent (diskless, fanless, no other parts moving or otherwise making sound). It was like being in a library. By far, the loudest remaining sound was the keystrokes from everyone typing.
I like silent (as opposed to quiet), PC's, although I've only used them in practice as thin clients.
When I see a vendor using the term "silent" to refer to a PC that is merely quiet, I consider that vendor to be a liar, and I generally can't trust them enough to be willing to do buy from them.
I also buy a lot of quiet hardware, like big slow fans, aluminum cases so I can disconnect more fans and so on, but I try never to buy from vendors that call these things "silent."
Personally, I doubt that any standard of patentability is high enough, duration short enough and scope narrow enough to make a patent system worth more than the value of the competition and incremental development that it stifles. Perhaps my views are skewed by being based so much on my seeing software development "before" and "after" software patents, but I think that with the advent of computer aided design and faster communications, the development of almost everything that is currently patentable is becoming more like software development or actually becoming software development.
Assuming that they're "on the same shelves" because it's an alphabetical filing system or some similar reason, I would have no problem with that.
If you want to argue for restricting people's freedoms or access to information, including children's, then the burden of proof is upon you to justify it. You didn't include any reasons that I could identify in your posting.
Product activation is amoral, any company doing that deserves a boycot.
Why does a company deserve a boycot for doing something amoral, "being neither moral nor immoral"?
"The law also supports the conclusion that Peak's loading of copyrighted software into RAM creates a "copy" of that software in violation of the Copyright Act." MAI Systems Corp. v. Peak Computer, US 9th Circuit federal appeals court.
By the way, I believe that the Digital Millenium Copyright Act slightly narrowed this ruling by putting a special case provision into title 17 United State Code section 117 that only applied to software that was necessary for booting when hardware maintenance was being done by a third party provider.
I am not a lawyer. Please do not use this as legal advice.
Although we discussed in another response that this could be covered by the transfer payments that I originally suggested, I want to point out that that I don't know if your claim is true anyhow, at least when you average in the additional unemployed people earning $0/hour due to minimum wage. I would be interested in hearing a basis for this prediction and seeing what that basis predicts if we were to raise the minimum wage to, say, $500/hour.
I don't think theres a problem finding jobs at the bottom right now, McDonalds seems to be perpetually hiring.
If that were true, then we should rescale welfare to assume that everyone at least works at McDonalds. I think that it's more likely that McDonalds franchises, like many businesses, find it economically worthwhile to give that impression so that many people will interview, and they can keep their turnover costs low. It probably is true that someone educated enough to read slashdot who wants to work at McDonald's will get hired immediately, but I don't think that's true for everyone who you describe as being "at the bottom."