WoW only does that for the test-drive client, but I found it worked quite well, I didn't find it laggy at all. Going to a new area could cause a short delay, but it appeared to cache it for use after that point.
So all someone has to do to hack your house is spoof your MAC/IP address? Oops! Makes much more sense that I have to make a secure connection to the "house controller", whether from inside or outside, including supplying security credentials (password, public key, Kerberos, SRP, whatever).
My point is that you already have to have a secure access point that authenticates you, and that can easily be done without IPv6. Using that as a REASON for deploying IPv6 is dumb.
You don't need to use dynamic DNS, either - you log in to a presence service which your home network is also logged in to and communicate with your home server(s) via that. That can either be a relayed tunnel or simply a connection service. You could set it up so you have to do that whether you're inside or outside your LAN and put all the security rules there. It's easy to get through a firewall if both sides can coordinate and cooperate to establish a direct connection through a NAT gateway, no UPNP required (it would be even easier if the TCP protocol directly supported a "meet" protocol rather than only a listen/connect style, but you can pretty much do it anyway).
BTW, my router does updates to dynamic DNS automatically whenever the address changes, for whatever reason. No need for a daemon process to do anything. It also supports inbound ad-hoc VPN connections; all my laptop at my in-law needs to do is connect to the inward VPN address (on the external IP address reported by whichever dynamic DNS service I'm using) and I'm on my LAN (the router also supports router-initiated inbound or outbound VPN connections for more permanent connections, of course).
But I don't want them addressable from the world at large; I want to go through a controller that verifies access and coordinates changes. That controller only needs one port on one address, everything INSIDE the house can all be on a local network with any type of addressing scheme you want. I'm not going to turn lights on or off by an IPv6 address, nor even by a DNS record that returns an IPv6 address, I'm going to turn lights on or off by saying "kitchen lights on" or "home theater lights fade to dark" or "dining room lights intimate". Then it can take the action as desired, including coordinating with automatic timers, climate control and everything else.
Same goes for the refrigerator, the thermostat, the home alarm system, the front-door camera or anything else you can think of. I don't need a lightbulb in the Ukraine talking to my refrigerator directly.
I don't see any problem at all with using a NAT layer at a gateway. Use a presence protocol and a host/service protocol (e.g. use something like an SSH tunnel, but instead of using addresses and ports, use a text name for what you're connecting to; this could simply be a local service (e.g. "home" for all functions controlling the home) or a "host:port" equivalent for connecting to specific machines (e.g. "MacG5:ssh") or almost any other scheme you can think of). You're going to have to go through an authentication structure anyway, and you'll probably want encryption as well; do you really want to have to log in to each lightbulb fixture you install and set up user names, passwords and security certificates?
"I don't want to xxx but yyy" is a common way of saying "when I say "yyy", it may sound like "xxx", but it's not. In other words, if that's what she really said, she was explicitly saying "I"m not going to hold you hostage". So she threatened him with not doing something. In addition, "holding someone hostage", absent physical restraint or threatening with a weapon, is a common expression of anything holding someone up, not unlawful restraint.
Bringing a cup of hot tea, "I don't want to scald you, so I'll hold on to this until you stand up." is also not threatening to scald someone with hot tea.
Finally, if she really did say what she claims, it sounds like the tech was giving lots of excuses, and her request to stay to exchange notes with the other tech was entirely reasonable.
If they're charging that much for Internet, and are using it to "pay for the infrastructure", makes you really wonder how much they're going to charge for flushing the toilet! That's gotta be a lot more expensive than an Internet connection.
No, I have no idea what you're talking about. You're saying wild things about NOAA regulating people taking photographs. We're talking about launching things into space. You said "legally launched a camera into space". If there's no person there with that camera, and it is sending back pictures to Earth, that is "remote sensing", thus it wasn't a legal launch unless you had a license from NOAA. You said NOAA doesn't have any jurisdiction, which means you haven't read the legislation.
Do you dispute the right of the US Government to control rocket launches from the United States or by its citizens and enforce regulations (based on international agreements) on where you can park a satellite so as not to cause problems for other space objects? Ok, you say, that's NASA that has jurisdiction over that. And guess what, NOAA has been given the task of overseeing (licensing) the subset of space launches that involve "remote sensing", just as the FCC has oversight on communications satellites. Why NOAA? Because they have the most experience with "remote sensing" technology.
It isn't remote sensing if you go up with the camera. I'm pretty sure the FAA, along with NASA, then has some jurisdiction. Since taking photographs potentially has security implications, there will no doubt be more regulations regarding "space tourists" taking photographs; and, no doubt, they won't cover hand-held cameras. Take all the pictures you want of Earth, the Moon, or anything else you want from space, NOAA won't care.
Note that part of the regulations regarding remote sensing requires a level of sharing of any data collected. Isn't that a good thing?
Well, your math is correct, but your numbers are suspect. Where'd you come up with 95% and a chain of 10 permits that you have to apply for one at a time?
Look, I'm a flight instructor, I've dealt with the FAA quite a bit, and for the most part they are dedicated people who really are out to serve the public. Occasionally you get an idiot, but most of the problems come from higher up, not with the "civil servants" you seem intent on bashing. From all accounts I've heard, NOAA people are not just mindless bureaucrats, but people interested in their field and work.
I think the people bitching about this are going to look awfully foolish when the permits start being approved for those groups that apply for them (I didn't check to see how long the licensing period is for, it may not even make sense to apply for this until they're a lot closer to launch).
BTW, with most permit situations where you need multiple permits/licenses, you usually get the approval process rolling on all of them at once; where there are prerequisites, most agencies I've dealt with can issue a permit/license that says "valid only with xxxx permit from xxxx agency" or whatever. You don't have to get one first, then get the next one, then the next one. Heck, for a flight instructor license, you need to have a commercial pilot's license. It isn't uncommon to get both of them on the same checkride. To take the knowledge exam (formerly known as the "written"), all you need is an instructor's authorization, and it's good for a year after that. So, you're getting close to the required hours for a commercial rating, you take the commercial and instructor knowledge exams, you go up for a checkride, and you come down with a new commercial license and instructor's license. Nobody requires you to get your commercial rating before you can take the instructor exams, heck, you can take the exams before you even get a STUDENT license.
So unless you have applications for all of the relevant permits and licenses you'd need to make a space launch, sitting there in your hands, and you can quote to me the sections where each says you can't send in the application until you've received other permits, you're just making stuff up. Four month lead time on a space launch simply doesn't sound like that big of a deal, what, you were planning on launching next weekend, maybe, and forgot to apply? Hmm, I'm not so sure I want you launching anywhere near me if you're that careless!
Let's see, you know that you can't even apply for a launch permit until you've got the remote sensing license in your hands? Where'd you find that out, and what reason do you have to believe that "a capricious civil servant" is going to be an ass and try to stop you and that you won't have any recourse if that should happen? What, did you fail a driving test after offering a beer to the examiner?
Remote. Remote. Remote. Remote. Damn, you even quoted me saying "remote" and ignored it. Guess what a person taking a picture isn't. Remote! Guess what the regulations regulate. Remote!
Guess what the regulations regarding manned space flight aren't going to cover. Remote! Yes, they'll probably put in requirements regarding using high-powered telescopes from space and what you can and can't do with them, but guess what won't be restricted. Your point-and-shoot Canon, that's for sure...
They have jurisdiction over people inside the United States who are engaged in activities that are covered by the regulations. These regulations, besides being concerned with security of photographs and such of sensitive areas, are also concerned with air-wave clutter, orbital debris, and even with ensuring that such data IS made available beyond what commercial interests might want to provide. There's nothing there trying to be protectionist that I see, neither of existing/future commercial interests nor of NOAA data itself. On the contrary.
Actually, the interest seems to be in PROMOTING it by making it available and regulating it so it is not a free-for-all, not "protecting it" it the way you're saying. There certainly needs to be SOME regulation, you don't want people sending up satellites in any-old-orbit, transmitting on any-old-frequency, shining laser lights down at your favorite observatory, or whatever...
Without such regulations, you'd be in a situation where they'd probably simply prohibit all such activities. Regulations like this are designed to PERMIT things to happen, while retaining enough control that it isn't chaotic. I didn't see any reference in the regulations to a fee for such a license.
The regulations in this part are
intended to:
(1) Preserve the national security of
the United States;
(2) Observe the foreign policies and
international obligations of the United
States;
(3) Advance and protect U.S. national
security and foreign policy interests by
maintaining U.S. leadership in remote
sensing space activities, and by
sustaining and enhancing the U.S.
remote sensing industry;
(4) Promote the broad use of remote
sensing data, their information products
and applications;
(5) Ensure that unenhanced data
collected by licensed private remote
sensing space systems concerning the
territory of any country are made
available to the government of that
country upon its request, as soon as
such data are available and on
reasonable commercial terms and
conditions as appropriate;
(6) Ensure that remotely sensed data
are widely available for civil and
scientific research, particularly
environmental and global change
research; and
(7) Maintain a permanent
comprehensive U.S. government archive
of global land remote sensing data for
long-term monitoring and study of the
changing global environment.
As for the space tourist taking along a camera, that's not "remote sensing".
I think you stated it well: "Being innocent is just as costly as being guilty". If innocent people trusted the justice system to exonerate them, even without having to hire a really expensive lawyer and put their life on hold for an indefinite amount of time, then you wouldn't see shady plea-bargain deals. The assumption, of course, is that someone who is really innocent would never plead guilty, but that just isn't true.
I think for the most part that the police and prosecutors are honest, really do try to find the guilty person, really do pay attention if they run across evidence that exonerates or points elsewhere. Most of them are dedicated to making the system really work. Unfortunately, that's only "most", there are too many who become convinced they know who did it, then only pay attention to evidence which supports their "certain" knowledge. This is a common human failing, but is very dangerous when present in people who have power; see, e.g. the Bush administration; ignoring aspects of incompetence, it is quite likely that some of the people there really do think that they're right, and that any cover-ups and deception are "end justifies the means", if only they can get people to leave them alone then all will be well with the world for everyone. Of course, some of them are just plain evil as well.
Dunno which CDC processor that might have been. With the various Cyber machines I worked with, one standard way to hang a process was to do an EQ * (branch to self); JP * was equally effective. It didn't hang the machine, just the process, and could easily be killed through NOS. In fact, standard practice was to make subroutine entry points be EQ *, so if you somehow entered the routine before it ever got called the process would hang. I never did that, I used PS, so the process would just halt immediately. All the operators ever did with a hung process was to drop it and cause the same crash dump that the PS did.
It really depends on the instruction set you're emulating. Once you go through the overhead of validating and picking apart a Java class, you could probably emulate it at around 10% of the native code a JIT compiler would emit. That's about the ratio I found when emulating versus (static Plenty-Of-Time) translation of one instruction set to another (including emulating one's complement arithmetic) (though the floating point routines were just translated into subroutine calls) (emulating a Cyber CDC CPU on an Alpha processor).
As for instruction cache, an emulator can actually be very efficient if the cache is large enough, running almost entirely out of that cache. In that case, the emulated processor has effectively a unified cache for data and instructions, in the form of the data cache of the native processor.
Why is that bogus? That kind of thing is sold with a pre-sale contract, with both sides completely clear on the terms. If you don't like such terms, don't sign the contract, get your jig from someone else or negotiate different terms. This is nothing like the typical EULA on consumer products where it is a simple sale, no opportunity for negotiation of terms, and often no way to even see the terms before you buy it, in an environment where there are significant barriers to competition due to compatibility concerns already exacerbated by copyright and patent issues.
The most amazing advances have really been in storage; my first computer was a Lisa with 1MB RAM and a 10MB hard drive; I can copy a 10MB chunk off my 1TB hard drive into what is now a relatively small 10MB buffer (out of 4GB of RAM) in about a second, and search that buffer for an arbitrary string in about.1 second. It used to take a long time to format that 10MB drive, and even longer to copy a directory hierarchy or search through a bunch of files. I can't think of many programs you can even buy that fit into 10MB now, much less run in 1MB of RAM, but that 10MB used to hold three different operating systems (the Lisa office suite with 7 different applications; the programming environment; and the Mac OS environment); and it booted into any one of them in a relative jiffy, running on a 68010 processor (at 6MHz I think it was).
Seems sort of odd to be trying to patent something that clearly can't be made to work at the present time. I don't see why this would be patentable anyway, similar devices have certainly been described in various science fiction stories in sufficient detail to be "prior art".
No, if they own the copyright (and the binary doesn't contain anyone else's code that is GPLed itself), they do not have to follow any license. License terms don't apply to the owner of the copyright. They own it, they can do what they want with it. They can, for example, license it to someone else with terms other than the GPL. That wouldn't affect any copies that they licensed under the GPL, of course.
If they do release their own code under the GPL, it would seem quite mad to object to providing source code. What they might do is avoid the "make available for 3 years" provision; provide source code now as people request it, but reserve the right to stop providing it at any time, which someone distributing GPL code written by others can do only by providing the source code WITH the binary, something that can be an annoying requirement when distributing the binary as part of a device (for example).
If they release a binary that does contain someone else's GPLed code, then of course they need to comply with the GPL and provide their own code as well, since otherwise they wouldn't have permission to use that other code.
Since this particular case is not one that fits the example given ("Consider a company that manages to create a de facto standard based on GPL software..."), it doesn't seem relevant to anti-trust.
Yeah, I had a game that my parents bought me for Christmas once, for the Mac. It mentioned something about changing the difficulty level to "impossible" if it was a pirated copy, but since I had a perfectly legitimate copy, I didn't worry about it. Until, after playing for a couple hours building up what I thought was a very good position (it was a strategy/resource-building type game) I was suddenly blitzed by an incredible number of enemy units and immediately lost. My only guess is that the copy-protection worked on a Mac Plus, but since I was using a Mac II it didn't work for some reason.
So, I never bothered to play that game again, never recommended it to anyone, never bought anything from them, and they're out of business. The net effect was that a legitimate customer never wants to give them any more money, a naive "pirate" would simply think the game sucked and wouldn't even consider buying it, and people who download a cracked copy don't have any problems. Brilliant!
Re:Blizzard may be my favorite company, but please
on
Who Owns Software?
·
· Score: 1
The problem with that is there are a wide range of players. If you make a game complicated/difficult enough that a bot starts to have problems with it, then some players will also find it too difficult, especially new players, and you can't automate a Turing test (an effective automated Turing test would itself be likely to pass a Turing test).
Because the cost to supply the network is based on peak usage, not actual usage. It costs more money to provide you with more electricity, it does not cost more money to transmit a byte of data when nothing else would be going over the line anyway. Charging for that byte is not efficient.
The solution is to do throttling such that anyone using less than the throttle limit is not affected at all, and everyone else is throttled at whatever limit is necessary to support the current demand. Measure the load for each customer with a fairly quick moving average, such that sitting and doing nothing for a couple minutes brings you down to zero, and you get maybe 30 seconds of max-rate (whatever your connection supports) before you start to be throttled down (and then, only when demand requires it; at low demand time you would get a much higher bandwidth with no "monthly caps" or anything like that).
Then price it based on actual cost and what people are willing to pay for a certain level of service. You can have different tiers of service still, just allow a multiplier on the throttle limit per customer; someone who pays twice as much gets twice as much bandwidth as the base (unless they need less than twice the base throttle level, in which case they have same priority as anyone else who is under the throttle limit).
This allows for a wide range of needs, a lot of people wouldn't need anything other than the base rate, high-bandwidth users would get better service at low-peak times, and even when using it during high-peak times would not be affecting lower-bandwidth users at all. The ISP can expand peak capacity based on what people are willing to pay for.
Do the Apple panels do temporal dithering? If so, they certainly have a case. Why is temporal dithering any worse than the spatial dithering that turns 3 discrete colors into a simulation of the colors we see?
"Your honor, we inspected the panel and didn't even find any YELLOW pixels; Apple cheats and creates yellow out of a combination of red and green pixels. They don't have millions of colors, they only have THREE colors!"
Slashdot Mirror
WoW only does that for the test-drive client, but I found it worked quite well, I didn't find it laggy at all. Going to a new area could cause a short delay, but it appeared to cache it for use after that point.
So all someone has to do to hack your house is spoof your MAC/IP address? Oops! Makes much more sense that I have to make a secure connection to the "house controller", whether from inside or outside, including supplying security credentials (password, public key, Kerberos, SRP, whatever).
My point is that you already have to have a secure access point that authenticates you, and that can easily be done without IPv6. Using that as a REASON for deploying IPv6 is dumb.
You don't need to use dynamic DNS, either - you log in to a presence service which your home network is also logged in to and communicate with your home server(s) via that. That can either be a relayed tunnel or simply a connection service. You could set it up so you have to do that whether you're inside or outside your LAN and put all the security rules there. It's easy to get through a firewall if both sides can coordinate and cooperate to establish a direct connection through a NAT gateway, no UPNP required (it would be even easier if the TCP protocol directly supported a "meet" protocol rather than only a listen/connect style, but you can pretty much do it anyway).
BTW, my router does updates to dynamic DNS automatically whenever the address changes, for whatever reason. No need for a daemon process to do anything. It also supports inbound ad-hoc VPN connections; all my laptop at my in-law needs to do is connect to the inward VPN address (on the external IP address reported by whichever dynamic DNS service I'm using) and I'm on my LAN (the router also supports router-initiated inbound or outbound VPN connections for more permanent connections, of course).
Having a unique identifier doesn't imply that it needs to be directly network-addressable from any other entity with a unique identifier.
For a unique-identifier scheme, see The Object Identifier Repository.
But I don't want them addressable from the world at large; I want to go through a controller that verifies access and coordinates changes. That controller only needs one port on one address, everything INSIDE the house can all be on a local network with any type of addressing scheme you want. I'm not going to turn lights on or off by an IPv6 address, nor even by a DNS record that returns an IPv6 address, I'm going to turn lights on or off by saying "kitchen lights on" or "home theater lights fade to dark" or "dining room lights intimate". Then it can take the action as desired, including coordinating with automatic timers, climate control and everything else.
Same goes for the refrigerator, the thermostat, the home alarm system, the front-door camera or anything else you can think of. I don't need a lightbulb in the Ukraine talking to my refrigerator directly.
I don't see any problem at all with using a NAT layer at a gateway. Use a presence protocol and a host/service protocol (e.g. use something like an SSH tunnel, but instead of using addresses and ports, use a text name for what you're connecting to; this could simply be a local service (e.g. "home" for all functions controlling the home) or a "host:port" equivalent for connecting to specific machines (e.g. "MacG5:ssh") or almost any other scheme you can think of). You're going to have to go through an authentication structure anyway, and you'll probably want encryption as well; do you really want to have to log in to each lightbulb fixture you install and set up user names, passwords and security certificates?
"I don't want to xxx but yyy" is a common way of saying "when I say "yyy", it may sound like "xxx", but it's not. In other words, if that's what she really said, she was explicitly saying "I"m not going to hold you hostage". So she threatened him with not doing something. In addition, "holding someone hostage", absent physical restraint or threatening with a weapon, is a common expression of anything holding someone up, not unlawful restraint.
Bringing a cup of hot tea, "I don't want to scald you, so I'll hold on to this until you stand up." is also not threatening to scald someone with hot tea.
Finally, if she really did say what she claims, it sounds like the tech was giving lots of excuses, and her request to stay to exchange notes with the other tech was entirely reasonable.
If they're charging that much for Internet, and are using it to "pay for the infrastructure", makes you really wonder how much they're going to charge for flushing the toilet! That's gotta be a lot more expensive than an Internet connection.
No, I have no idea what you're talking about. You're saying wild things about NOAA regulating people taking photographs. We're talking about launching things into space. You said "legally launched a camera into space". If there's no person there with that camera, and it is sending back pictures to Earth, that is "remote sensing", thus it wasn't a legal launch unless you had a license from NOAA. You said NOAA doesn't have any jurisdiction, which means you haven't read the legislation.
Do you dispute the right of the US Government to control rocket launches from the United States or by its citizens and enforce regulations (based on international agreements) on where you can park a satellite so as not to cause problems for other space objects? Ok, you say, that's NASA that has jurisdiction over that. And guess what, NOAA has been given the task of overseeing (licensing) the subset of space launches that involve "remote sensing", just as the FCC has oversight on communications satellites. Why NOAA? Because they have the most experience with "remote sensing" technology.
It isn't remote sensing if you go up with the camera. I'm pretty sure the FAA, along with NASA, then has some jurisdiction. Since taking photographs potentially has security implications, there will no doubt be more regulations regarding "space tourists" taking photographs; and, no doubt, they won't cover hand-held cameras. Take all the pictures you want of Earth, the Moon, or anything else you want from space, NOAA won't care.
Note that part of the regulations regarding remote sensing requires a level of sharing of any data collected. Isn't that a good thing?
Well, your math is correct, but your numbers are suspect. Where'd you come up with 95% and a chain of 10 permits that you have to apply for one at a time?
Look, I'm a flight instructor, I've dealt with the FAA quite a bit, and for the most part they are dedicated people who really are out to serve the public. Occasionally you get an idiot, but most of the problems come from higher up, not with the "civil servants" you seem intent on bashing. From all accounts I've heard, NOAA people are not just mindless bureaucrats, but people interested in their field and work.
I think the people bitching about this are going to look awfully foolish when the permits start being approved for those groups that apply for them (I didn't check to see how long the licensing period is for, it may not even make sense to apply for this until they're a lot closer to launch).
BTW, with most permit situations where you need multiple permits/licenses, you usually get the approval process rolling on all of them at once; where there are prerequisites, most agencies I've dealt with can issue a permit/license that says "valid only with xxxx permit from xxxx agency" or whatever. You don't have to get one first, then get the next one, then the next one. Heck, for a flight instructor license, you need to have a commercial pilot's license. It isn't uncommon to get both of them on the same checkride. To take the knowledge exam (formerly known as the "written"), all you need is an instructor's authorization, and it's good for a year after that. So, you're getting close to the required hours for a commercial rating, you take the commercial and instructor knowledge exams, you go up for a checkride, and you come down with a new commercial license and instructor's license. Nobody requires you to get your commercial rating before you can take the instructor exams, heck, you can take the exams before you even get a STUDENT license.
So unless you have applications for all of the relevant permits and licenses you'd need to make a space launch, sitting there in your hands, and you can quote to me the sections where each says you can't send in the application until you've received other permits, you're just making stuff up. Four month lead time on a space launch simply doesn't sound like that big of a deal, what, you were planning on launching next weekend, maybe, and forgot to apply? Hmm, I'm not so sure I want you launching anywhere near me if you're that careless!
Let's see, you know that you can't even apply for a launch permit until you've got the remote sensing license in your hands? Where'd you find that out, and what reason do you have to believe that "a capricious civil servant" is going to be an ass and try to stop you and that you won't have any recourse if that should happen? What, did you fail a driving test after offering a beer to the examiner?
Remote. Remote. Remote. Remote. Damn, you even quoted me saying "remote" and ignored it. Guess what a person taking a picture isn't. Remote! Guess what the regulations regulate. Remote!
Guess what the regulations regarding manned space flight aren't going to cover. Remote! Yes, they'll probably put in requirements regarding using high-powered telescopes from space and what you can and can't do with them, but guess what won't be restricted. Your point-and-shoot Canon, that's for sure...
The regulations don't prohibit photography by people, standing, sitting, flying or orbiting. Do you have any idea what you're talking about? At all?
They have jurisdiction over people inside the United States who are engaged in activities that are covered by the regulations. These regulations, besides being concerned with security of photographs and such of sensitive areas, are also concerned with air-wave clutter, orbital debris, and even with ensuring that such data IS made available beyond what commercial interests might want to provide. There's nothing there trying to be protectionist that I see, neither of existing/future commercial interests nor of NOAA data itself. On the contrary.
This law wouldn't apply to space tourists. By definition, they aren't "remote sensing".
Actually, the interest seems to be in PROMOTING it by making it available and regulating it so it is not a free-for-all, not "protecting it" it the way you're saying. There certainly needs to be SOME regulation, you don't want people sending up satellites in any-old-orbit, transmitting on any-old-frequency, shining laser lights down at your favorite observatory, or whatever...
Without such regulations, you'd be in a situation where they'd probably simply prohibit all such activities. Regulations like this are designed to PERMIT things to happen, while retaining enough control that it isn't chaotic. I didn't see any reference in the regulations to a fee for such a license.
As for the space tourist taking along a camera, that's not "remote sensing".
I think you stated it well: "Being innocent is just as costly as being guilty". If innocent people trusted the justice system to exonerate them, even without having to hire a really expensive lawyer and put their life on hold for an indefinite amount of time, then you wouldn't see shady plea-bargain deals. The assumption, of course, is that someone who is really innocent would never plead guilty, but that just isn't true.
I think for the most part that the police and prosecutors are honest, really do try to find the guilty person, really do pay attention if they run across evidence that exonerates or points elsewhere. Most of them are dedicated to making the system really work. Unfortunately, that's only "most", there are too many who become convinced they know who did it, then only pay attention to evidence which supports their "certain" knowledge. This is a common human failing, but is very dangerous when present in people who have power; see, e.g. the Bush administration; ignoring aspects of incompetence, it is quite likely that some of the people there really do think that they're right, and that any cover-ups and deception are "end justifies the means", if only they can get people to leave them alone then all will be well with the world for everyone. Of course, some of them are just plain evil as well.
Dunno which CDC processor that might have been. With the various Cyber machines I worked with, one standard way to hang a process was to do an EQ * (branch to self); JP * was equally effective. It didn't hang the machine, just the process, and could easily be killed through NOS. In fact, standard practice was to make subroutine entry points be EQ *, so if you somehow entered the routine before it ever got called the process would hang. I never did that, I used PS, so the process would just halt immediately. All the operators ever did with a hung process was to drop it and cause the same crash dump that the PS did.
It really depends on the instruction set you're emulating. Once you go through the overhead of validating and picking apart a Java class, you could probably emulate it at around 10% of the native code a JIT compiler would emit. That's about the ratio I found when emulating versus (static Plenty-Of-Time) translation of one instruction set to another (including emulating one's complement arithmetic) (though the floating point routines were just translated into subroutine calls) (emulating a Cyber CDC CPU on an Alpha processor).
As for instruction cache, an emulator can actually be very efficient if the cache is large enough, running almost entirely out of that cache. In that case, the emulated processor has effectively a unified cache for data and instructions, in the form of the data cache of the native processor.
Why is that bogus? That kind of thing is sold with a pre-sale contract, with both sides completely clear on the terms. If you don't like such terms, don't sign the contract, get your jig from someone else or negotiate different terms. This is nothing like the typical EULA on consumer products where it is a simple sale, no opportunity for negotiation of terms, and often no way to even see the terms before you buy it, in an environment where there are significant barriers to competition due to compatibility concerns already exacerbated by copyright and patent issues.
The most amazing advances have really been in storage; my first computer was a Lisa with 1MB RAM and a 10MB hard drive; I can copy a 10MB chunk off my 1TB hard drive into what is now a relatively small 10MB buffer (out of 4GB of RAM) in about a second, and search that buffer for an arbitrary string in about .1 second. It used to take a long time to format that 10MB drive, and even longer to copy a directory hierarchy or search through a bunch of files. I can't think of many programs you can even buy that fit into 10MB now, much less run in 1MB of RAM, but that 10MB used to hold three different operating systems (the Lisa office suite with 7 different applications; the programming environment; and the Mac OS environment); and it booted into any one of them in a relative jiffy, running on a 68010 processor (at 6MHz I think it was).
Seems sort of odd to be trying to patent something that clearly can't be made to work at the present time. I don't see why this would be patentable anyway, similar devices have certainly been described in various science fiction stories in sufficient detail to be "prior art".
No, if they own the copyright (and the binary doesn't contain anyone else's code that is GPLed itself), they do not have to follow any license. License terms don't apply to the owner of the copyright. They own it, they can do what they want with it. They can, for example, license it to someone else with terms other than the GPL. That wouldn't affect any copies that they licensed under the GPL, of course.
If they do release their own code under the GPL, it would seem quite mad to object to providing source code. What they might do is avoid the "make available for 3 years" provision; provide source code now as people request it, but reserve the right to stop providing it at any time, which someone distributing GPL code written by others can do only by providing the source code WITH the binary, something that can be an annoying requirement when distributing the binary as part of a device (for example).
If they release a binary that does contain someone else's GPLed code, then of course they need to comply with the GPL and provide their own code as well, since otherwise they wouldn't have permission to use that other code.
Since this particular case is not one that fits the example given ("Consider a company that manages to create a de facto standard based on GPL software..."), it doesn't seem relevant to anti-trust.
Yeah, I had a game that my parents bought me for Christmas once, for the Mac. It mentioned something about changing the difficulty level to "impossible" if it was a pirated copy, but since I had a perfectly legitimate copy, I didn't worry about it. Until, after playing for a couple hours building up what I thought was a very good position (it was a strategy/resource-building type game) I was suddenly blitzed by an incredible number of enemy units and immediately lost. My only guess is that the copy-protection worked on a Mac Plus, but since I was using a Mac II it didn't work for some reason.
So, I never bothered to play that game again, never recommended it to anyone, never bought anything from them, and they're out of business. The net effect was that a legitimate customer never wants to give them any more money, a naive "pirate" would simply think the game sucked and wouldn't even consider buying it, and people who download a cracked copy don't have any problems. Brilliant!
The problem with that is there are a wide range of players. If you make a game complicated/difficult enough that a bot starts to have problems with it, then some players will also find it too difficult, especially new players, and you can't automate a Turing test (an effective automated Turing test would itself be likely to pass a Turing test).
Because the cost to supply the network is based on peak usage, not actual usage. It costs more money to provide you with more electricity, it does not cost more money to transmit a byte of data when nothing else would be going over the line anyway. Charging for that byte is not efficient.
The solution is to do throttling such that anyone using less than the throttle limit is not affected at all, and everyone else is throttled at whatever limit is necessary to support the current demand. Measure the load for each customer with a fairly quick moving average, such that sitting and doing nothing for a couple minutes brings you down to zero, and you get maybe 30 seconds of max-rate (whatever your connection supports) before you start to be throttled down (and then, only when demand requires it; at low demand time you would get a much higher bandwidth with no "monthly caps" or anything like that).
Then price it based on actual cost and what people are willing to pay for a certain level of service. You can have different tiers of service still, just allow a multiplier on the throttle limit per customer; someone who pays twice as much gets twice as much bandwidth as the base (unless they need less than twice the base throttle level, in which case they have same priority as anyone else who is under the throttle limit).
This allows for a wide range of needs, a lot of people wouldn't need anything other than the base rate, high-bandwidth users would get better service at low-peak times, and even when using it during high-peak times would not be affecting lower-bandwidth users at all. The ISP can expand peak capacity based on what people are willing to pay for.
Do the Apple panels do temporal dithering? If so, they certainly have a case. Why is temporal dithering any worse than the spatial dithering that turns 3 discrete colors into a simulation of the colors we see?
"Your honor, we inspected the panel and didn't even find any YELLOW pixels; Apple cheats and creates yellow out of a combination of red and green pixels. They don't have millions of colors, they only have THREE colors!"