Not true. I've got a Android (Samsung Galaxy recent model), and I've never activated an internet connection. I must have one somehow, because it's had me update my software a couple of times, but I never connect, and I haven't given it any permissions on my network. I suspect it's calling over the cell phone lines.
I'll grant that it keep asking me to do an internet connection, but I got it because I needed a phone. If I'd been able to get a simple phone as quickly, I would have preferred that.
This does, however, prove that you don't need to enable internet access on an Android phone...perhaps you bought yours from a vendor who customized things so that it does require it.
Troll or incompetent? I can't tell. I've known a few people incompetent enough to feel that way, but they feel the same way every time they get a MSWind upgrade.
No. A mass driver is a catapult based engine. I'm looking for an ion rocket based on fumes boiled out of rock, probably by a laser. Fast ejection velocity, not large amounts of ejection. I'd agree that a mass driver would probably be easier to build, but it's not at all what I was looking for.
FWIW, I'm dreaming of far ahead, with space habitats being long term living facilities that move slowly between the stars scavenging from free planets and asteroids, and moving at just enough faster than the local drift to continually encounter new areas to mine for resources. So my dream engine is quite conservative with respect to it's need for mass, but also doesn't require exotic substances that are hard to find outside of a deep gravity well. Hydrogen (relatively) is abundant in the form of methane, ammonia, etc. Silicon dioxide is reasonably available in the form of rocks. Fuse hydrogen for energy, use the energy to split Silicon Dioxide into Silicon and Oxygen. Save the Oxygen for breathing, etc. eject the Silicon as high velocity (as high as you can reasonably get) ions. A closed ecology can only go so far. There will always be leaks and the need for new resources. But you don't want to expend more than you need to.
We aren't near that point yet, but that's the direction I'd like our tools to be developing.
Perhaps I misunderstand the assertion. Perhaps you are measuring total RAM rather than directly addressable fast RAM. If that's so, then the rest of this response is inappropriate,
But my first reaction was as follows:
Well, if you amend that to "people who feel the must", I'll agree, but virtual memory means that this is a silly attitude. I don't believe that anybody is actively working with 64TB of data at once (i.e., within, say, the same second). If they think you need that then either they're wrong or the machine is improperly designed. I suppose there could be cases where you can't predict which bytes you'll need next, but 64TB is a big cache to have in high-speed storage at once, and should more than suffice for anything. What you need backing it is a huge intermediate speed storage between it and the disk, and I will agree that that intermediate speed storage might well be more sizable. Actually, I suppose that "intermediate" speed storage would be a third level of speed, with CPU registers being fast, followed by a larger fast RAM cache, followed by slower RAM, followed by more permanent storage, with each successive layer being larger but slower than the one before it, but I can see no reason for the fast RAM to be larger than 64TB for anything. In fact that seems overkill. There might be reasons for the lowest RAM level to be as large as the backing store (disks?), but I doubt it. It sounds like poor design. The backing disks should always be considerably larger than the RAM that they are backing, and the slow RAM should be considerably larger than the fast RAM that they are backing, and the fast RAM should be considerably larger than the CPU registers that they are backing. And if the disks are backed up to tape, the tape library should be considerably larger than the disks that they are backing.
Some of the programs I could run in emulation, though not well, since many of them were timing dependent. But I couldn't get the data from the files.
Recently I took an old computer, booted it up, and *printed* a bunch of the files to pdf. This gave me the data, but not in a very usable form, then I transferred the pdfs to a different computer, and now I'm in the process of recovering from the pdfs. This has not left me very happy about programs that won't usably export the data. And by usably I mean in a form that I, as a programmer, can use. Other data is for systems that died before virtualization became practical. There's less of that because most has already been thrown away. But how would you go about reading a hard sectored 8-inch floppy in Wang format? There has never been a way to read that since the system died.
I tried for quite awhile to run MSWind95 in a VM. It never quite worked. There didn't seem to be an option to run MacOS 7.2 in a VM.
So, yes, it is a reason. And the particular software that I'm running under a VM in Linux is a game, so I didn't pay attention to it being closed source. Then it acted the way one needs to expect closed source software to act...but since it was Linux I have been able to keep it usable.
Hydrogen has the advantage that it's available anywhere you can find frozen methane. Xenon can be a bit harder to scavenge. What I'd really like is a "high" power ion rocket that could use rocks for exhaust. This, though, is a big problem because rocks aren't a simple element, but a complex mix that varies. (By "high power" I'm thinking of about 30 pounds thrust, but that's probably dreaming.)
Actually, that's probably true, though there might be corner cases...
The thing is, when you've got that much addressable space you should probably be doing paging with an LRU cache flush to an intermediate level of memory, which should save itself to disk in idle moments. This would take about 64 bits/block. One bit for "changed since read from/written to disk" and a bunch for "time of last access".
OTOH, I have my doubts that they actually have 64TB of RAM. I expect they just have a memory-mapped disk with an LRU cache, so you can address it as if there were 64TB or RAM, and in that case I can see cases where 64TB wouldn't be enough. As far as the OS was concerned, it would look like hardware with a HUGE RAM, even though it was actually a hardware paged disk. This would speed up both power up and power down, and would have other benefits. And for a system like that I can see that an addressable space of n*64TB might always end up being too small for certain classes of problems (for any fixed n).
But for an actual RAM vs. disk split I think that 64TB RAM should be enough for anyone. Even that would slow down power up and power down tremendously unless you don't care about losing what currently in RAM, and don't care that you're initializing the system with lots of random garbage.
MUCH better. At least you can figure out how to read the data files, even if nothing else. I've ended up with a bunch of files from dropped closed source projects that represent a huge amount of work and are totally worthless. Now I won't depend on any software that isn't FOSS.
A problem with FOSS software is that it often *does* have gating dependencies on the current developer. They aren't so much legal as informational, and sometimes interest, but they're there.
That said, this is much less severe with FOSS projects than with closed source projects, and the legal obstacles are absent, but that doesn't mean that there aren't any dependencies.
E.g. (a story from the 1970's). There was this shoe store that used software to do its taxes, and since computer time was expensive the compiler output was hand tuned for speed. At some point the original source code was lost, and all that was left was a binary program that did their taxes quickly and efficiently. The guy who wrote it had left for another job years ago. Then the IRS decided to audit them... Whoops!... but they had all rights to the code, so that wasn't a problem. And the binary was isomorphic to the tuned assembler code, so that wasn't a problem. But *NOBODY* knew how it worked, what it did, or why it did it.
If nobody can understand it, it *will* rot. This may be a disaster as far as the project is concerned. But if nobody really understands what it is doing, replacing it can be...problematic.
Before I switched of Linux my projects were continually depending on software that people and companies just dropped for one reason or another. Going out of business was one of the major reasons, but not the only one. I've still got files in various of them that I haven't got around to throwing away yet, but which are totally useless because there's no possible way to run them. TurboCAD is one that I just threw away today. I think they went out of business. Deluxe Music Construction Set I finished cleaning out earlier this year, they haven't worked since MacOS 7.5. The list is long and unhappy, full of work that was lost because of companies that stopped supporting their users. This hasn't happened once since I moved to Linux...though there are a couple of programs that only run in a virtualized Red Hat Linux 2.4 environment.
Things don't happen that way, but most "open source" projects don't end up with ownership problems so much as maintainer problems. It's quite rare that the legal inheritor of the copyright has either the desire or the ability to maintain it. With most open source licenses, certainly with GPL, the next maintainer retains the right to alter and release new versions of the code even without legal ownership of the copyright to the existing code. But capability is an entirely separate matter. There can also be problems with the right to use the name, but, again, that doesn't usually arise with death.
Actually I believe that for FOSS projects the two main problems in this area are the right to continue using the name and the capability to maintain the code (and interest in doing so). And that doesn't always wait for somebody to die to become a problem. I know Debian is frequently asking for someone to adopt a package that has lost its maintainer...and often the reason is loss of interest rather than death.
All that said, I know that in the Python community awhile ago (over a decade now?) there was significant concern about "What do we do if Guido is hit by a bus?", and the result was the current Python Software Foundation. Smaller projects, however, would probably find that solution expensive overkill.
The problem with that argument is that "intelligence" is an undefined term, unless you are arguing that IQ is intelligence, I which case I deny your original assertion.
What people have that has been our mainstay of success is culture. Human culture depends on language, though it clearly includes a lot of other things too. (Various monkeys and apes have been shown to have a primitive culture, but without language cultural transmission is labored and limited.) And unlike intelligence, what culture is at the basis is easily defined, it's the ability to pick up ways of doing things that are different from the ways your ancestors did. So some monkeys moving to the sea shore learn to crack open shell-fish. That's a primitive cultural advance. But notice that culture is the characteristic of a group rather than of an individual. One monkey learning to crack open a shell isn't a cultural advance. It only becomes such when it is transmitted to other monkeys.
You overstate the case. IQ is correlated with many useful skills. But it sure hasn't been shown to correlate with others.
For that matter, IQ itself is not a unitary measure. The tests measure different, and perhaps independent (but certainly not varying identically) capabilities. Sometimes the capabilities are correlated, but not identical, as, e.g., the ability to maintain focus and the ability to memorize. But as I know of no accepted separation of the capabilities measured by IQ, I doubt that there's even been significant research as to exactly WHAT is measured. But it does predict with reasonable accuracy your ability to do well on the average school test (without even attempting to measure motivation to do well over time).
FWIW, the editors and reporters were always biased, back as far as we have records. But they used to be separately biased for each paper, and they used to worry about their reputation. Now the papers are owned by conglomerates who are mainly interested in the papers as pushing their particular message, whatever that it. And telecommunications allows micromanaging by owners at a distance.
Even the purely local papers around here have been bought up by chains, and local news has dropped off precipitously.
This trend was first (to my knowledge) commented on during the 1960's. Up until then most large cities had more than one locally owned and operated paper. Mergers had already happened more than once, though, (in the 1950s Honolulu had the Star-Bulletin, which was the merger of the Star and the Bulletin) so it may have been an on-going process.
Not for their own code. Developers are lousy at testing their own code, because they handle the errors that they think of, and won't make the ones they don't.
I suspect that developers are even poor choices for testing other developers code, because they're less likely to make ID10T errors.
What changed is something that didn't change. Apple has no interest in allowing non-Apple software to run, so their "security" application disallowed Audacity, despite that Audacity had been running under MacOS for a long time as a useful tool.
Perhaps Apple has a competing product. I don't know, they didn't when I was using them a bit over a decade ago. But my guess is that just just broke anything that wasn't too prominent that they weren't selling.
No, but it does prevent it from being very useful. For that technology to be useful in preventing accidents probably well over half the cars on the road would need to use it. Possibly over 3/4. So even if it were enforced it would take several years before it would do much good. And it mainly helps the vehicles avoiding someone else behaving illegally...and possibly mainly helps vehicles under automatic control. That would depend on implementation details.
I'd have to check, but I've probably got over 1000 passwords. Most of them are defunct, but sometimes I'm surprised that one I haven't used for a decade will still work. Some of them predate commercial access on the internet. Some of them are from when I was sysadmining a multi-user CP/M machine (with 16 processors!). But there are far too many to remember, especially for sites that I haven't accessed in a decade...and, surprise!, sometimes I need access again, and it's really nice if the old password still works. And you can't predict which ones.
You can't be sure of that. I'll admit that was my first take on the article, but it's not something I feel certain about.
That said, my second take on the article was "Oh, no. A *different* group of people will be in charge."
I didn't consider that it might be his actual opinion until my third take. And I see no way to choose between those possibilities (and there are probably a few more). I can't even estimate the degree of probability. But that he was stating his honest opinion is one plausible choice, even if it's not the most probable.
Remote location: Periodically mail a backup to your sister on the other coast.
It all depends on who you are, how rapidly your important data changes, and how much those changes are worth to you. I didn't address secrecy, which would, indeed, complicate matters.
OK, I'd be more willing to believe Apple than some others. That, however, is a long way from saying I'd believe their claim without actual proof for anything important.
Slashdot Mirror
Not true. I've got a Android (Samsung Galaxy recent model), and I've never activated an internet connection. I must have one somehow, because it's had me update my software a couple of times, but I never connect, and I haven't given it any permissions on my network. I suspect it's calling over the cell phone lines.
I'll grant that it keep asking me to do an internet connection, but I got it because I needed a phone. If I'd been able to get a simple phone as quickly, I would have preferred that.
This does, however, prove that you don't need to enable internet access on an Android phone...perhaps you bought yours from a vendor who customized things so that it does require it.
Troll or incompetent? I can't tell. I've known a few people incompetent enough to feel that way, but they feel the same way every time they get a MSWind upgrade.
No. A mass driver is a catapult based engine. I'm looking for an ion rocket based on fumes boiled out of rock, probably by a laser. Fast ejection velocity, not large amounts of ejection. I'd agree that a mass driver would probably be easier to build, but it's not at all what I was looking for.
FWIW, I'm dreaming of far ahead, with space habitats being long term living facilities that move slowly between the stars scavenging from free planets and asteroids, and moving at just enough faster than the local drift to continually encounter new areas to mine for resources. So my dream engine is quite conservative with respect to it's need for mass, but also doesn't require exotic substances that are hard to find outside of a deep gravity well. Hydrogen (relatively) is abundant in the form of methane, ammonia, etc. Silicon dioxide is reasonably available in the form of rocks. Fuse hydrogen for energy, use the energy to split Silicon Dioxide into Silicon and Oxygen. Save the Oxygen for breathing, etc. eject the Silicon as high velocity (as high as you can reasonably get) ions. A closed ecology can only go so far. There will always be leaks and the need for new resources. But you don't want to expend more than you need to.
We aren't near that point yet, but that's the direction I'd like our tools to be developing.
Perhaps I misunderstand the assertion. Perhaps you are measuring total RAM rather than directly addressable fast RAM. If that's so, then the rest of this response is inappropriate,
But my first reaction was as follows:
Well, if you amend that to "people who feel the must", I'll agree, but virtual memory means that this is a silly attitude. I don't believe that anybody is actively working with 64TB of data at once (i.e., within, say, the same second). If they think you need that then either they're wrong or the machine is improperly designed. I suppose there could be cases where you can't predict which bytes you'll need next, but 64TB is a big cache to have in high-speed storage at once, and should more than suffice for anything. What you need backing it is a huge intermediate speed storage between it and the disk, and I will agree that that intermediate speed storage might well be more sizable. Actually, I suppose that "intermediate" speed storage would be a third level of speed, with CPU registers being fast, followed by a larger fast RAM cache, followed by slower RAM, followed by more permanent storage, with each successive layer being larger but slower than the one before it, but I can see no reason for the fast RAM to be larger than 64TB for anything. In fact that seems overkill. There might be reasons for the lowest RAM level to be as large as the backing store (disks?), but I doubt it. It sounds like poor design. The backing disks should always be considerably larger than the RAM that they are backing, and the slow RAM should be considerably larger than the fast RAM that they are backing, and the fast RAM should be considerably larger than the CPU registers that they are backing. And if the disks are backed up to tape, the tape library should be considerably larger than the disks that they are backing.
Some of the programs I could run in emulation, though not well, since many of them were timing dependent. But I couldn't get the data from the files.
Recently I took an old computer, booted it up, and *printed* a bunch of the files to pdf. This gave me the data, but not in a very usable form, then I transferred the pdfs to a different computer, and now I'm in the process of recovering from the pdfs. This has not left me very happy about programs that won't usably export the data. And by usably I mean in a form that I, as a programmer, can use. Other data is for systems that died before virtualization became practical. There's less of that because most has already been thrown away. But how would you go about reading a hard sectored 8-inch floppy in Wang format? There has never been a way to read that since the system died.
I tried for quite awhile to run MSWind95 in a VM. It never quite worked. There didn't seem to be an option to run MacOS 7.2 in a VM.
So, yes, it is a reason. And the particular software that I'm running under a VM in Linux is a game, so I didn't pay attention to it being closed source. Then it acted the way one needs to expect closed source software to act...but since it was Linux I have been able to keep it usable.
Hydrogen has the advantage that it's available anywhere you can find frozen methane. Xenon can be a bit harder to scavenge. What I'd really like is a "high" power ion rocket that could use rocks for exhaust. This, though, is a big problem because rocks aren't a simple element, but a complex mix that varies. (By "high power" I'm thinking of about 30 pounds thrust, but that's probably dreaming.)
Actually, that's probably true, though there might be corner cases...
The thing is, when you've got that much addressable space you should probably be doing paging with an LRU cache flush to an intermediate level of memory, which should save itself to disk in idle moments. This would take about 64 bits/block. One bit for "changed since read from/written to disk" and a bunch for "time of last access".
OTOH, I have my doubts that they actually have 64TB of RAM. I expect they just have a memory-mapped disk with an LRU cache, so you can address it as if there were 64TB or RAM, and in that case I can see cases where 64TB wouldn't be enough. As far as the OS was concerned, it would look like hardware with a HUGE RAM, even though it was actually a hardware paged disk. This would speed up both power up and power down, and would have other benefits. And for a system like that I can see that an addressable space of n*64TB might always end up being too small for certain classes of problems (for any fixed n).
But for an actual RAM vs. disk split I think that 64TB RAM should be enough for anyone. Even that would slow down power up and power down tremendously unless you don't care about losing what currently in RAM, and don't care that you're initializing the system with lots of random garbage.
MUCH better. At least you can figure out how to read the data files, even if nothing else. I've ended up with a bunch of files from dropped closed source projects that represent a huge amount of work and are totally worthless. Now I won't depend on any software that isn't FOSS.
A problem with FOSS software is that it often *does* have gating dependencies on the current developer. They aren't so much legal as informational, and sometimes interest, but they're there.
That said, this is much less severe with FOSS projects than with closed source projects, and the legal obstacles are absent, but that doesn't mean that there aren't any dependencies.
E.g. (a story from the 1970's). ... but they had all rights to the code, so that wasn't a problem. And the binary was isomorphic to the tuned assembler code, so that wasn't a problem. But *NOBODY* knew how it worked, what it did, or why it did it.
There was this shoe store that used software to do its taxes, and since computer time was expensive the compiler output was hand tuned for speed. At some point the original source code was lost, and all that was left was a binary program that did their taxes quickly and efficiently. The guy who wrote it had left for another job years ago. Then the IRS decided to audit them... Whoops!
If nobody can understand it, it *will* rot. This may be a disaster as far as the project is concerned. But if nobody really understands what it is doing, replacing it can be...problematic.
Before I switched of Linux my projects were continually depending on software that people and companies just dropped for one reason or another. Going out of business was one of the major reasons, but not the only one. I've still got files in various of them that I haven't got around to throwing away yet, but which are totally useless because there's no possible way to run them. TurboCAD is one that I just threw away today. I think they went out of business. Deluxe Music Construction Set I finished cleaning out earlier this year, they haven't worked since MacOS 7.5. The list is long and unhappy, full of work that was lost because of companies that stopped supporting their users. This hasn't happened once since I moved to Linux...though there are a couple of programs that only run in a virtualized Red Hat Linux 2.4 environment.
Things don't happen that way, but most "open source" projects don't end up with ownership problems so much as maintainer problems. It's quite rare that the legal inheritor of the copyright has either the desire or the ability to maintain it. With most open source licenses, certainly with GPL, the next maintainer retains the right to alter and release new versions of the code even without legal ownership of the copyright to the existing code. But capability is an entirely separate matter. There can also be problems with the right to use the name, but, again, that doesn't usually arise with death.
Actually I believe that for FOSS projects the two main problems in this area are the right to continue using the name and the capability to maintain the code (and interest in doing so). And that doesn't always wait for somebody to die to become a problem. I know Debian is frequently asking for someone to adopt a package that has lost its maintainer...and often the reason is loss of interest rather than death.
All that said, I know that in the Python community awhile ago (over a decade now?) there was significant concern about "What do we do if Guido is hit by a bus?", and the result was the current Python Software Foundation. Smaller projects, however, would probably find that solution expensive overkill.
The problem with that argument is that "intelligence" is an undefined term, unless you are arguing that IQ is intelligence, I which case I deny your original assertion.
What people have that has been our mainstay of success is culture. Human culture depends on language, though it clearly includes a lot of other things too. (Various monkeys and apes have been shown to have a primitive culture, but without language cultural transmission is labored and limited.) And unlike intelligence, what culture is at the basis is easily defined, it's the ability to pick up ways of doing things that are different from the ways your ancestors did. So some monkeys moving to the sea shore learn to crack open shell-fish. That's a primitive cultural advance. But notice that culture is the characteristic of a group rather than of an individual. One monkey learning to crack open a shell isn't a cultural advance. It only becomes such when it is transmitted to other monkeys.
You overstate the case. IQ is correlated with many useful skills. But it sure hasn't been shown to correlate with others.
For that matter, IQ itself is not a unitary measure. The tests measure different, and perhaps independent (but certainly not varying identically) capabilities. Sometimes the capabilities are correlated, but not identical, as, e.g., the ability to maintain focus and the ability to memorize. But as I know of no accepted separation of the capabilities measured by IQ, I doubt that there's even been significant research as to exactly WHAT is measured. But it does predict with reasonable accuracy your ability to do well on the average school test (without even attempting to measure motivation to do well over time).
But I don't trust the authority that you reference in your "argument from authority". Can you point to some actual evidence?
The problem is false positives. Otherwise it's trivial on any decent email system. But the false positive problem can be significant.
FWIW, the editors and reporters were always biased, back as far as we have records. But they used to be separately biased for each paper, and they used to worry about their reputation. Now the papers are owned by conglomerates who are mainly interested in the papers as pushing their particular message, whatever that it. And telecommunications allows micromanaging by owners at a distance.
Even the purely local papers around here have been bought up by chains, and local news has dropped off precipitously.
This trend was first (to my knowledge) commented on during the 1960's. Up until then most large cities had more than one locally owned and operated paper. Mergers had already happened more than once, though, (in the 1950s Honolulu had the Star-Bulletin, which was the merger of the Star and the Bulletin) so it may have been an on-going process.
Not for their own code. Developers are lousy at testing their own code, because they handle the errors that they think of, and won't make the ones they don't.
I suspect that developers are even poor choices for testing other developers code, because they're less likely to make ID10T errors.
What changed is something that didn't change. Apple has no interest in allowing non-Apple software to run, so their "security" application disallowed Audacity, despite that Audacity had been running under MacOS for a long time as a useful tool.
Perhaps Apple has a competing product. I don't know, they didn't when I was using them a bit over a decade ago. But my guess is that just just broke anything that wasn't too prominent that they weren't selling.
No, but it does prevent it from being very useful. For that technology to be useful in preventing accidents probably well over half the cars on the road would need to use it. Possibly over 3/4. So even if it were enforced it would take several years before it would do much good. And it mainly helps the vehicles avoiding someone else behaving illegally...and possibly mainly helps vehicles under automatic control. That would depend on implementation details.
I'd have to check, but I've probably got over 1000 passwords. Most of them are defunct, but sometimes I'm surprised that one I haven't used for a decade will still work. Some of them predate commercial access on the internet. Some of them are from when I was sysadmining a multi-user CP/M machine (with 16 processors!). But there are far too many to remember, especially for sites that I haven't accessed in a decade...and, surprise!, sometimes I need access again, and it's really nice if the old password still works. And you can't predict which ones.
You can't be sure of that. I'll admit that was my first take on the article, but it's not something I feel certain about.
That said, my second take on the article was "Oh, no. A *different* group of people will be in charge."
I didn't consider that it might be his actual opinion until my third take. And I see no way to choose between those possibilities (and there are probably a few more). I can't even estimate the degree of probability. But that he was stating his honest opinion is one plausible choice, even if it's not the most probable.
Remote location: Periodically mail a backup to your sister on the other coast.
It all depends on who you are, how rapidly your important data changes, and how much those changes are worth to you. I didn't address secrecy, which would, indeed, complicate matters.
OK, I'd be more willing to believe Apple than some others. That, however, is a long way from saying I'd believe their claim without actual proof for anything important.