How long before the disgruntled sysadmin replaces the disgruntled postal worker in the zeitgeist?
Only when disgruntled sysadmins start damaging meatspace.
I think a point is being missed here:
Damage to healthcare information IS damage to meatspace.
Whether it's actual health records or administrative databases (like appointment information, billing, supplies inventory and restock ordering, etc.), fouling the data can foul the medical treatment - causing incorrect treatment or delaying it when such a delay can be permanently damaging or fatal.
Prison time for planting a "logic bomb" intended to delete such data? HELL yes!
A cryonicist of my association once remarked that a far future civilization might find it useful to simulate people and events in its past - repeatedly with variations - for a number of reasons. A person experiencing such a simulated existence would not be able to tell the difference.
Now if the person had one real life, and a large number of simulated lives, in each of which he came to consider the question of whether he was simulated, the probability that the answer is "simulated" in any given case becomes close to 1. B-)
Philosophical, unprovable arguements are by nature not worth more than discussion, and can not by nature lead to any outcome...
However if the simulation is buggy it could lead to some useful special-cases in the (simulated) natural laws. "Special cases" that violate, say, conservation of momentum, or mass/energy, or a host of other stuff. Think of the technologies you could build on exploiting such bugs: Free power. Teleportation. Duplication of organized matter. Etc.
Such bugs might have a form that would expose the buggy simulation as a simulation. And a model that presumes "the universe may be a buggy simulation" might lead to searching for the bugs in different parts of the search space than one saying merely "the universe's laws may have some odd kinks".
The "insolubility" of the halting problem has to do with the impossibility of constructing a general algorithm that can examine another algorithm to determine whether it will halt.
Being able to do this is not a prerequisite for simulating the actual operation of the algorithm. So solving the halting problem is not necessary to run a universe-simulation containing such an algorithm. The simulator doesn't need to know in advance what the outcome will be to compute the steps going forward. It can just crunch away. If it stops, fine. If it hasn't stopped in a few million simulated years that says nothing about whether it will stop the next simulated nanosecond.
In fact that's usually the whole reason a simulation is done: To find out what happens via brute force computation rather than analysis. B-)
I completely disagree. The calculus on the simulation argument is surprisingly solid when you think about it (Bostrom, for instance, has some pretty good arguments for it). You say, "It's just not a useful avenue for speculation...."
Hear hear!
One interesting avenue for speculation: What if there are bugs in the simulation? Perhaps algorithmic, perhaps the equivalent of the "pentium floating-point bug" or the lack of denormals in the Weitek floating-point acceleration coprocessor chip that was used in the Sun4.
Bugs enable exploits. Exploits of a bug in the (simulation of the) physical laws of the universe would be the equivalent of magic: Do this incantation, get that result which violates the otherwise consistent physical laws in some radical way.
And if the bug is later fixed "the magic goes away".
Perhaps this has already happened. (What passes for the historical record a couple millennia or more back certainly seems at odds with a lot of science developed in the last 1500 years.)
And perhaps this might happen again.
(I have joked for decades that "The universe is a computer simulation and quantum numbers are as far as the machine takes the arithmetic." and had once done a plot sketch for a novel based on this concept - where a move of the simulation to a new machine with higher resolution changes the scale of quantization - somehow managing to avoid breaking the chemistry on which our lives depend but causing all the current semiconductor electronics to fail due to the change in bandgap and tunneling scale. This leads (along with the retooling of electronics) to the identification of the simulated nature of the universe and the successful hunt for simulation bugs that enable industrial magic and eventual communication with the operators of the simulation.)
In order to access the Watch Now service, I had to give Microsoft's DRM sniffing program access to all of the files on my hard drive. If the software found any non-Netflix video files, it would revoke my rights to the content and invalidate the DRM.
It will destroy your paid-for content if you have other content from another vendor? Sounds like an antitrust violation AND consumer fraud.
I'd bring this up with the FTC.
And I'd sue Netflix in small claims for everything you've paid so far. B-)
= = = =
Imagine if fifty, just fifty, people a day did that. They might think it's a movement. And that's what it is. The Alice's Restaurant Anti-DRM-masaccree movement. And you can join just by singing it, the next time it comes around on the guitar... B-)
Perhaps their target audience is portable media players, not executives or geeks needing to do random presentations. You know, practical, moneymaking considerations.
With scanning mirrors capable of +- atan(0.5) deflection and fixed lasers, one per color, the resolution is largely a matter of clocking rate for the laser driver / video interfaces and relative amplitude selection for the vertical and horizontal mirror drivers. The biggest cost for higher resolution would be to increase the scanning speed of one dimension (probably horizontal) to handle the extra lines required.
I don't really understand why they don't just make it capable of all the stock resolutions for computer monitors and also HDTV. B-(
... in real life tests apparently the GSM technology still outperforms analog in terms of range,...
Range, maybe yes. Though lower power it's also less susceptable to interference. So perhaps the range from the tower to the phone is comparable, despite the factor of 6 power difference.
But coverage? Hell no!
To convert equivalent range to equivalent coverage you have to convert all the cell sites to digital. This has NOT happened.
Analog cell sites cover virtually all of the central 48 states. GSM and other digital services are concentrated in urban areas and the most heavily-settled routes between them. (Look up the coverage maps of the various carriers to see this.)
Post telecom-crash (and the resulting reorganizations and mergers) the telecoms have concentrated on the high-density markets of the cities and the major commute routes, ignoring the rural areas and lightly cross-country routes (exactly the places where you NEED your cellphone or OnStar to work if you get into trouble).
The digital networks had only been lightly deployed before the crash. After it the telecoms apparently treated the individual cell towers as profit centers. They upgraded those where the bandwidth crunch required it or the added channels could bring in added revenue. But the lightly-used cells in rural areas - which didn't have enough traffic to pay for themselves - were not upgraded.
This completely ignores the main point of cell phones: Maintaining communication when traveling or working away from the fixed infrastructure. Yes, the lightly-used cells don't have enough traffic to pay off directly - but by filling in the holes they increase the value of the subscription to the overall network. That's how they earn their keep.
The result is that there are large chunks of the nation where cellphone service will "go dark" when/if the analog cells are shut down.
(This happens to include my country house, by the way. It's in one of AT&T's "last cell"s on the periphery of their old AMPS/TDMA network and, though they've been charging extra to try to force me into changing to GSM, as of the last time I checked - which I do every couple months - they still hadn't upgraded it. My expectation is that the service there will just die when/if they pull the plug on AMPS/TDMA.)
It's also a secondary nuclear fuel: The capture reaction n + Li6 -> H3 + He4 has leftover energy. It's not a self-sustaining primary reaction, but it is very exothermic (4.8 Mev).
But in addition to shielding and secondary energy production, it acts as a transducer, converting the kinetic energy of both the neutrons and the n + Li6 -> H3 + He4 reaction products to heat.
Some friends of mine were involved in a private launch company back in the mid '80s. (At their request it will remain unnamed.)
One of their major problems was obtaining components for avionics and for handling cryogenic liquids. These were made mainly by the companies who contracted to NASA for various parts of the (very lucrative) shuttle program.
One of their contacts told them that a NASA administrator had let them know that if they supplied any parts to a private rocket company they wouldn't be supplying any more for the shuttle.
The company thus had to make do without components that had been developed with tax money, and (on their shoestring budget) develop their own from scratch or convert stuff intended for other purposes - none of them space-rated.
They did some amazing stuff on that shoestring. But it was the failure of one of those re-purposed parts that ended up trashing their effort and running them out of money.
Now NASA was SUPPOSED to be ENCOURAGING the private development of space capability, as they had air flight. But the government space programs had put them in a position where doing so would undercut the funding for their own programs. So it was in their interest to keep the suppliers on a short leash and kill off any company trying to assemble and operate their own craft.
Pulling the plug on NASA as the government-run space transportation company (and boondoggle) would, IMHO, not just open up the field to private companies, but is a necessary step in getting to affordable private space travel in what remains of my lifetime.
Which is not necessarily to say kill it off completely. But putting it out of the transportation business and back to R&D, with private enterprise actually running the spacelines, seems to me to be a necessary minimum for turning space exploration from a government-funded boondoggle (ala Columbus) to an ongoing enterprise (ala private cargo and passenger ships crossing the Atlantic and Pacific ocean).
Have you ever wondered why we haven't encountered intelligent life forms other than ourselves? An advanced race with regular slower-than-light starships would be able to colonize an entire galaxy within a few million years (barely an instant on a geological timescale).
My preferred answer to the Fermi paradox is a corollary of that:
The thought of one such mentally ill leader having access to the largest stock of nuclear weapons in the world is... disturbing.
It's supposed to be.
The MAD doctrine deters nuclear war by threatening a retaliation that would likely bring down civilization and possibly end the human race and much of life on Earth.
For it to work, US presidents have to put on a show, looking crazy enough that they'd actually do it - but sane enough that the won't shoot first and can be reasoned with on issues that otherwise would have been "solved" by the outcome of a war. (IMHO it's likely the term "Mutually Assured Destruction" was chosen at least partly for the acronym, to help put on this show. Psych warfare was pretty well developed by the start of the Cold War.)
MAD is pretty terrifying. But it reversed the ongoing escalation of wars right after the bombs were proven to work under battle conditions (and two fried cities were substituted for the years of war that had been expected to be necessary to end the Japan part of WWII). It's been over half a century and no nukes have been used in war since those two.
... that story... starts with the premise that returning soldiers would essentially take over the world and everything would be wonderful thereafter. History has shown quite clearly that every time this occurs things go badly.
Capsule summary: The returning WWII veterans came back to a totally corrupt county political machine - which oppressed them with fake traffic and other tickets with large fines. Having just fought against oppressive dictatorships in Europe they decided to fix it. First they ran their own reform slate. But when the political machine set up to jigger the ballot counting they thew them out by force of arms (a literal shooting war to recover the ballot boxes) and then counted the ballots in public.
I call bullshit. Telco switches record the calls to CDR (call data records) files before sending the data on to the billing systems.
As it was explained to me the CDRs were being recorded in the mainframe live, not buffered in the switches (except for network streaming).
I was on the OS side of things, not the apps, and told this by another OS type. So it's possible you're right and I'm propagating a myth.
But I do note that storing them in the switches and later uploading them to the mainframe can increase the vulnerability: They're subject to corruption on the switches before uploading, and the switches have a higher failure rate. (They just recover fast, and calls in progress don't get dropped, since the switch brains are involved in CHANGING the connections, not maintaining them.) Also, a two-step process also increases the chance of the converse failure: billing for the same call twice. That can cost them a BUNCH more than giving away a call. So I can easily believe they'd have transitioned to a system where the CDRs were hot-potatoed to the mainframe, even if it was done in a way that risked dropping them through the cracks if the mainframe was belly-up for 25 min or so.
= = = =
Which doesn't say a thing about the transaction support apps and databases on the brokerage mainframes, does it? You need a host somewhere, and last I heard the distributed update problem had not been solved. Which means distributing the operation across a number of low-reliability machines doesn't improve the failure rate to something above that of the machines: It just partitions the accounts into groups that are each subject to the same bad failure rate and then multiplies the chances that some customers get bit. It also opens the brokerage to charges of favoritism if some customer accounts are frozen due to machine failure and others have their transactions processed.
Better to put them all on one giant, heavily backed up, hyper-reliable internally-redundant machine and take them all down at once - but almost never.
"If there's one thing the market hates, it's crashes."
No fooling.
I used to work on Amdhall's unix for their mainframes. Among other things it was used by brokerages to support trading and all the Baby Bells to support data collection for billing.
If a baby bell's billing system went down all the phone calls dialed, started, or completed while it was down were free. This made downtime cost something like $4 million / hour.
Brokerage support going down cost far more.
So imagine a trading system going down (equivalent to all the brokerages going down at the same time...)
Needless to say, much of the point of mainframes is to keep this from ever happening.
So the hardware is built so it performs the correct computation despite component failures, radiation-flipped bits, or on-the-fly hardware changes (adding/deleting/resizing peripherals, CPUs memory, switching out failing components), etc. And the software is built to similar standards.
This can cause problems. Like sizing event counters to stand uptime measured in decades. Or getting non-critical patches installed. (I recall a minor patch to a driver, too small to rate forcing a couple million bux worth of reboot, that had been installed on all the customers' machines to go live at the next reboot. Two years later (last I heard) they were still supporting the bug because some systems hadn't rebooted yet...)
USB is great for small devices -- thumb drives, mice, webcams and such --
I beg to differ.
I wired my new (2001) house with a couple runs of cat-5E from the comp room to each corner, expecting inexpensive single-chip cameras to become available to be used for security cams.
Well they became available, all right. But all the cheap ones were USB, not Ethernet, and USB has a distance limit suitable for a workstation's desk rather than a house.
If it was just an electrical issue I could have built suitable level-shifters, baluns, etc. to extend the reach. But the limit is apparently timing of the poll/response rather than just signal integrity, so I'm hosed.
In fairness:
- There are ethernet webcams - but they're not cheap.
- There are active USB extenders to bridge USB to cat-5/5e and back for long runs - but they cost more than the cameras and I'd need one for each corner.
- I haven't had the spare round-TUITs to look into whether the timing issue is programmable in the driver or hardwired into the chips...
It doesn't matter if we evolve, because we change the environment around us as opposed to adapting to it.
Just for starters, infectious diseases will continue to drive human evolution. The bugs evolve immunity to technological helpers (such as antibiotics) and our large population and easy travel means lots of potential for plagues. And bugs swap protective genes around - even between drastically different species (such as e-coli and salmonella) and collect them (on plasmid "charm bracelets"). In reasonably short order they can be expected to evolve to the point that some are resistant to pretty much every chemical compound our own cells can survive. As a result our immune systems are still undergoing evolutionary pressure - and evolving at an accelerated rate compared to the times when humans were scattered bands of hunter/gatherers.
Example: Bubonic Plague caused a strong selective pressure for a previously rare form of one receptor in the immune system's communication net. Those with two copies of the common form died off in droves as the bugs used it to ride into the lymph system and undermine active immunity. But those with one each of that and a particular variant got very sick but survived, while those with two copies of the variant didn't have enough of an infection to be identified as having the disease.
Nowdays plague is mildly endemic in the US southwest and still very susceptible to antibiotics. But evolutionary pressures are at work again - on the same two forms of the gene and working in the same direction. HIV targets the same same receptor and misses the same variant, so people with two copies of the variant don't develop AIDS despite exposure to HIV. (Which is what led to the immunity to Plague being discovered.)
That one won't fit on a pistol grip - or even a tank-based mobile platform. What kind of raygun is that?
Fixed installation beam weapons might be OK for shooting down incoming stuff. But you need to go on the offense if you want to finish the war with a win.
Do you know if they can somehow take a picture of the atoms while it is at a freeze frame? or will the photon cause the atoms to move again?
Yes and yes.
The burst of x-ray/gamma-ray photons will no doubt blast the molecules being observed into their component nuclei and electrons, which will scatter like billiard balls during the "break".
But it's a very SHORT flash. You get your "picture" of where they were when the flash hit, by the scattering of the incident massless photons, before the particles with rest-mass have a chance to go somewhere else.
One significant difference between Linux on a PC and Linux on a mobile is that it is illegal to expose the core baseband processor architecture to open software, because that would make it trivial to create network destroying devices.
I've heard that preventing such tampering was the reason much of the firmware is closed, but this is the first time I've seen the assertion of an actual law or regulation requiring the firmware internals be kept secret.
Can you (or someone) give us a pointer to the law and/or regulation in question?
Last year I attended a conference. woot.com was one of the sponsors. At the closing ceremony they passed out some swag from that company to the attendees - in a container boldly labeled with the company logo.
When I got home and she saw it my wife was ROTFL.
She's one of the couple hundred remaining speakers of the west-coast American Indian trade jargon. And it seems that, in that language, the word for the male organ is (approximately) WOOT-`let.
To use a computer analogy (which seems appropriate in this case):
IMHO they're only testing the visual preprocessor speed, in a situation where the expected effect is bringing online more processing power and/or modifying the task scheduling and priorities for better response time on normally background tasks that have become life-critical.
Skimming TFA I was struck by how most of the arguments and propaganda techniques could be applied - even moreso - to talking an IT department into replacing Microsoft OSes and apps with FOSS solutions.
Slashdot Mirror
How long before the disgruntled sysadmin replaces the disgruntled postal worker in the zeitgeist?
Only when disgruntled sysadmins start damaging meatspace.
I think a point is being missed here:
Damage to healthcare information IS damage to meatspace.
Whether it's actual health records or administrative databases (like appointment information, billing, supplies inventory and restock ordering, etc.), fouling the data can foul the medical treatment - causing incorrect treatment or delaying it when such a delay can be permanently damaging or fatal.
Prison time for planting a "logic bomb" intended to delete such data? HELL yes!
In fact that's not just the default, but they don't even offer a driver as optional equipment.
A cryonicist of my association once remarked that a far future civilization might find it useful to simulate people and events in its past - repeatedly with variations - for a number of reasons. A person experiencing such a simulated existence would not be able to tell the difference.
Now if the person had one real life, and a large number of simulated lives, in each of which he came to consider the question of whether he was simulated, the probability that the answer is "simulated" in any given case becomes close to 1. B-)
Philosophical, unprovable arguements are by nature not worth more than discussion, and can not by nature lead to any outcome ...
However if the simulation is buggy it could lead to some useful special-cases in the (simulated) natural laws. "Special cases" that violate, say, conservation of momentum, or mass/energy, or a host of other stuff. Think of the technologies you could build on exploiting such bugs: Free power. Teleportation. Duplication of organized matter. Etc.
Such bugs might have a form that would expose the buggy simulation as a simulation. And a model that presumes "the universe may be a buggy simulation" might lead to searching for the bugs in different parts of the search space than one saying merely "the universe's laws may have some odd kinks".
The "insolubility" of the halting problem has to do with the impossibility of constructing a general algorithm that can examine another algorithm to determine whether it will halt.
Being able to do this is not a prerequisite for simulating the actual operation of the algorithm. So solving the halting problem is not necessary to run a universe-simulation containing such an algorithm. The simulator doesn't need to know in advance what the outcome will be to compute the steps going forward. It can just crunch away. If it stops, fine. If it hasn't stopped in a few million simulated years that says nothing about whether it will stop the next simulated nanosecond.
In fact that's usually the whole reason a simulation is done: To find out what happens via brute force computation rather than analysis. B-)
I completely disagree. The calculus on the simulation argument is surprisingly solid when you think about it (Bostrom, for instance, has some pretty good arguments for it). You say, "It's just not a useful avenue for speculation. ..."
Hear hear!
One interesting avenue for speculation: What if there are bugs in the simulation? Perhaps algorithmic, perhaps the equivalent of the "pentium floating-point bug" or the lack of denormals in the Weitek floating-point acceleration coprocessor chip that was used in the Sun4.
Bugs enable exploits. Exploits of a bug in the (simulation of the) physical laws of the universe would be the equivalent of magic: Do this incantation, get that result which violates the otherwise consistent physical laws in some radical way.
And if the bug is later fixed "the magic goes away".
Perhaps this has already happened. (What passes for the historical record a couple millennia or more back certainly seems at odds with a lot of science developed in the last 1500 years.)
And perhaps this might happen again.
(I have joked for decades that "The universe is a computer simulation and quantum numbers are as far as the machine takes the arithmetic." and had once done a plot sketch for a novel based on this concept - where a move of the simulation to a new machine with higher resolution changes the scale of quantization - somehow managing to avoid breaking the chemistry on which our lives depend but causing all the current semiconductor electronics to fail due to the change in bandgap and tunneling scale. This leads (along with the retooling of electronics) to the identification of the simulated nature of the universe and the successful hunt for simulation bugs that enable industrial magic and eventual communication with the operators of the simulation.)
In order to access the Watch Now service, I had to give Microsoft's DRM sniffing program access to all of the files on my hard drive. If the software found any non-Netflix video files, it would revoke my rights to the content and invalidate the DRM.
It will destroy your paid-for content if you have other content from another vendor? Sounds like an antitrust violation AND consumer fraud.
I'd bring this up with the FTC.
And I'd sue Netflix in small claims for everything you've paid so far. B-)
= = = =
Imagine if fifty, just fifty, people a day did that. They might think it's a movement. And that's what it is. The Alice's Restaurant Anti-DRM-masaccree movement. And you can join just by singing it, the next time it comes around on the guitar... B-)
Perhaps their target audience is portable media players, not executives or geeks needing to do random presentations.
You know, practical, moneymaking considerations.
With scanning mirrors capable of +- atan(0.5) deflection and fixed lasers, one per color, the resolution is largely a matter of clocking rate for the laser driver / video interfaces and relative amplitude selection for the vertical and horizontal mirror drivers. The biggest cost for higher resolution would be to increase the scanning speed of one dimension (probably horizontal) to handle the extra lines required.
I don't really understand why they don't just make it capable of all the stock resolutions for computer monitors and also HDTV. B-(
... in real life tests apparently the GSM technology still outperforms analog in terms of range, ...
Range, maybe yes. Though lower power it's also less susceptable to interference. So perhaps the range from the tower to the phone is comparable, despite the factor of 6 power difference.
But coverage? Hell no!
To convert equivalent range to equivalent coverage you have to convert all the cell sites to digital. This has NOT happened.
Analog cell sites cover virtually all of the central 48 states. GSM and other digital services are concentrated in urban areas and the most heavily-settled routes between them. (Look up the coverage maps of the various carriers to see this.)
Post telecom-crash (and the resulting reorganizations and mergers) the telecoms have concentrated on the high-density markets of the cities and the major commute routes, ignoring the rural areas and lightly cross-country routes (exactly the places where you NEED your cellphone or OnStar to work if you get into trouble).
The digital networks had only been lightly deployed before the crash. After it the telecoms apparently treated the individual cell towers as profit centers. They upgraded those where the bandwidth crunch required it or the added channels could bring in added revenue. But the lightly-used cells in rural areas - which didn't have enough traffic to pay for themselves - were not upgraded.
This completely ignores the main point of cell phones: Maintaining communication when traveling or working away from the fixed infrastructure. Yes, the lightly-used cells don't have enough traffic to pay off directly - but by filling in the holes they increase the value of the subscription to the overall network. That's how they earn their keep.
The result is that there are large chunks of the nation where cellphone service will "go dark" when/if the analog cells are shut down.
(This happens to include my country house, by the way. It's in one of AT&T's "last cell"s on the periphery of their old AMPS/TDMA network and, though they've been charging extra to try to force me into changing to GSM, as of the last time I checked - which I do every couple months - they still hadn't upgraded it. My expectation is that the service there will just die when/if they pull the plug on AMPS/TDMA.)
It's also a secondary nuclear fuel: The capture reaction n + Li6 -> H3 + He4 has leftover energy. It's not a self-sustaining primary reaction, but it is very exothermic (4.8 Mev).
But in addition to shielding and secondary energy production, it acts as a transducer, converting the kinetic energy of both the neutrons and the n + Li6 -> H3 + He4 reaction products to heat.
Sweet system.
And if the other athletes were taking performance-enhancing drugs the one that got caught might get an overdose.
Some friends of mine were involved in a private launch company back in the mid '80s. (At their request it will remain unnamed.)
One of their major problems was obtaining components for avionics and for handling cryogenic liquids. These were made mainly by the companies who contracted to NASA for various parts of the (very lucrative) shuttle program.
One of their contacts told them that a NASA administrator had let them know that if they supplied any parts to a private rocket company they wouldn't be supplying any more for the shuttle.
The company thus had to make do without components that had been developed with tax money, and (on their shoestring budget) develop their own from scratch or convert stuff intended for other purposes - none of them space-rated.
They did some amazing stuff on that shoestring. But it was the failure of one of those re-purposed parts that ended up trashing their effort and running them out of money.
Now NASA was SUPPOSED to be ENCOURAGING the private development of space capability, as they had air flight. But the government space programs had put them in a position where doing so would undercut the funding for their own programs. So it was in their interest to keep the suppliers on a short leash and kill off any company trying to assemble and operate their own craft.
Pulling the plug on NASA as the government-run space transportation company (and boondoggle) would, IMHO, not just open up the field to private companies, but is a necessary step in getting to affordable private space travel in what remains of my lifetime.
Which is not necessarily to say kill it off completely. But putting it out of the transportation business and back to R&D, with private enterprise actually running the spacelines, seems to me to be a necessary minimum for turning space exploration from a government-funded boondoggle (ala Columbus) to an ongoing enterprise (ala private cargo and passenger ships crossing the Atlantic and Pacific ocean).
Have you ever wondered why we haven't encountered intelligent life forms other than ourselves? An advanced race with regular slower-than-light starships would be able to colonize an entire galaxy within a few million years (barely an instant on a geological timescale).
My preferred answer to the Fermi paradox is a corollary of that:
Somebody had to be first. Looks like it's us.
(For this galaxy at least.)
The thought of one such mentally ill leader having access to the largest stock of nuclear weapons in the world is... disturbing.
It's supposed to be.
The MAD doctrine deters nuclear war by threatening a retaliation that would likely bring down civilization and possibly end the human race and much of life on Earth.
For it to work, US presidents have to put on a show, looking crazy enough that they'd actually do it - but sane enough that the won't shoot first and can be reasoned with on issues that otherwise would have been "solved" by the outcome of a war. (IMHO it's likely the term "Mutually Assured Destruction" was chosen at least partly for the acronym, to help put on this show. Psych warfare was pretty well developed by the start of the Cold War.)
MAD is pretty terrifying. But it reversed the ongoing escalation of wars right after the bombs were proven to work under battle conditions (and two fried cities were substituted for the years of war that had been expected to be necessary to end the Japan part of WWII). It's been over half a century and no nukes have been used in war since those two.
For a counterexample see the Battle of Athens.
Capsule summary: The returning WWII veterans came back to a totally corrupt county political machine - which oppressed them with fake traffic and other tickets with large fines. Having just fought against oppressive dictatorships in Europe they decided to fix it. First they ran their own reform slate. But when the political machine set up to jigger the ballot counting they thew them out by force of arms (a literal shooting war to recover the ballot boxes) and then counted the ballots in public.
I call bullshit. Telco switches record the calls to CDR (call data records) files before sending the data on to the billing systems.
As it was explained to me the CDRs were being recorded in the mainframe live, not buffered in the switches (except for network streaming).
I was on the OS side of things, not the apps, and told this by another OS type. So it's possible you're right and I'm propagating a myth.
But I do note that storing them in the switches and later uploading them to the mainframe can increase the vulnerability: They're subject to corruption on the switches before uploading, and the switches have a higher failure rate. (They just recover fast, and calls in progress don't get dropped, since the switch brains are involved in CHANGING the connections, not maintaining them.) Also, a two-step process also increases the chance of the converse failure: billing for the same call twice. That can cost them a BUNCH more than giving away a call. So I can easily believe they'd have transitioned to a system where the CDRs were hot-potatoed to the mainframe, even if it was done in a way that risked dropping them through the cracks if the mainframe was belly-up for 25 min or so.
= = = =
Which doesn't say a thing about the transaction support apps and databases on the brokerage mainframes, does it? You need a host somewhere, and last I heard the distributed update problem had not been solved. Which means distributing the operation across a number of low-reliability machines doesn't improve the failure rate to something above that of the machines: It just partitions the accounts into groups that are each subject to the same bad failure rate and then multiplies the chances that some customers get bit. It also opens the brokerage to charges of favoritism if some customer accounts are frozen due to machine failure and others have their transactions processed.
Better to put them all on one giant, heavily backed up, hyper-reliable internally-redundant machine and take them all down at once - but almost never.
"If there's one thing the market hates, it's crashes."
No fooling.
I used to work on Amdhall's unix for their mainframes. Among other things it was used by brokerages to support trading and all the Baby Bells to support data collection for billing.
If a baby bell's billing system went down all the phone calls dialed, started, or completed while it was down were free. This made downtime cost something like $4 million / hour.
Brokerage support going down cost far more.
So imagine a trading system going down (equivalent to all the brokerages going down at the same time...)
Needless to say, much of the point of mainframes is to keep this from ever happening.
So the hardware is built so it performs the correct computation despite component failures, radiation-flipped bits, or on-the-fly hardware changes (adding/deleting/resizing peripherals, CPUs memory, switching out failing components), etc. And the software is built to similar standards.
This can cause problems. Like sizing event counters to stand uptime measured in decades. Or getting non-critical patches installed. (I recall a minor patch to a driver, too small to rate forcing a couple million bux worth of reboot, that had been installed on all the customers' machines to go live at the next reboot. Two years later (last I heard) they were still supporting the bug because some systems hadn't rebooted yet...)
USB is great for small devices -- thumb drives, mice, webcams and such --
I beg to differ.
I wired my new (2001) house with a couple runs of cat-5E from the comp room to each corner, expecting inexpensive single-chip cameras to become available to be used for security cams.
Well they became available, all right. But all the cheap ones were USB, not Ethernet, and USB has a distance limit suitable for a workstation's desk rather than a house.
If it was just an electrical issue I could have built suitable level-shifters, baluns, etc. to extend the reach. But the limit is apparently timing of the poll/response rather than just signal integrity, so I'm hosed.
In fairness:
- There are ethernet webcams - but they're not cheap.
- There are active USB extenders to bridge USB to cat-5/5e and back for long runs - but they cost more than the cameras and I'd need one for each corner.
- I haven't had the spare round-TUITs to look into whether the timing issue is programmable in the driver or hardwired into the chips...
It doesn't matter if we evolve, because we change the environment around us as opposed to adapting to it.
Just for starters, infectious diseases will continue to drive human evolution. The bugs evolve immunity to technological helpers (such as antibiotics) and our large population and easy travel means lots of potential for plagues. And bugs swap protective genes around - even between drastically different species (such as e-coli and salmonella) and collect them (on plasmid "charm bracelets"). In reasonably short order they can be expected to evolve to the point that some are resistant to pretty much every chemical compound our own cells can survive. As a result our immune systems are still undergoing evolutionary pressure - and evolving at an accelerated rate compared to the times when humans were scattered bands of hunter/gatherers.
Example: Bubonic Plague caused a strong selective pressure for a previously rare form of one receptor in the immune system's communication net. Those with two copies of the common form died off in droves as the bugs used it to ride into the lymph system and undermine active immunity. But those with one each of that and a particular variant got very sick but survived, while those with two copies of the variant didn't have enough of an infection to be identified as having the disease.
Nowdays plague is mildly endemic in the US southwest and still very susceptible to antibiotics. But evolutionary pressures are at work again - on the same two forms of the gene and working in the same direction. HIV targets the same same receptor and misses the same variant, so people with two copies of the variant don't develop AIDS despite exposure to HIV. (Which is what led to the immunity to Plague being discovered.)
I've got a real one in the next building!
That one won't fit on a pistol grip - or even a tank-based mobile platform. What kind of raygun is that?
Fixed installation beam weapons might be OK for shooting down incoming stuff. But you need to go on the offense if you want to finish the war with a win.
B-)
Do you know if they can somehow take a picture of the atoms while it is at a freeze frame? or will the photon cause the atoms to move again?
Yes and yes.
The burst of x-ray/gamma-ray photons will no doubt blast the molecules being observed into their component nuclei and electrons, which will scatter like billiard balls during the "break".
But it's a very SHORT flash. You get your "picture" of where they were when the flash hit, by the scattering of the incident massless photons, before the particles with rest-mass have a chance to go somewhere else.
One significant difference between Linux on a PC and Linux on a mobile is that it is illegal to expose the core baseband processor architecture to open software, because that would make it trivial to create network destroying devices.
I've heard that preventing such tampering was the reason much of the firmware is closed, but this is the first time I've seen the assertion of an actual law or regulation requiring the firmware internals be kept secret.
Can you (or someone) give us a pointer to the law and/or regulation in question?
Last year I attended a conference. woot.com was one of the sponsors. At the closing ceremony they passed out some swag from that company to the attendees - in a container boldly labeled with the company logo.
When I got home and she saw it my wife was ROTFL.
She's one of the couple hundred remaining speakers of the west-coast American Indian trade jargon. And it seems that, in that language, the word for the male organ is (approximately) WOOT-`let.
Shades of 18th century viagra ads.
To use a computer analogy (which seems appropriate in this case):
IMHO they're only testing the visual preprocessor speed, in a situation where the expected effect is bringing online more processing power and/or modifying the task scheduling and priorities for better response time on normally background tasks that have become life-critical.
Skimming TFA I was struck by how most of the arguments and propaganda techniques could be applied - even moreso - to talking an IT department into replacing Microsoft OSes and apps with FOSS solutions.
Thanks, Microsoft!