When I was coming back from San Fran a couple weeks ago (after Macworld) there was a multi-million dollar NASA camera that'll be on the next space ship going through security at the same time I was. You'd think they'd check those things before hand.
And let the baggage handlers throw it around and/or put it on the next plane to Botswanaland by mistake? Hardly!
... why not Luxembourg? With a population of only 465,000 we'd made a majority of the population and be able to form a governmenmt.
For starters because Luxembourg won't let you move in and get citizenship all that easily.
And the population is rich enough to enforce their will: Pretty much every adult is an officer of several international corporations, at some serious pay each. This is because Luxembourg's laws make it advantageous to headquarter there, but require at least one citizen as a major officer.
Besides: Taking over by settling creates serious (sometimes deadly) opposition from those already there who FORMERLY ran their own government.
If you want to create a settlement where you can run your own government up to a significant level, try Oregon. If they remain true to their history, once you've established a significant colony of like-minded people, if you have a beef with the rest of your county they'll split it and give you your own county composed of you and your like-minded settlers. Then you can elect your own supervisors and sheriff, tax each other, maintain the roads your way, etc.
(Which is what makes the Ragneeshi's attempted takeover of Wasco county - by food-poisoning a salad bar at a local restaurant shortly before the election - such a stupid move: The state had already offered them a county composed of their own settlement and the roads to it.)
But software bloat increases faster than single-thread performance, thus making software run slower.
But if your application can run in its own thread the ongoing software bloat of the underlying OS won't affect its internal performance. (And the OS vendor may bloat it, but count on him to use enough multithreading to keep it out from underfoot - even if it chews up most of the processors in the multicore cluster.)
As long as you or your software vendor doesn't add inline bloating with the upgrades the application's performance will track that of the individual processors in the cluster - which will continue improving, though more gradually.
(If your vendor can't keep his engineers' hands off the critical code's performance his product is toast. Start looking for a replacement.)
What a good way to extract atmospheric moisture. And drifting pesticides, emmision fumes, particulate pollution blowing from every industrial site in the area... its like a pollutent-concentrator!
This is far superior to the "miss-to-kill" technology they were employing in previous models
Joke all you want, but that's what we've actually been doing. Didn't anybody ever tell you that close counts with hand grenades?
And nuclear weapons.
Which, unfortunately, was exactly the "big boom" "kinda close" that had been contemplated in some previous ABM designs.
After all, if it has blossomed, MIRV style, into a cloud of decoys and multiple real nuclear bombs on independent trajectories, spread out by quite a bit by the time your missile gets there, you need a BIG boom to disable all of them that matter. And without an atmosphere to carry a shockwave it helps if you can irradiate with heat, gammas, and neutrons, and can vaporize the whole antimissile and hit the targets with the vapor.
Problem with that is you're setting off your OWN nukes above your OWN targets - and high enough above the atmosphere to do a major electromagnetic pulse when the gamma burst makes a sheet of electrons the size of a continent jump upward by a few miles. (With defenses like that who needs an enemy missile with a real warhead? Other than to provoke the defense.)
So, yes, "hit to kill" is a BIG improvement over the "miss-to-kill technology they were using in previous models". (Assuming you have at least as many anti-missiles as they sent warheads and convincing decoys.)
And I'm sure that with a little web surfing we could come up with enough other black hats to fill all the MP slots.
Not that we need to, of course - and they'd be the wrong names, anyhow. Think what an opportunity this is for the trillionaire-"businessmen" who deploy the malware and run the phising scams and spam botnets.
How cool. So Microsoft is going to use their "genuine (dis)advantage" tools to get serious about collecting their license fees.
So what does this do to the "total cost of ownership" of windows versus open source solutions?
How much of those calculations especially at the PHB level - are done assuming either that all their installations are paid for (and nobody installed any extras or forged their identification) or that they can get away with extras - and in either case didn't factor in being audited? (That's a BIG cost even (especially) if it turns out you're squeaky-clean.)
Perhaps this will create additional incentives to switch.
I'm presuming that, if the machine doesn't do everything the customers want (ESPECIALLY VoIP over whatever network connection combined with being able to access home router and commercial hotspots), it will be hacked almost as soon as it's out there. B-)
(Actually I'm presuming it will be hacked almost as soon as it's out there, regardless - especially if it won't allow non-Apple software add-ons. B-) But being unable to use VoIP over your LAN will put the pressure on with a specific target from day zero.)... what about feature support on other networks ? For example, the graphical voicemail.
(I see we agree on what's the most important phone-like feature after VoIP roaming.)
That's already in place with plenty of phone systems. (For instance, our PBX does it at work.) You bring up the interface on a web browser, are presented with your voicemails, and when you select one it plays through your phone.
Such prior art should keep this from being one of their 200+ patents - or at least should keep any patent claims on the way they did it from simultaneously blocking such a web/phone hybrid and standing up to a challenge. (If it's already patented in the PBX context it may even expose Apple to a suit for doing it on the iPhone.)
If Verizon is smart it will try to head this off by dropping prices to where they're just covering network connectivity rather than subsidizing the non-existent "free" crippled phone.
Oops! Meant "Cingular".
And if Cingular/ATT and Verizon are BOTH clueless and leave the plans at regular cellphone rates, watch for users to start migrating to WiFi hotspot operators and WISPs.
(Watch for that anyhow, once people start hacking. B-) Even if Apple doesn't support it - or doesn't support it well - nobody in their right mind with a VoIP account and a WiFi AP at home is going to chew up cell minutes when at home when they can make the iPhone use their broadband and existing accounts, getting hax from their VoIP providers or third parties to make it work well.)
Since apple's phone isn't subsidized by the cell companies they have no leverage on how its firmware operates. This means:
- It's not locked into a carrier. You can switch in a heartbeat and/or put more than one plan on it.
- It's not locked OUT of using other systems than cellphone - like VoIP over WiFi or WiMax.
This means that the cellphone carriers are not just in competition with other cellphone and cellphone/data carriers. They're also in competition with Wireless ISPs (WISPs).
Even between the cellphone carriers the lack of the lock-in means they're in straight competition on price of service. (They had to do the lockin and early termination fee to pay for the handset tie-in.)
This will produce significant downward market pressure on cellphone companies.
Market forces don't produce a heavy drive toward marginal cost until there are at least THREE competing providers of the good or service. (For two the strategy is to track each other's prices and split the market about 50/50. For three or more the incentive is for the little guy to try to undercut the two biggest players and steal market from the pair - and for them to retaliate using their economy of scale.)
While there are several cellphone players now there are typically only two dominant players in most markets. The original bandwidth licensing regime was set up for "competition of two" (the incumbent phone company and ONE competitor) and the early rollout gave two players incumbent status in most markets. They then had an analog of the government-subsidized copper buildout of the wireline phone companies that gave them an advantage in coverage as the upstarts started up - leading to sickly third players and rounds of consolidation.
This device lets WISPs with significant coverage play in the cellphone space - and use their bandwidth cost advantage to become major players. If Verizon is smart it will try to head this off by dropping prices to where they're just covering network connectivity rather than subsidizing the non-existent "free" crippled phone.
Strongarm states that don't conform their own laws to national 'standard' election laws by threatening Federal funding sanctions. That's been the common practice for quite a while, as far as I understand it.
They did that with speed limits and highway funds, and some states resisted or ignored it. (The states collect much of the federal money - such as gas taxes - and they can turn the tables and refuse to turn it in.)
Interference in election procedures is a MUCH bigger hotbutton issue for state officials than traffic laws. It's fundamental to the governmental underpinning and their own selection. States with honest politicians will resist on constitutional issues. (States with political machines in place will resist to maintain their lock on elections and claim to be resisting on constitutional issues. B-) )
Even if the change were benign it might affect the outcome. The people in charge in the states were elected by the old procedures and have an interest in not sinking the boat that ferries them into their offices every election cycle.
The decentralized form of elections might have made sense for the age of horse coaches but in the age of internet it is not too tough to have thge same standards everywhere in the US.
The US is not a country. It is a confederation of countries ("the several States") united by a set of treaties (ratifications of the Constitution or treaties of incorporation, depending on when/how they joined up).
Each of these treaty-bound but independent countries has its own election procedures, and they did not give up their right to control their own election procedures when they joined. The constitution doesn't authorize the Fed to control election procedures in general (though there are things it DOES authorize that let the Federal courts to intervene in details occasionally).
Part of the reason that the president is chosen by the electoral college, rather than the popular vote, is to provide a firewall between election corruption in particular states and the selection of the president (the one case where the population of the whole country votes on a common federal-level decision).
With each state selecting only its own electors, a corrupt voting system can only affect the electors from that state. A state can do no more than throw all its electors to one candidate (which most of them do) and even in close races most of the states aren't in a position to swing the election. (Political machines are run by the party in power, which generally has a significant majority among the overall population of the state. So the state's electors will generally be chosen their way and there's nothing to be gained by election fraud beyond the point that insures it.) If things were close (typically when the machine has lost the population but isn't out yet due to election corruption or jerrymandering) and they swung their own state's election through vote fraud, you only have to recount that state to try to fix things. Also: If the electoral college vote is close, you only have to recount the particular states that were close AND could have swung it.
If the president were chosen by a national popular vote, a corrupt machine in just one of the large states (NY, CA, etc.) could dump a bunch of fake or switched popular votes into the common pool, essentially controlling the presidency. Further, if the election is close and fraud or error is suspected, you may have to recount the whole COUNTRY to see if you "correct" enough votes to change or clarify the outcome. (If you thought Florida was a disaster: imagine that multiplied by fifty.)
Pardon my ignorance but why are inverters so damn expensive? [compared to things like stereo equipment]... Are there some crazy expensive components in there or is it more of a low volume/high margin economic reason?
- Need more reliability: Lives can be at stake.
- Harsher physical environment (heat/cold, battery acid in the air,...)
- Harsher electrical environment (switching spikes, lightning surges,...)
- Higher power. (Means bigger and more expensive components.)
- Lower volume.
- Regulatory requirement testing. (More and harsher tests, fewer units sold over which to distribute the cost of the testing and equipment sacrificed.)
- Less competition.
Will come down if a lot of 'em are installed. But probably won't approach a low-end Hi-Fi.
Where they're weak, however, is with energy storage. Compared with lithium-ion batteries, high-end ultracapacitors on the market today store 25 times less energy per pound.
I fail to see how a battery with less capacity than the actual ones can be of any use. Anyone care to explain?
Easy:
The factor-of-25-less-than-batteries power-to-weight ratio is for CURRENT PRODUCTION ultracapacitors from other vendors. That's the competition.
The 'way-better-than-batteries number is for their NEW one. They claim a power-to-weight-ratio improvement of more than a factor of 25 for their new gadget compared to the current ones.
Isn't there other capicitors in an earlier story..
Nano fibers used to increase a capacitors surface area i believe was the concept.
There was also a similar development with lithium cells, also using nano-fiber graphite forests for electrodes, producing hysterical energy densities and recharge rates (like 80% in a minute or so), high efficiency (since they'd slag down at that rate otherwise), and both long lives and a large numbers of cycles (since the graphite nanotubes don't tend to degrade anywhere but at the tips, and very slowly there.)
Somebody also did something similar with lead-acids, of all things. Built the plates' base structure by plating the lead onto a graphite (non-nano) fiber base, rather than starting from a lead skeleton. Greatly increased charge/discharge rate and efficiency (since the graphite conducts better than lead) significantly reduced weight (like well under half of a regular battery if I recall it right) and enormous increase in number of charge/discharge cycles before failure (since the graphite skeleton holds its shape rather than participating in the chemical reaction, which is what's behind some of a lead-acid's failure mechanisms - thus letting the plates "heal".)
And then there's vanadium redox...
Lots of good alternatives in the pipe. A conglomerate of oil companies would be hard pressed to buy them ALL up and bury 'em. B-)
Key has to be decrypted somewhere. Where else do you want to put it?
In a CPU register, such as those found in AltiVec or Streaming SIMD Extensions (SSE) that are capable of storing 128 bits.
And the next time the task switches the register gets stored in the task control block - which is in RAM.
If it's out on dedicated hardware somewhere but you're doing software decryption the task still has to be able to bring it back into a register, so it's still accessible to other software.
And if the decryption is done by dedicated hardware that is write-only for the key, the software still has to store the key. So you shim the driver and catch it on its way. Harder than dumping memory or sniffing the OS's task structures, but still easier than reverse-engineering the key decryption code and writing a mimic.
Locking down software for a GP computing platform from snooping by its owner, armed with sufficient privileges on the machine and whatever tools (soft or hard) that he choses to install, is a very tough problem. B-)
Big fat myth. They make far more power than it takes to make them.
But even if they didn't, you're not using the right comparison.
First: There are two basic kinds of power: Low-quality heat and high-quality stuff like shaft horsepower or electricity. To go from the low- to the high-quality form you have to pay the "carnot-cycle tax". Most of the energy used to make panels is in the form of heat - to smelt metal, refine silicon, and the like.
If you want heat from the sun for an industrial process you DON'T use photovoltaic to get high-quality electricity and then turn it into low-quality heat energy in a resistor. You can get a LOT more at a FAR lower cost and higher efficiency by collecting it directly.
Grids start out with fuel, burn it to generate low-quality heat, then run that through a heat engine to generate high quality energy, first as horsepower and then as electricity. Generation plants and power grids are FAR from a hundred percent efficient (and much of the inefficiency is that heat engine), so they consume far more energy than they deliver. Yet you don't hear the greenies claiming that makes them a net loss, eh?
Second: Panels deliver their output as high-quality electricity at the load site. You may have to store it to cover cloudy and dark times, but that's about it. Grids generate it somewhere else, then transport it. So grids have transmission losses. More inefficiency, more fuel burned to make the delivered high-quality electricity.
But (as with panels) you also have the energy cost of construction of the grid - or at least your pro-rated share of it: Smelting metal for transmission lines, generators, boilers, turbines, transformers, guy wires, towers, fittings, electric meters, breakers, and surge arresters. Melting sand for glass insulators. Cutting trees and treating them to make poles. Fueling the machinery to cut the right-of-way for the lines, haul the parts to the site and drive the workers back-and-forth, and erect and test the lines. (Not to mention consumption of the land under the right-of-way, which is largely taken out of service for other purposes, whether residential, agricultural, industrial, recreational, or ecological.
The cost of the instalation is a pretty good measure of the combined costs - energy, labor, and resource-consumption - of building the infrastructure, generating, and delivering the high-quality electric power to the load. Energy costs are the bulk of it in both cases. You find that, even with current technology, photovoltaic power beats grid power in a number of applications. Generally that's new construction in remote areas (where you can avoid the cost of building the last mile(s) of the grid feed and use that to pay for the instalation), small loads (road signs, yard lights, emergency phones, etc.) where it's cheaper to slap a panel on 'em than run a line, and maybe sites where you need backup power - and can convert it to full renewable energy by adding generation and tweaking the choice of battery size and inverter design. With technology improvements and rising fuel costs the breakeven point is reached in more areas. With enough improvement the grid may become obsolete in most places where reliable sun and/or wind is available.
Imagine if your inverter is pushing out the full -120v when the incoming mains comes back at +120...
It would be possible to build an inverter that would disconnect the incoming mains supply in the event of a power failure, and "slip" the inverter until it's in phase before dropping it back in, but you'd need something like a 100A contactor for that to work.
Actually, they drop it because grid-tie inverters are REQUIRED to disconnect from the grid when the grid goes down. This is to prevent backfeeding the disconnected island and frying a lineman who's trying to fix the downed wire for your block and thinks the lines are dead when YOU kept them live. (Those pole-pig transformers work just FINE in reverse, so a lineman might grab a line with 12,000 volts on it and a couple kilowatts to keep it that way while he's dancing and trying to breathe.)
Now the EASY way to do this is just to monitor the frequency and voltage, and shut the inverter off when it goes out of spec (meaning the grid is probably dead and the line only looks hot because of the inverter backfeeding it).
For a couple grand more, in the case of some good inverters that are designed for it (such as some of the Xantrex models), you can add a box with a relay, a phase-difference monitor, and a subsidiary "brain" board (or get an inverter with the function built in). (Actually the box in question usually also has the line monitoring circuit and combines with inverters that are otherwise stand-alone non-grid-tie.) That box will disconnect the inverter-and-keepalive-lodds from the line and let it keep going during an outage, then tell it to drift phase until it matches and hook it back up once the grid is back and has stabilized.
By their own stats only 47% of the spam tries just the primary MX of an unresponding 2-MX system, while 36% tries only the secondary and 17% tries both. So even before the spammers work around it they'd stop less than half the spam.
It looks like there might be a few spambots out there that only try the primary, but that about 3/4ths of those that only try one on each attempt make a random choice. Having only the secondary down rejects 36% rather than 47% of the spam, so the approach seems to have little to recommend it.
(You might stop something like 83% by implementing a stateful double-knock system - but again only until the spammers deploy a followon version of their bots rehacked to try all the MXes until they get through rather than just randomly pick one and poke it.)
I thought Al was the father of the internet, not this interloper!
No, he invented the internet. He's the father of Global Warming. Sheesh, I thought everyone knew that.
Actually it could be argued that Gore's legislation (opening the internet to commercial traffic, including advertising, and hamstringing its operators' ability to disconnect or penalize abusers) makes him the father of Spam.
What he missed is the biggest reason to oppose net neutrality legislation... any legislation is another step to the government fully regulating and controling the Internet. This is something we must avoid at all costs.
My impression (if only from the article) is that he gets it just fine and that's exactly his underlying point.
So far we've been lucky that the government hasn't come in and totally regulated it. We've got to work to push back what it does control now.
Actually the FCC commissioners have been actively pushing to keep governmental hands off the Internet at all levels. They've sued other branches of the federal government and knocked down state and local laws and regulations whenever governmental personnel attempt to get the camel's nose into the tent.
The bulk of the congresscritters also seem to have figured out that the Internet is currently the largest goose that's laying golden eggs, and keep voting down attempts by others of their number to regulate and/or tax it and the things that happen using it.
But the temptation to regulate is always there and regulations, once imposed, tend to ratchet ever upward. Then you're stuck (unless/until you can create a new infrastructure for sidestepping the regulations, as was done with internetworking).
= = = =
There are two parts to network "non-neutrality":
- One is the ability to treat different types of packets different treatment when they need it to perform their function correctly. This is what drives innovation, creating the ability to do more things well over the internet and cutting costs by combining many different types of networking into a single, less-expensive, infrastructure. This is a GOOD idea - even if it sometimes means some packets get better treatment than others.
- The other is making use of the tools built to treat packets differently to perform anti-competitive or customer-milking activities, such as giving packets for a service they sell better treatment than equivalent packets for their competitors' services or peer-to-peer replacements, blocking or downgrading Quality of Service (QoS) for some end-to-end services and then charging extra to re-enable or un-hobble them, and so on. This is the driver for possible anticompetitive unfairness.
Attempts to block the second type of "unfairness" are likely to block the first, and stifle innovation.
Even attempts to legislate against the second type must be carefully worded. Otherwise you might create pathologies like blocking a telephone company from moving their legacy traffic (with its higher QoS guarantees AND revenue streams to finance further infrastructure upgrades) onto their IP infrastructure and retiring their older network - using the resulting economies of scale to expand the IP net and thus reducing costs and increasing bandwidth for the best-effort IP traffic that shares the bigger, faster backbone, even though the "phone service packets" and "IPTV packets" are getting preferential routing and queueing to meet their higher QoS guarantees.
Meanwhile, the FCC asserts that it is already empowered to block unfairness of the second sort when it is creating actual unfair competition. Further, it gives examples where it has already doing so.
But Kahn's point is that the internet's own robustness and massive economies occured, despite being built on diverse networking platforms with diverse handling of information, precisely BECAUSE the designers weren't limited by such regulations and were thus free, in each case, to do the best they could invent.
If you RTFA you'll see that he explicitly disclaims attempts to label him the internet's daddy, pointing out that it came together through the efforts of a whole bunch of people, of which he and his co-inventor of TCP/IP were merely two - who happened to build a particularly high-profile piece as one of their contributions.
Thats easy, they plan on turning it over to the DOD.
Giving it to the NSA makes more sense.
Imagine: Instead of tagging conversations for human review when a keyword is present, they could have the acres of supercomputers analyze them for agreement on action items.
Then the automated agents could maintain and analyze a database of who agreed to what, flagging a collection of conversations for human review only if/when it amounted to a conspiracy to prepare for, support, or execute a military, geopolitical, or criminal activity of interest to the government.
Automated "big-brother". Pervasive area and phone bugs are much more useful if they can be monitored 24/7 by uniformly diligent and objective (even if not maximally perceptive and efficient) AIs, rather than occasionally by high-priced, error-prone, boredom-prone, subjective, and easily-corruptible humans.
Slashdot Mirror
When I was coming back from San Fran a couple weeks ago (after Macworld) there was a multi-million dollar NASA camera that'll be on the next space ship going through security at the same time I was. You'd think they'd check those things before hand.
And let the baggage handlers throw it around and/or put it on the next plane to Botswanaland by mistake? Hardly!
... why not Luxembourg? With a population of only 465,000 we'd made a majority of the population and be able to form a governmenmt.
For starters because Luxembourg won't let you move in and get citizenship all that easily.
And the population is rich enough to enforce their will: Pretty much every adult is an officer of several international corporations, at some serious pay each. This is because Luxembourg's laws make it advantageous to headquarter there, but require at least one citizen as a major officer.
Besides: Taking over by settling creates serious (sometimes deadly) opposition from those already there who FORMERLY ran their own government.
If you want to create a settlement where you can run your own government up to a significant level, try Oregon. If they remain true to their history, once you've established a significant colony of like-minded people, if you have a beef with the rest of your county they'll split it and give you your own county composed of you and your like-minded settlers. Then you can elect your own supervisors and sheriff, tax each other, maintain the roads your way, etc.
(Which is what makes the Ragneeshi's attempted takeover of Wasco county - by food-poisoning a salad bar at a local restaurant shortly before the election - such a stupid move: The state had already offered them a county composed of their own settlement and the roads to it.)
But software bloat increases faster than single-thread performance, thus making software run slower.
But if your application can run in its own thread the ongoing software bloat of the underlying OS won't affect its internal performance. (And the OS vendor may bloat it, but count on him to use enough multithreading to keep it out from underfoot - even if it chews up most of the processors in the multicore cluster.)
As long as you or your software vendor doesn't add inline bloating with the upgrades the application's performance will track that of the individual processors in the cluster - which will continue improving, though more gradually.
(If your vendor can't keep his engineers' hands off the critical code's performance his product is toast. Start looking for a replacement.)
What a good way to extract atmospheric moisture. And drifting pesticides, emmision fumes, particulate pollution blowing from every industrial site in the area... its like a pollutent-concentrator!
Kinda like a rainstorm, isn't it?
This is far superior to the "miss-to-kill" technology they were employing in previous models
Joke all you want, but that's what we've actually been doing. Didn't anybody ever tell you that close counts with hand grenades?
And nuclear weapons.
Which, unfortunately, was exactly the "big boom" "kinda close" that had been contemplated in some previous ABM designs.
After all, if it has blossomed, MIRV style, into a cloud of decoys and multiple real nuclear bombs on independent trajectories, spread out by quite a bit by the time your missile gets there, you need a BIG boom to disable all of them that matter. And without an atmosphere to carry a shockwave it helps if you can irradiate with heat, gammas, and neutrons, and can vaporize the whole antimissile and hit the targets with the vapor.
Problem with that is you're setting off your OWN nukes above your OWN targets - and high enough above the atmosphere to do a major electromagnetic pulse when the gamma burst makes a sheet of electrons the size of a continent jump upward by a few miles. (With defenses like that who needs an enemy missile with a real warhead? Other than to provoke the defense.)
So, yes, "hit to kill" is a BIG improvement over the "miss-to-kill technology they were using in previous models". (Assuming you have at least as many anti-missiles as they sent warheads and convincing decoys.)
And I'm sure that with a little web surfing we could come up with enough other black hats to fill all the MP slots.
Not that we need to, of course - and they'd be the wrong names, anyhow. Think what an opportunity this is for the trillionaire-"businessmen" who deploy the malware and run the phising scams and spam botnets.
Their own country...
How cool. So Microsoft is going to use their "genuine (dis)advantage" tools to get serious about collecting their license fees.
So what does this do to the "total cost of ownership" of windows versus open source solutions?
How much of those calculations especially at the PHB level - are done assuming either that all their installations are paid for (and nobody installed any extras or forged their identification) or that they can get away with extras - and in either case didn't factor in being audited? (That's a BIG cost even (especially) if it turns out you're squeaky-clean.)
Perhaps this will create additional incentives to switch.
Remember, no custom application on the iPhone
... what about feature support on other networks ? For example, the graphical voicemail.
I'm presuming that, if the machine doesn't do everything the customers want (ESPECIALLY VoIP over whatever network connection combined with being able to access home router and commercial hotspots), it will be hacked almost as soon as it's out there. B-)
(Actually I'm presuming it will be hacked almost as soon as it's out there, regardless - especially if it won't allow non-Apple software add-ons. B-) But being unable to use VoIP over your LAN will put the pressure on with a specific target from day zero.)
(I see we agree on what's the most important phone-like feature after VoIP roaming.)
That's already in place with plenty of phone systems. (For instance, our PBX does it at work.) You bring up the interface on a web browser, are presented with your voicemails, and when you select one it plays through your phone.
Such prior art should keep this from being one of their 200+ patents - or at least should keep any patent claims on the way they did it from simultaneously blocking such a web/phone hybrid and standing up to a challenge. (If it's already patented in the PBX context it may even expose Apple to a suit for doing it on the iPhone.)
If Verizon is smart it will try to head this off by dropping prices to where they're just covering network connectivity rather than subsidizing the non-existent "free" crippled phone.
Oops! Meant "Cingular".
And if Cingular/ATT and Verizon are BOTH clueless and leave the plans at regular cellphone rates, watch for users to start migrating to WiFi hotspot operators and WISPs.
(Watch for that anyhow, once people start hacking. B-) Even if Apple doesn't support it - or doesn't support it well - nobody in their right mind with a VoIP account and a WiFi AP at home is going to chew up cell minutes when at home when they can make the iPhone use their broadband and existing accounts, getting hax from their VoIP providers or third parties to make it work well.)
Since apple's phone isn't subsidized by the cell companies they have no leverage on how its firmware operates. This means:
- It's not locked into a carrier. You can switch in a heartbeat and/or put more than one plan on it.
- It's not locked OUT of using other systems than cellphone - like VoIP over WiFi or WiMax.
This means that the cellphone carriers are not just in competition with other cellphone and cellphone/data carriers. They're also in competition with Wireless ISPs (WISPs).
Even between the cellphone carriers the lack of the lock-in means they're in straight competition on price of service. (They had to do the lockin and early termination fee to pay for the handset tie-in.)
This will produce significant downward market pressure on cellphone companies.
Market forces don't produce a heavy drive toward marginal cost until there are at least THREE competing providers of the good or service. (For two the strategy is to track each other's prices and split the market about 50/50. For three or more the incentive is for the little guy to try to undercut the two biggest players and steal market from the pair - and for them to retaliate using their economy of scale.)
While there are several cellphone players now there are typically only two dominant players in most markets. The original bandwidth licensing regime was set up for "competition of two" (the incumbent phone company and ONE competitor) and the early rollout gave two players incumbent status in most markets. They then had an analog of the government-subsidized copper buildout of the wireline phone companies that gave them an advantage in coverage as the upstarts started up - leading to sickly third players and rounds of consolidation.
This device lets WISPs with significant coverage play in the cellphone space - and use their bandwidth cost advantage to become major players. If Verizon is smart it will try to head this off by dropping prices to where they're just covering network connectivity rather than subsidizing the non-existent "free" crippled phone.
Strongarm states that don't conform their own laws to national 'standard' election laws by threatening Federal funding sanctions. That's been the common practice for quite a while, as far as I understand it.
They did that with speed limits and highway funds, and some states resisted or ignored it. (The states collect much of the federal money - such as gas taxes - and they can turn the tables and refuse to turn it in.)
Interference in election procedures is a MUCH bigger hotbutton issue for state officials than traffic laws. It's fundamental to the governmental underpinning and their own selection. States with honest politicians will resist on constitutional issues. (States with political machines in place will resist to maintain their lock on elections and claim to be resisting on constitutional issues. B-) )
Even if the change were benign it might affect the outcome. The people in charge in the states were elected by the old procedures and have an interest in not sinking the boat that ferries them into their offices every election cycle.
(If you thought Florida was a disaster: imagine that multiplied by fifty.)
AND happening EVERY TIME!
The decentralized form of elections might have made sense for the age of horse coaches but in the age of internet it is not too tough to have thge same standards everywhere in the US.
The US is not a country. It is a confederation of countries ("the several States") united by a set of treaties (ratifications of the Constitution or treaties of incorporation, depending on when/how they joined up).
Each of these treaty-bound but independent countries has its own election procedures, and they did not give up their right to control their own election procedures when they joined. The constitution doesn't authorize the Fed to control election procedures in general (though there are things it DOES authorize that let the Federal courts to intervene in details occasionally).
Part of the reason that the president is chosen by the electoral college, rather than the popular vote, is to provide a firewall between election corruption in particular states and the selection of the president (the one case where the population of the whole country votes on a common federal-level decision).
With each state selecting only its own electors, a corrupt voting system can only affect the electors from that state. A state can do no more than throw all its electors to one candidate (which most of them do) and even in close races most of the states aren't in a position to swing the election. (Political machines are run by the party in power, which generally has a significant majority among the overall population of the state. So the state's electors will generally be chosen their way and there's nothing to be gained by election fraud beyond the point that insures it.) If things were close (typically when the machine has lost the population but isn't out yet due to election corruption or jerrymandering) and they swung their own state's election through vote fraud, you only have to recount that state to try to fix things. Also: If the electoral college vote is close, you only have to recount the particular states that were close AND could have swung it.
If the president were chosen by a national popular vote, a corrupt machine in just one of the large states (NY, CA, etc.) could dump a bunch of fake or switched popular votes into the common pool, essentially controlling the presidency. Further, if the election is close and fraud or error is suspected, you may have to recount the whole COUNTRY to see if you "correct" enough votes to change or clarify the outcome. (If you thought Florida was a disaster: imagine that multiplied by fifty.)
Pardon my ignorance but why are inverters so damn expensive? [compared to things like stereo equipment] ... Are there some crazy expensive components in there or is it more of a low volume/high margin economic reason?
...) ...)
- Need more reliability: Lives can be at stake.
- Harsher physical environment (heat/cold, battery acid in the air,
- Harsher electrical environment (switching spikes, lightning surges,
- Higher power. (Means bigger and more expensive components.)
- Lower volume.
- Regulatory requirement testing. (More and harsher tests, fewer units sold over which to distribute the cost of the testing and equipment sacrificed.)
- Less competition.
Will come down if a lot of 'em are installed. But probably won't approach a low-end Hi-Fi.
Where they're weak, however, is with energy storage. Compared with lithium-ion batteries,
high-end ultracapacitors on the market today store 25 times less energy per pound.
I fail to see how a battery with less capacity than the actual ones can be of any use.
Anyone care to explain?
Easy:
The factor-of-25-less-than-batteries power-to-weight ratio is for CURRENT PRODUCTION ultracapacitors from other vendors. That's the competition.
The 'way-better-than-batteries number is for their NEW one. They claim a power-to-weight-ratio
improvement of more than a factor of 25 for their new gadget compared to the current ones.
Make sense now?
Geez.. ultracapacitors.. we had supercapacitors till now.. whats next.. ubercapacitors? ubersuperultracapacitors..
That's easy: Hypercapacitors.
(You can tell when they're coming by the amount of hype.)
Isn't there other capicitors in an earlier story..
Nano fibers used to increase a capacitors surface area i believe was the concept.
There was also a similar development with lithium cells, also using nano-fiber graphite forests for electrodes, producing hysterical energy densities and recharge rates (like 80% in a minute or so), high efficiency (since they'd slag down at that rate otherwise), and both long lives and a large numbers of cycles (since the graphite nanotubes don't tend to degrade anywhere but at the tips, and very slowly there.)
Somebody also did something similar with lead-acids, of all things. Built the plates' base structure by plating the lead onto a graphite (non-nano) fiber base, rather than starting from a lead skeleton. Greatly increased charge/discharge rate and efficiency (since the graphite conducts better than lead) significantly reduced weight (like well under half of a regular battery if I recall it right) and enormous increase in number of charge/discharge cycles before failure (since the graphite skeleton holds its shape rather than participating in the chemical reaction, which is what's behind some of a lead-acid's failure mechanisms - thus letting the plates "heal".)
And then there's vanadium redox...
Lots of good alternatives in the pipe. A conglomerate of oil companies would be hard pressed to buy them ALL up and bury 'em. B-)
Key has to be decrypted somewhere. Where else do you want to put it?
In a CPU register, such as those found in AltiVec or Streaming SIMD Extensions (SSE) that are capable of storing 128 bits.
And the next time the task switches the register gets stored in the task control block - which is in RAM.
If it's out on dedicated hardware somewhere but you're doing software decryption the task still has to be able to bring it back into a register, so it's still accessible to other software.
And if the decryption is done by dedicated hardware that is write-only for the key, the software still has to store the key. So you shim the driver and catch it on its way. Harder than dumping memory or sniffing the OS's task structures, but still easier than reverse-engineering the key decryption code and writing a mimic.
Locking down software for a GP computing platform from snooping by its owner, armed with sufficient privileges on the machine and whatever tools (soft or hard) that he choses to install, is a very tough problem. B-)
Big fat myth. They make far more power than it takes to make them.
But even if they didn't, you're not using the right comparison.
First: There are two basic kinds of power: Low-quality heat and high-quality stuff like shaft horsepower or electricity. To go from the low- to the high-quality form you have to pay the "carnot-cycle tax". Most of the energy used to make panels is in the form of heat - to smelt metal, refine silicon, and the like.
If you want heat from the sun for an industrial process you DON'T use photovoltaic to get high-quality electricity and then turn it into low-quality heat energy in a resistor. You can get a LOT more at a FAR lower cost and higher efficiency by collecting it directly.
Grids start out with fuel, burn it to generate low-quality heat, then run that through a heat engine to generate high quality energy, first as horsepower and then as electricity. Generation plants and power grids are FAR from a hundred percent efficient (and much of the inefficiency is that heat engine), so they consume far more energy than they deliver. Yet you don't hear the greenies claiming that makes them a net loss, eh?
Second: Panels deliver their output as high-quality electricity at the load site. You may have to store it to cover cloudy and dark times, but that's about it. Grids generate it somewhere else, then transport it. So grids have transmission losses. More inefficiency, more fuel burned to make the delivered high-quality electricity.
But (as with panels) you also have the energy cost of construction of the grid - or at least your pro-rated share of it: Smelting metal for transmission lines, generators, boilers, turbines, transformers, guy wires, towers, fittings, electric meters, breakers, and surge arresters. Melting sand for glass insulators. Cutting trees and treating them to make poles. Fueling the machinery to cut the right-of-way for the lines, haul the parts to the site and drive the workers back-and-forth, and erect and test the lines. (Not to mention consumption of the land under the right-of-way, which is largely taken out of service for other purposes, whether residential, agricultural, industrial, recreational, or ecological.
The cost of the instalation is a pretty good measure of the combined costs - energy, labor, and resource-consumption - of building the infrastructure, generating, and delivering the high-quality electric power to the load. Energy costs are the bulk of it in both cases. You find that, even with current technology, photovoltaic power beats grid power in a number of applications. Generally that's new construction in remote areas (where you can avoid the cost of building the last mile(s) of the grid feed and use that to pay for the instalation), small loads (road signs, yard lights, emergency phones, etc.) where it's cheaper to slap a panel on 'em than run a line, and maybe sites where you need backup power - and can convert it to full renewable energy by adding generation and tweaking the choice of battery size and inverter design. With technology improvements and rising fuel costs the breakeven point is reached in more areas. With enough improvement the grid may become obsolete in most places where reliable sun and/or wind is available.
Imagine if your inverter is pushing out the full -120v when the incoming mains comes back at +120...
It would be possible to build an inverter that would disconnect the incoming mains supply in the event of a power failure, and "slip" the inverter until it's in phase before dropping it back in, but you'd need something like a 100A contactor for that to work.
Actually, they drop it because grid-tie inverters are REQUIRED to disconnect from the grid when the grid goes down. This is to prevent backfeeding the disconnected island and frying a lineman who's trying to fix the downed wire for your block and thinks the lines are dead when YOU kept them live. (Those pole-pig transformers work just FINE in reverse, so a lineman might grab a line with 12,000 volts on it and a couple kilowatts to keep it that way while he's dancing and trying to breathe.)
Now the EASY way to do this is just to monitor the frequency and voltage, and shut the inverter off when it goes out of spec (meaning the grid is probably dead and the line only looks hot because of the inverter backfeeding it).
For a couple grand more, in the case of some good inverters that are designed for it (such as some of the Xantrex models), you can add a box with a relay, a phase-difference monitor, and a subsidiary "brain" board (or get an inverter with the function built in). (Actually the box in question usually also has the line monitoring circuit and combines with inverters that are otherwise stand-alone non-grid-tie.) That box will disconnect the inverter-and-keepalive-lodds from the line and let it keep going during an outage, then tell it to drift phase until it matches and hook it back up once the grid is back and has stabilized.
It's not even a temporary solution.
By their own stats only 47% of the spam tries just the primary MX of an unresponding 2-MX system, while 36% tries only the secondary and 17% tries both. So even before the spammers work around it they'd stop less than half the spam.
It looks like there might be a few spambots out there that only try the primary, but that about 3/4ths of those that only try one on each attempt make a random choice. Having only the secondary down rejects 36% rather than 47% of the spam, so the approach seems to have little to recommend it.
(You might stop something like 83% by implementing a stateful double-knock system - but again only until the spammers deploy a followon version of their bots rehacked to try all the MXes until they get through rather than just randomly pick one and poke it.)
I thought Al was the father of the internet, not this interloper!
No, he invented the internet. He's the father of Global Warming. Sheesh, I thought everyone knew that.
Actually it could be argued that Gore's legislation (opening the internet to commercial traffic, including advertising, and hamstringing its operators' ability to disconnect or penalize abusers) makes him the father of Spam.
What he missed is the biggest reason to oppose net neutrality legislation ... any legislation is another step to the government fully regulating and controling the Internet. This is something we must avoid at all costs.
My impression (if only from the article) is that he gets it just fine and that's exactly his underlying point.
So far we've been lucky that the government hasn't come in and totally regulated it. We've got to work to push back what it does control now.
Actually the FCC commissioners have been actively pushing to keep governmental hands off the Internet at all levels. They've sued other branches of the federal government and knocked down state and local laws and regulations whenever governmental personnel attempt to get the camel's nose into the tent.
The bulk of the congresscritters also seem to have figured out that the Internet is currently the largest goose that's laying golden eggs, and keep voting down attempts by others of their number to regulate and/or tax it and the things that happen using it.
But the temptation to regulate is always there and regulations, once imposed, tend to ratchet ever upward. Then you're stuck (unless/until you can create a new infrastructure for sidestepping the regulations, as was done with internetworking).
= = = =
There are two parts to network "non-neutrality":
- One is the ability to treat different types of packets different treatment when they need it to perform their function correctly. This is what drives innovation, creating the ability to do more things well over the internet and cutting costs by combining many different types of networking into a single, less-expensive, infrastructure. This is a GOOD idea - even if it sometimes means some packets get better treatment than others.
- The other is making use of the tools built to treat packets differently to perform anti-competitive or customer-milking activities, such as giving packets for a service they sell better treatment than equivalent packets for their competitors' services or peer-to-peer replacements, blocking or downgrading Quality of Service (QoS) for some end-to-end services and then charging extra to re-enable or un-hobble them, and so on. This is the driver for possible anticompetitive unfairness.
Attempts to block the second type of "unfairness" are likely to block the first, and stifle innovation.
Even attempts to legislate against the second type must be carefully worded. Otherwise you might create pathologies like blocking a telephone company from moving their legacy traffic (with its higher QoS guarantees AND revenue streams to finance further infrastructure upgrades) onto their IP infrastructure and retiring their older network - using the resulting economies of scale to expand the IP net and thus reducing costs and increasing bandwidth for the best-effort IP traffic that shares the bigger, faster backbone, even though the "phone service packets" and "IPTV packets" are getting preferential routing and queueing to meet their higher QoS guarantees.
Meanwhile, the FCC asserts that it is already empowered to block unfairness of the second sort when it is creating actual unfair competition. Further, it gives examples where it has already doing so.
But Kahn's point is that the internet's own robustness and massive economies occured, despite being built on diverse networking platforms with diverse handling of information, precisely BECAUSE the designers weren't limited by such regulations and were thus free, in each case, to do the best they could invent.
If you RTFA you'll see that he explicitly disclaims attempts to label him the internet's daddy, pointing out that it came together through the efforts of a whole bunch of people, of which he and his co-inventor of TCP/IP were merely two - who happened to build a particularly high-profile piece as one of their contributions.
Thats easy, they plan on turning it over to the DOD.
Giving it to the NSA makes more sense.
Imagine: Instead of tagging conversations for human review when a keyword is present, they could have the acres of supercomputers analyze them for agreement on action items.
Then the automated agents could maintain and analyze a database of who agreed to what, flagging a collection of conversations for human review only if/when it amounted to a conspiracy to prepare for, support, or execute a military, geopolitical, or criminal activity of interest to the government.
Automated "big-brother". Pervasive area and phone bugs are much more useful if they can be monitored 24/7 by uniformly diligent and objective (even if not maximally perceptive and efficient) AIs, rather than occasionally by high-priced, error-prone, boredom-prone, subjective, and easily-corruptible humans.