Additionally, managers in silicon valley have a track record of strong bias in favor of graduates from a single-digit list of colleges, and of Caucasian, Oriental, or Indian descent males.
(The university business is particularly blecherous since the actual pioneers of the information age were almost entirely NOT ivy-leaguers, and had more than a sprinkling of non wasp-males among their number.)
If you include women, blacks, American Indians, Hispanic-descent citizens, various "halfbreeds", and graduates of other fine universities (especially state universities) - rather than reserving them to support (or janitorial) positions, there is no shortage whatsoever.
But there's another part of this: You have to PAY them on the basis of performance, respect their opinions, and avoid filing the serial numbers off their ideas and crediting them to the stars from that tiny pool of ivy-league whites and orientals. If you hire them and then systematically abuse them and pay them 2/3 of what you pay the in crowd, they'll burn out and drop out.
(I watched this happen to an exceptional talent. Woman. Part Amerind. State universities. IQ so high a psych professor had to roll a special test to estimate it. Four degrees, one advanced and from a top U, in diverse subjects (computer science among them). Sharp as a tack and total grasp of the subtleties of software engineering. Yet administrators systematically ignored or rejected her ideas or credited her colleagues for them when they were finally accepted. Last straw was in a windows application development shop when she found out the clueless-about-Windows-programs unix people she was teaching were paid more than half again what she got. She left the field.)
If you realy wanted to make money and you had a fantastic back catalog, you could sell flash player MP3 CDs with 4 or more albums on each CD at above 128K bitrate for under $10. Man I would be first in line for the back collections of Pink Floyd, ELO, Styx, Guns & Roses, Chicago, etc. They are happy with $0 instead of 100's of dollars for the back catalog at value pricing.
They either haven't heard, or don't believe in, an old saying:
"Fast nickels are better than slow dimes."
For two reasons:
- You get far more money once you add it up.
- You start getting it earlier.
They're so afraid of losing the slow dimes that they're missing the fast nickels.
Odd that you think off-grid solar is cost effective and grid-tie isn't. Off-grid solar usually has batteries. Since people don't want their batteries to be constantly in discharge, they often end up with the panels throwing away the output because the batteries are charged or close to charged. This really screws up the economics of panels.
What makes off-grid cost-effective is when it saves you enough by NOT running the grid to the site to pay for much or all of the system.
Example: Suppose the cost of the system - panels, batteries, inverter, wiring (excluding the house wiring), instalation, and all comes to exactly the same as the grid hookup. Now your instalation is FREE. Your power cost become the cost of maintainence for the system - mainly replacing the batteries every five to ten years. That's a drop in the bucket compared to a power bill.
With grid-tie, all power generated by the panels is always used, either in the house, or by the grid. The grid is your 100% efficient storage.
Not really, though it's close. Two main losses:
If you feed more than you use in a given year the excess is lost. (Like the dump load on the batteries.)
And you still pay the connect fee. (In the case of Sierra power in Nevada that's currently $6/month. $72/year would cover the periodic replacement costs for about 5-10 KWHr of your deep-cycle battery capacity.)
Thousands upon thousands of people have done it, and continue to do so. The economics are well-settled at this point.
Photovoltaic generation on a house-load level IS cost-effective in one situation: New construction in rural areas, where it displaces running a long (and high-priced) grid connection. Then the money that would have been spent on the grid tie can be spent on the capital cost of the photovoltaic system instead.
This one requires a grid tie for net metering, so that displacement is not available.
Unfortunately, equipment costs for grid-tied photovoltaic equipment is still high enough that you're ahead to invest the money it would have cost and spend the interest buying power for the life of the system you didn't install. (This could change with enough lowering of equipment costs or raising of electric power prices.)
If you have a source for equipment inexpensive enough to back up your claim, please let us know what it is. I have two houses where I'd LOVE to install such a system.
The plan is not feasible, with or without the multi-level scheme.
Solar installations of house-size with a net-metering grid hookup are not cost-competitive with grid power, even with government subsidies and without paying for the space under them. Otherwise people would be able to save money by doing this themselves, without the middleman and his pyramid scheme.
The difference currently is a factor of several - too large for even an exceedingly efficient company's economy of scale to overcome. It's dropping. But it's still far from crossover. (When it DOES cross over there will be efficient companies building customer-owned installations that homeowners finance with the mortgage, as part of their houses.)
Imagine a point-to-point connection where every link in the network connecting them was unsaturated, all the routers had spare CPU capacity, and there was 0% packet loss.
Imagine a world without war.
Imagine more money than you can spend.
Imagine no possessions I wonder if you can Nothing to kill or die for A brotherhood of man Imagine all the people Sharing all the world
Or all that bandwidth so great that "every link in the network [] was unsaturated, all the routers had spare CPU capacity, and there was 0% packet loss"
Trust me:
1) Bandwidth expands to occasionally congest an installed link of any size. (The TCP throttling algorithm actually is written to do precisely that, in order to maximize throughput.)
2) The transport providers aren't going to install, and the ISPs aren't going to buy, enough backbone bandwidth to guarantee zero packet drop to all best-effort traffic, despite total saturation of all edge links to customers. (Their investors won't let them - and if they did they'd still be bankrupted by competitors who didn't, and could thus undersell them.)
3) Even if the backbone links didn't saturate, file transfers over TCP alone will (automatically) ramp up until they saturate edge links, killing the QoS needed for other services on those links unless there is QoS-driven queueing.
But if you'd actually bothered to read my previous post you'd understand that, since I made the points there.
Meanwhile, legacy services include contractual guarantees of QoS levels. In order to live up to those contractual committments while converging that traffic onto a common IP-based backbone the ISP must have similar guarantees that the designated packets will go thorough. To do that he must (first) insure that there is bandwidth to carry them and (second) insure that competing UN-guaranteed packets don't create congestion that bumps them.
He can do the first easily. But the second requires that un-guaranteed packets receive lower service levels than guaranteed packets whenever there are more packets than capacity, period.
Bland assertions that he COULD have installed enough bandwidth that the packets would never be congested won't cut it. The installed bandwidth is what it is, and any momentary burst of traffic that exceeds it can not bump the packets for the guaranteed service or the contract is violated.
So the ISP requires QoS-guaranteed service or he can't converge.
These whines are in fact "special consideration" pressure for the telcos to get "Net Doublecharge". They don't need service tiers,...
I have to disagree with that.
Their legacy services are connection-oriented. They have quality of service guarantees on packet delivery that can NOT be met on a best-effort network, period.
So they have three choices:
1) Maintain separate networks.
2) Split the bandwidth on a common transport between a best-effort packet network and a connection-oriented network.
3) Pool the bandwidth in a common tranport, with the connection-oriented traffic implemented as packet streams with overriding QoS guarantees and the best-effort traffic taking all the rest.
1) Is enormously expensive. Companies that stick with that will be eaten alive by those that switch.
2) Is a big cost savings. But it means there is a bunch of bandwidth reserved for connection-oriented traffic that is not being used and thus wasted. (It's also a pain to administer and requires either more complicated software or more equipment, both of which increase costs.)
3) Is still cheaper per bit. Yet it makes all the bandwidth that isn't actually assigned to a connection at any given moment available to the best-effort network. Same or lower infrastructure cost providing more bandwidth to the best-effort network means a better price/performance ratio.
But 3) means treating some packets better than others. That's forbidden by "Network Neutrality". Oops!
Increasing the bandwidth of the backbone will NOT fix this - at least until the backbone capacity exceeds the total of the last-mile capacity - a "won't happen". (Even THEN it would still be broken across the last mile connection to your house.) In the absence of per-stream QoS enforcement by the routers the TCP streams will grab all the available bandwidth and divide it among themselves. Start some file transfers and you kill your audio and video feeds and phone line. A popular new torrent release kills EVERYBODY's audio and video feeds and phone lines, including the major networks and 911 service.
"3)" is "The Convergence". It lets the carriers put ALL their traffic on a single infrastructure, achieving enormous economies of scale and simplification of equipment and operatioin. Yes, it means the backbone must be upgraded to make QoS decisions on a packet-by-packet basis (in addition to the routing, queueing, and throttling decisions they already make). But the backbone bandwidth has to be upgraded by several orders of magnitude anyway, just to handle the currently foreseeable load of a converged infrastructure. So adding QoS at the same time is just a matter of more checkboxes for the equipment manufacturers' software departments.
IF the pressure to legislate "Network Neutrality" doesn't succeed (or at least doesn't spread beyond requiring equivalent SERVICES by different end providers to pay different rates.)
Treating the competition's packets worse than your own for a particular service is ALREADY an antitrust violation, a false-advertising issue if it wasn't explicitly warned about in advertising, and something the FCC jumps on and fines carriers for whenever it rears its ugly head. So new legislation is not required to deal with it.
I hear that much of algebra was indeed developed by Muslims - for a reason:
Both men and women owned property which wasn't pooled by marriage and on death a person's assets were divided among the spouse(s) and offspring in various prescribed proportions. With up to four wives (at a time) in a marriage, some of whom may have brought children from a previous marriage, figuring this out got very complicated despite the simple initial rules - until you had algebra to set up and solve the problems.
Short version is that defending a libel case in the states is usually easy unless the case in truly eggregious.
Conversely, brining a suit would almost certainly be disastrous if there were a hint of truth to the defamation - achieving far wider publication of the allegations while uncovering the truth behind them. So the fact that he's suing at all is a strong indication that the smear is false.
A fair point, but even a successful libel case isn't going to remove that stigma from your name.
It will for anyone familiar with the law - provided the case receives as much coverage as the original libel or slander.
Truth is an absolute defense against suits for libel or slander. So a defamed person would be a fool to sue if there was a trace of truth to the smear: Such truth would almost certainly come out, with MUCH wider coverage, at the trial. And then he'd also be on the hook for filing a bogus suit. Thus just the fact that he's suing is a strong indication that the smear is completely false.
Meanwhile, the fact of the suit replacing the smear in the Wiki entry serves as a long-lived retraction, with at least the reach of the original smear. Even old caches are automatically linked to the updated entry, while the news articles also link back and give it broader coverage. No doubt the entry will eventually be updated with the result of the suit, as well, and last for a very long time.
It's nice to see that Wikipedia's structure so strongly resists use as a defamation tool. It's good both for the project itself and for a world of potential victims.
"Thee shall know the truth, and the truth shall make you free."
Looks as though the defamatory remarks are sort of back in the article. Except now they are there as a factual statement that he is suing for someone posting those remarks. I find that ironic and funny.
The reporting of the remarks doesn't quote them. Instead it paraphrases them in a way that makes it clear that they were an over-the-top vandalism rather than having any connection to Zoeller's actual behavior.
Further, anybody who is at all familiar with the law on defamation would be aware that truth is an absolute defense. Zoeller would be a fool to file such a suit if there were any truth at all to the allegations.
So, no, I don't see any irony or humor in the entry containing the facts of the suit including a paraphrase of the remarks in question. If anything it serves as a retraction and a defense of Zoeller's character.
When was the last time you saw, say, the New York Times give as much prominence to a retraction of an erroneous story as they gave to the original story. Yet this is exactly what Wikipedia is doing.
Further, the retraction will no doubt remain posted far longer than the original smear, and (thanks to both the expanding fame of Wikipedia and the press coverage of this issue) come to the attention of far more people.
"AT&T is willing to concede that software isn't patentable, if the Court will conclude that the things that software does - the methods and procedures and instructions that a processor carries out once software is installed - are patentable.
Or for us mere developers: even if software is not patentable the lawyers will still have a way with the words to F*ck you.
I doubt the supremes would go for that.
The law explicitly exempts "mathematical algorithms" from patentability. But a "mathematical algorithm" is just a set of instructions for performing a mathematical computation. AT&T's argument would make it possible to patent performing the computation. If that's true, what was Congress saying when they blocked patenting mathematical algorithms?
"It's OK to contemplate the algorithm while meditating but not to actually USE it." No, I don't think so. And the Supremes are the country's experts at seeing through specious legal arguments.
If they rule in favor of Microsoft, sure there is a lot of different rulings they could make, but it still comes down to one of two things: (1) They say software patents are invalid, the issue is moot, case dismissed, or (2) software patents are valid, but here's why we ruled for Microsoft.
I wish it were that way. But (2) would likely come out this way:
"We ruled for Microsoft because of this other reason, so we didn't have to resolve whether software patents are valid."
Being NAL, I don't recall the pseudo-Latin legalize for it. But I understand that one of the court's principles is to avoid making a decision that shakes things up unless it's the only way to settle a case.
(Of course they might still pick patentability of software as the reason to rule - especially if they decided the Federal District had egregiously ignored or misinterpreted their previous rulings, in order to clean up the mess and/or reassert their authority. But they don't necessarily have to settle the issue if they rule for Microsoft on some other grounds.)
Perhaps Steve Ballmer ought to have checked with legal before mouthing off about Linux and intellectual property yesterday?
Maybe legal WANTS him to mouth off about Linux and IP.
Just think: If he raises enough stink about how Microsoft is ready to use patents to break Linux and other FOSS projects if software IS held to be patentable, and the justices hear about it...
Once you have an IP address, you look up who owns it and you call them. They do their research, looking at things such as DNS records, DHCP assignments, DSLAM logs, etc... They then look up which customer that was, and there you go.
Note that the bad guy may be using a dynamic IP address assigned through DHCP - especially if the new possessor of the laptop is using a dialup ISP, or if somebody just plugs it in at work. If so you also need the time at which the IP address was captured, so the owner of the IP can figure out which of the users who happened to get that IP address was the culprit.
"I always knew that a geek would make a great husband." Cheer up bunky, it could happen to a 'Dotter. Some day. The odds are certainly no worse than finding, say, extraterrestrial life.
Happened to me.
Pre-slashdot, though. We first met on a netnews group but didn't really hit it off until introduced in person by a friend at a party.
Still married, too. Despite losing most of the first nest egg in the internet bubble burst.
That anyone who doesn't like the inefficiency of voting shouldn't be allowed to vote? Voting should *not* take an hour.
I am saying that voting has a very important purpose (preventing civil war) and making it too convenient defeats the purpose (increasing the risk of civil war).
But if you prefer intermittent civil war to waiting in line to vote, feel free to vote for easier voting.
I had to wait in line for over an hour just to cast my ballot. I had to stand outside (in South Florida), listen to old people bitch about everything around them, and then get harassed by the security guards---all to vote in a free election.
Suppose the issues were so important that, in the absence of the election, you'd have volunteered to fight in a civil war over them. How much discomfort would you endure to fight for your beliefs in that war?
Elections are NOT about being nice, or being fair, or because there's some magic reason why (50%+1) voters should run the lives of the other (50%-1). Elections are about convincing the losers that they can't reverse the decision - on the issues, or on picking decision-makers - by going to war to overthrow the government and install their party/dictator. This stabilizes the republic.
That's why, until a few years ago in the last canton to repeal the requirement, the Swiss carried swords to their elections: They used to vote by raising them to be counted. That's why, in the US, the franchise was originally only for the classes of people who were the major fighters in the revolution (property owners and merchants), and was extended to one group after another as they proved themselves able to engage in organized violence to advance their political agendas.
That's why (until recently), registration was about as hard and inconvenient as signing up with a militia, and voting was about as hard and inconvenient as showing up for a muster.
Does the outcome matter to you, enough to fight a war over it? If so, it should matter enough to stand in line for an hour or two. If not, you shouldn't vote.
Now if voting is made harder for some than for others you have a gripe: Such bias lowers the accuracy of the model - until it may NOT predict the outcome of the civil war, and thus not prevent it. Such distortions need to be hunted down and eliminated.
But making it so easy that masses of people who would never be bothered to fight still vote on issues that don't bother them, and count as much as smaller groups of people who really care and are chomping at the bit to promote their agenda - by violence if necessary, also biases the model and destabilizes the country. So it's a really bad idea.
In hydrocarbons burning the hydrogen provides most of the energy. Burning the carbon provides some, but the carbon is mainly useful for packing the hydrogen in a form more dense than H2 gas for convenient storage and handling.
As hydrocarbons go, CH4 has a higher ratio of hydrogen to carbon than any other molecule: Every bond on every carbon holds a hydrogen, none are "wasted" connecting to other carbons.
So if you're going to burn hydrocarbons for energy, methane releases the least CO2 for a given amount of energy produced.
It seems to me that such things can easily be replaced by a small heating element, perhaps a small ceramic one like one uses for gas ovens or a foil battery warmer. Not as cheap as a lightbulb but certainly would outlast one by decades.
You really don't understand the situation out in the sticks.
- Things have to be maintained - and usually built - using supplies available at rural stores.
- Even the rural stores may be an hour or two away by road in the best of weather, and inaccessible for days in the winter. The nearest city may be a small town, too, with something substantial enough to have a suitable industrial supply maybe a couple hundred mile round trip. Add the gas price and the opportunity cost of spending a day driving just to get a part...
When was the last time you saw a "small ceramic heating element" available in a mom-and-pop grocery? When was the last time you DIDN'T find light bulbs in one - especially if it was the only store for 40 miles?
You also don't want the type used in gas ovens: In addition to the high drain, they reach a temperature suitable for igniting gas (or anything else flammable that comes in contact with them. That's not what you want in the battery shed when the wind is blowing and the windcharger has topped of the batteries and the dump load controller has decided to give them an equalizing charge - so they're now putting out enough hydrogen gas to create an explosion if the venting got clogged by a snowdrift or silverthaw.
Fluorescents aren't suitable: In addition to having trouble starting at low temperatures (exactly when you need them), the last thing you want about boot-toe high is a target that will shatter easily, dumping mercury in the pumphouse for your place's water - right next to the well, within the circle that must be kept clean of toxics due to rain seeping down. (Bulbs take a lot to break them.
Slashdot Mirror
They're not interested in the absolute temperature, they're interested in the change in temperature over both short and long time periods.
Of course to measure the change you first have to measure the temperature at all. Then you wait a while and do it again.
Since they couldn't get there to measure it before this is that first measurement. Any comparisons are against models.
Meanwhile the definition of amount of information obtained is "how much it surprised the receiver".
Additionally, managers in silicon valley have a track record of strong bias in favor of graduates from a single-digit list of colleges, and of Caucasian, Oriental, or Indian descent males.
(The university business is particularly blecherous since the actual pioneers of the information age were almost entirely NOT ivy-leaguers, and had more than a sprinkling of non wasp-males among their number.)
If you include women, blacks, American Indians, Hispanic-descent citizens, various "halfbreeds", and graduates of other fine universities (especially state universities) - rather than reserving them to support (or janitorial) positions, there is no shortage whatsoever.
But there's another part of this: You have to PAY them on the basis of performance, respect their opinions, and avoid filing the serial numbers off their ideas and crediting them to the stars from that tiny pool of ivy-league whites and orientals. If you hire them and then systematically abuse them and pay them 2/3 of what you pay the in crowd, they'll burn out and drop out.
(I watched this happen to an exceptional talent. Woman. Part Amerind. State universities. IQ so high a psych professor had to roll a special test to estimate it. Four degrees, one advanced and from a top U, in diverse subjects (computer science among them). Sharp as a tack and total grasp of the subtleties of software engineering. Yet administrators systematically ignored or rejected her ideas or credited her colleagues for them when they were finally accepted. Last straw was in a windows application development shop when she found out the clueless-about-Windows-programs unix people she was teaching were paid more than half again what she got. She left the field.)
If you realy wanted to make money and you had a fantastic back catalog, you could sell flash player MP3 CDs with 4 or more albums on each CD at above 128K bitrate for under $10. Man I would be first in line for the back collections of Pink Floyd, ELO, Styx, Guns & Roses, Chicago, etc. They are happy with $0 instead of 100's of dollars for the back catalog at value pricing.
They either haven't heard, or don't believe in, an old saying:
"Fast nickels are better than slow dimes."
For two reasons:
- You get far more money once you add it up.
- You start getting it earlier.
They're so afraid of losing the slow dimes that they're missing the fast nickels.
Odd that you think off-grid solar is cost effective and grid-tie isn't. Off-grid solar usually has batteries. Since people don't want their batteries to be constantly in discharge, they often end up with the panels throwing away the output because the batteries are charged or close to charged. This really screws up the economics of panels.
What makes off-grid cost-effective is when it saves you enough by NOT running the grid to the site to pay for much or all of the system.
Example: Suppose the cost of the system - panels, batteries, inverter, wiring (excluding the house wiring), instalation, and all comes to exactly the same as the grid hookup. Now your instalation is FREE. Your power cost become the cost of maintainence for the system - mainly replacing the batteries every five to ten years. That's a drop in the bucket compared to a power bill.
With grid-tie, all power generated by the panels is always used, either in the house, or by the grid. The grid is your 100% efficient storage.
Not really, though it's close. Two main losses:
If you feed more than you use in a given year the excess is lost. (Like the dump load on the batteries.)
And you still pay the connect fee. (In the case of Sierra power in Nevada that's currently $6/month. $72/year would cover the periodic replacement costs for about 5-10 KWHr of your deep-cycle battery capacity.)
Thousands upon thousands of people have done it, and continue to do so. The economics are well-settled at this point.
Photovoltaic generation on a house-load level IS cost-effective in one situation: New construction in rural areas, where it displaces running a long (and high-priced) grid connection. Then the money that would have been spent on the grid tie can be spent on the capital cost of the photovoltaic system instead.
This one requires a grid tie for net metering, so that displacement is not available.
Unfortunately, equipment costs for grid-tied photovoltaic equipment is still high enough that you're ahead to invest the money it would have cost and spend the interest buying power for the life of the system you didn't install. (This could change with enough lowering of equipment costs or raising of electric power prices.)
If you have a source for equipment inexpensive enough to back up your claim, please let us know what it is. I have two houses where I'd LOVE to install such a system.
The plan is not feasible, with or without the multi-level scheme.
Solar installations of house-size with a net-metering grid hookup are not cost-competitive with grid power, even with government subsidies and without paying for the space under them. Otherwise people would be able to save money by doing this themselves, without the middleman and his pyramid scheme.
The difference currently is a factor of several - too large for even an exceedingly efficient company's economy of scale to overcome. It's dropping. But it's still far from crossover. (When it DOES cross over there will be efficient companies building customer-owned installations that homeowners finance with the mortgage, as part of their houses.)
This is a ponzi scheme.
Imagine a point-to-point connection where every link in the network connecting them was unsaturated, all the routers had spare CPU capacity, and there was 0% packet loss.
Imagine a world without war.
Imagine more money than you can spend.
Imagine no possessions
I wonder if you can
Nothing to kill or die for
A brotherhood of man
Imagine all the people
Sharing all the world
Or all that bandwidth so great that "every link in the network [] was unsaturated, all the routers had spare CPU capacity, and there was 0% packet loss"
Trust me:
1) Bandwidth expands to occasionally congest an installed link of any size. (The TCP throttling algorithm actually is written to do precisely that, in order to maximize throughput.)
2) The transport providers aren't going to install, and the ISPs aren't going to buy, enough backbone bandwidth to guarantee zero packet drop to all best-effort traffic, despite total saturation of all edge links to customers. (Their investors won't let them - and if they did they'd still be bankrupted by competitors who didn't, and could thus undersell them.)
3) Even if the backbone links didn't saturate, file transfers over TCP alone will (automatically) ramp up until they saturate edge links, killing the QoS needed for other services on those links unless there is QoS-driven queueing.
But if you'd actually bothered to read my previous post you'd understand that, since I made the points there.
Meanwhile, legacy services include contractual guarantees of QoS levels. In order to live up to those contractual committments while converging that traffic onto a common IP-based backbone the ISP must have similar guarantees that the designated packets will go thorough. To do that he must (first) insure that there is bandwidth to carry them and (second) insure that competing UN-guaranteed packets don't create congestion that bumps them.
He can do the first easily. But the second requires that un-guaranteed packets receive lower service levels than guaranteed packets whenever there are more packets than capacity, period.
Bland assertions that he COULD have installed enough bandwidth that the packets would never be congested won't cut it. The installed bandwidth is what it is, and any momentary burst of traffic that exceeds it can not bump the packets for the guaranteed service or the contract is violated.
So the ISP requires QoS-guaranteed service or he can't converge.
So adding QoS at the same time is just a matter of more checkboxes for the equipment manufacturers' software departments.
And of providing more crunch in the processors so software can throw some more instructions at each packet.
These whines are in fact "special consideration" pressure for the telcos to get "Net Doublecharge". They don't need service tiers, ...
I have to disagree with that.
Their legacy services are connection-oriented. They have quality of service guarantees on packet delivery that can NOT be met on a best-effort network, period.
So they have three choices:
1) Maintain separate networks.
2) Split the bandwidth on a common transport between a best-effort packet network and a connection-oriented network.
3) Pool the bandwidth in a common tranport, with the connection-oriented traffic implemented as packet streams with overriding QoS guarantees and the best-effort traffic taking all the rest.
1) Is enormously expensive. Companies that stick with that will be eaten alive by those that switch.
2) Is a big cost savings. But it means there is a bunch of bandwidth reserved for connection-oriented traffic that is not being used and thus wasted. (It's also a pain to administer and requires either more complicated software or more equipment, both of which increase costs.)
3) Is still cheaper per bit. Yet it makes all the bandwidth that isn't actually assigned to a connection at any given moment available to the best-effort network. Same or lower infrastructure cost providing more bandwidth to the best-effort network means a better price/performance ratio.
But 3) means treating some packets better than others. That's forbidden by "Network Neutrality". Oops!
Increasing the bandwidth of the backbone will NOT fix this - at least until the backbone capacity exceeds the total of the last-mile capacity - a "won't happen". (Even THEN it would still be broken across the last mile connection to your house.) In the absence of per-stream QoS enforcement by the routers the TCP streams will grab all the available bandwidth and divide it among themselves. Start some file transfers and you kill your audio and video feeds and phone line. A popular new torrent release kills EVERYBODY's audio and video feeds and phone lines, including the major networks and 911 service.
"3)" is "The Convergence". It lets the carriers put ALL their traffic on a single infrastructure, achieving enormous economies of scale and simplification of equipment and operatioin. Yes, it means the backbone must be upgraded to make QoS decisions on a packet-by-packet basis (in addition to the routing, queueing, and throttling decisions they already make). But the backbone bandwidth has to be upgraded by several orders of magnitude anyway, just to handle the currently foreseeable load of a converged infrastructure. So adding QoS at the same time is just a matter of more checkboxes for the equipment manufacturers' software departments.
IF the pressure to legislate "Network Neutrality" doesn't succeed (or at least doesn't spread beyond requiring equivalent SERVICES by different end providers to pay different rates.)
Treating the competition's packets worse than your own for a particular service is ALREADY an antitrust violation, a false-advertising issue if it wasn't explicitly warned about in advertising, and something the FCC jumps on and fines carriers for whenever it rears its ugly head. So new legislation is not required to deal with it.
I hear that much of algebra was indeed developed by Muslims - for a reason:
Both men and women owned property which wasn't pooled by marriage and on death a person's assets were divided among the spouse(s) and offspring in various prescribed proportions. With up to four wives (at a time) in a marriage, some of whom may have brought children from a previous marriage, figuring this out got very complicated despite the simple initial rules - until you had algebra to set up and solve the problems.
Short version is that defending a libel case in the states is usually easy unless the case in truly eggregious.
Conversely, brining a suit would almost certainly be disastrous if there were a hint of truth to the defamation - achieving far wider publication of the allegations while uncovering the truth behind them. So the fact that he's suing at all is a strong indication that the smear is false.
A fair point, but even a successful libel case isn't going to remove that stigma from your name.
It will for anyone familiar with the law - provided the case receives as much coverage as the original libel or slander.
Truth is an absolute defense against suits for libel or slander. So a defamed person would be a fool to sue if there was a trace of truth to the smear: Such truth would almost certainly come out, with MUCH wider coverage, at the trial. And then he'd also be on the hook for filing a bogus suit. Thus just the fact that he's suing is a strong indication that the smear is completely false.
Meanwhile, the fact of the suit replacing the smear in the Wiki entry serves as a long-lived retraction, with at least the reach of the original smear. Even old caches are automatically linked to the updated entry, while the news articles also link back and give it broader coverage. No doubt the entry will eventually be updated with the result of the suit, as well, and last for a very long time.
It's nice to see that Wikipedia's structure so strongly resists use as a defamation tool. It's good both for the project itself and for a world of potential victims.
"Thee shall know the truth, and the truth shall make you free."
Looks as though the defamatory remarks are sort of back in the article. Except now they are there as a factual statement that he is suing for someone posting those remarks. I find that ironic and funny.
The reporting of the remarks doesn't quote them. Instead it paraphrases them in a way that makes it clear that they were an over-the-top vandalism rather than having any connection to Zoeller's actual behavior.
Further, anybody who is at all familiar with the law on defamation would be aware that truth is an absolute defense. Zoeller would be a fool to file such a suit if there were any truth at all to the allegations.
So, no, I don't see any irony or humor in the entry containing the facts of the suit including a paraphrase of the remarks in question. If anything it serves as a retraction and a defense of Zoeller's character.
When was the last time you saw, say, the New York Times give as much prominence to a retraction of an erroneous story as they gave to the original story. Yet this is exactly what Wikipedia is doing.
Further, the retraction will no doubt remain posted far longer than the original smear, and (thanks to both the expanding fame of Wikipedia and the press coverage of this issue) come to the attention of far more people.
The problem lies in possible transmission errors, of which there are many.
Can you say "Forward Error Correction"? (I KNEW you could!) This stuff has been under development at LEAST since Hamming, just after the end of WW II.
Take a look at the guts of WiMax and WiFi, for just a sampling of the ways you can use a little redundancy to repair a lot of broken bits.
From there handling interplanetary distances and channel conditions is just a matter of degree, not of additional breakthroughs.
"AT&T is willing to concede that software isn't patentable, if the Court will conclude that the things that software does - the methods and procedures and instructions that a processor carries out once software is installed - are patentable.
Or for us mere developers: even if software is not patentable the lawyers will still have a way with the words to F*ck you.
I doubt the supremes would go for that.
The law explicitly exempts "mathematical algorithms" from patentability. But a "mathematical algorithm" is just a set of instructions for performing a mathematical computation. AT&T's argument would make it possible to patent performing the computation. If that's true, what was Congress saying when they blocked patenting mathematical algorithms?
"It's OK to contemplate the algorithm while meditating but not to actually USE it." No, I don't think so. And the Supremes are the country's experts at seeing through specious legal arguments.
If they rule in favor of Microsoft, sure there is a lot of different rulings they could make, but it still comes down to one of two things: (1) They say software patents are invalid, the issue is moot, case dismissed, or (2) software patents are valid, but here's why we ruled for Microsoft.
I wish it were that way. But (2) would likely come out this way:
"We ruled for Microsoft because of this other reason, so we didn't have to resolve whether software patents are valid."
Being NAL, I don't recall the pseudo-Latin legalize for it. But I understand that one of the court's principles is to avoid making a decision that shakes things up unless it's the only way to settle a case.
(Of course they might still pick patentability of software as the reason to rule - especially if they decided the Federal District had egregiously ignored or misinterpreted their previous rulings, in order to clean up the mess and/or reassert their authority. But they don't necessarily have to settle the issue if they rule for Microsoft on some other grounds.)
Perhaps Steve Ballmer ought to have checked with legal before mouthing off about Linux and intellectual property yesterday?
Maybe legal WANTS him to mouth off about Linux and IP.
Just think: If he raises enough stink about how Microsoft is ready to use patents to break Linux and other FOSS projects if software IS held to be patentable, and the justices hear about it...
%LTB-)
But I guess lifesaving medical research, including drug design, doesn't qualify as "noteworthy". B-(
Once you have an IP address, you look up who owns it and you call them. They do their research, looking at things such as DNS records, DHCP assignments, DSLAM logs, etc... They then look up which customer that was, and there you go.
Note that the bad guy may be using a dynamic IP address assigned through DHCP - especially if the new possessor of the laptop is using a dialup ISP, or if somebody just plugs it in at work. If so you also need the time at which the IP address was captured, so the owner of the IP can figure out which of the users who happened to get that IP address was the culprit.
"I always knew that a geek would make a great husband." Cheer up bunky, it could happen to a 'Dotter. Some day. The odds are certainly no worse than finding, say, extraterrestrial life.
Happened to me.
Pre-slashdot, though. We first met on a netnews group but didn't really hit it off until introduced in person by a friend at a party.
Still married, too. Despite losing most of the first nest egg in the internet bubble burst.
That anyone who doesn't like the inefficiency of voting shouldn't be allowed to vote? Voting should *not* take an hour.
I am saying that voting has a very important purpose (preventing civil war) and making it too convenient defeats the purpose (increasing the risk of civil war).
But if you prefer intermittent civil war to waiting in line to vote, feel free to vote for easier voting.
I had to wait in line for over an hour just to cast my ballot. I had to stand outside (in South Florida), listen to old people bitch about everything around them, and then get harassed by the security guards---all to vote in a free election.
Suppose the issues were so important that, in the absence of the election, you'd have volunteered to fight in a civil war over them. How much discomfort would you endure to fight for your beliefs in that war?
Elections are NOT about being nice, or being fair, or because there's some magic reason why (50%+1) voters should run the lives of the other (50%-1). Elections are about convincing the losers that they can't reverse the decision - on the issues, or on picking decision-makers - by going to war to overthrow the government and install their party/dictator. This stabilizes the republic.
That's why, until a few years ago in the last canton to repeal the requirement, the Swiss carried swords to their elections: They used to vote by raising them to be counted. That's why, in the US, the franchise was originally only for the classes of people who were the major fighters in the revolution (property owners and merchants), and was extended to one group after another as they proved themselves able to engage in organized violence to advance their political agendas.
That's why (until recently), registration was about as hard and inconvenient as signing up with a militia, and voting was about as hard and inconvenient as showing up for a muster.
Does the outcome matter to you, enough to fight a war over it? If so, it should matter enough to stand in line for an hour or two. If not, you shouldn't vote.
Now if voting is made harder for some than for others you have a gripe: Such bias lowers the accuracy of the model - until it may NOT predict the outcome of the civil war, and thus not prevent it. Such distortions need to be hunted down and eliminated.
But making it so easy that masses of people who would never be bothered to fight still vote on issues that don't bother them, and count as much as smaller groups of people who really care and are chomping at the bit to promote their agenda - by violence if necessary, also biases the model and destabilizes the country. So it's a really bad idea.
B-(
In hydrocarbons burning the hydrogen provides most of the energy. Burning the carbon provides some, but the carbon is mainly useful for packing the hydrogen in a form more dense than H2 gas for convenient storage and handling.
As hydrocarbons go, CH4 has a higher ratio of hydrogen to carbon than any other molecule: Every bond on every carbon holds a hydrogen, none are "wasted" connecting to other carbons.
So if you're going to burn hydrocarbons for energy, methane releases the least CO2 for a given amount of energy produced.
It seems to me that such things can easily be replaced by a small heating element, perhaps a small ceramic one like one uses for gas ovens or a foil battery warmer. Not as cheap as a lightbulb but certainly would outlast one by decades.
You really don't understand the situation out in the sticks.
- Things have to be maintained - and usually built - using supplies available at rural stores.
- Even the rural stores may be an hour or two away by road in the best of weather, and inaccessible for days in the winter. The nearest city may be a small town, too, with something substantial enough to have a suitable industrial supply maybe a couple hundred mile round trip. Add the gas price and the opportunity cost of spending a day driving just to get a part...
When was the last time you saw a "small ceramic heating element" available in a mom-and-pop grocery? When was the last time you DIDN'T find light bulbs in one - especially if it was the only store for 40 miles?
You also don't want the type used in gas ovens: In addition to the high drain, they reach a temperature suitable for igniting gas (or anything else flammable that comes in contact with them. That's not what you want in the battery shed when the wind is blowing and the windcharger has topped of the batteries and the dump load controller has decided to give them an equalizing charge - so they're now putting out enough hydrogen gas to create an explosion if the venting got clogged by a snowdrift or silverthaw.
Fluorescents aren't suitable: In addition to having trouble starting at low temperatures (exactly when you need them), the last thing you want about boot-toe high is a target that will shatter easily, dumping mercury in the pumphouse for your place's water - right next to the well, within the circle that must be kept clean of toxics due to rain seeping down. (Bulbs take a lot to break them.