Was Apple really the first to place the keys at the top of the open clamshell, and the pointing device on the lower half under your thumbs? I used a couple of PC notebooks before getting access to a Powerbook, and remember thinking "now this makes *&^%$#@! sense".
It is rare for a heat wave in the US to be blamed for more than 20 deaths. The worst one that turned up in a casual Google search was 1936, where 5000+ people died. That was before the wide availability of air conditioning.
So employers must do one of two things: 1. Pay more. 2. Train more.
Training and paying market-plus wages have to both be considered a long-term investment in the employee. Businesses are reluctant to train because a trained employee can leave and take that training to the competition; the way to stop that is to compensate trained, valuable employees well enough that they don't want to leave.
Companies that make assistive devices like this know:
* Their customer is actually the government agency that pays them
* The customer will pay an incredible premium for a single-purpose device instead of software on a general-purpose device
That last one is important. In some states, agencies are precluded from giving their clients laptops with, say, scanners for the blind or textspeech software for the deaf, because the client might use the device to do things other than handle their disability.
The Watson model looks a lot like how expert systems were supposed to work back in the 1970's and 80's. Both of them get high-level performance at specific tasks out of a computer system by encoding expert knowledge and drawing inferences from it.
Watson has several big advantages over previous expert systems work:
* It has a lot more data available * It reasons probabilistically from that data, so its conclusions are less brittle * The data starts out mostly as raw text, so it's easier to update * Watson can deliver results via the Internet
Those last two are actually the biggest win for Watson. What killed early expert systems was the maintenance effort required to keep them up to date, both in improving the knowledge base and in distributing it to users. Having Watson-like services delivered via the internet makes that maintenance much easier.
A language/methodology can catch on fast if it is the price of entry to a hot field. Broad, long-term popularity requires finding applications outside the original field.
Objective C is the current price of entry for iOS development. It will stay around as long as Apple allows no alternatives, but is unlikely to grow outside that niche.
OOP was the price of entry for GUI toolkits. It has since proved its worth as a general structure and analysis method.
Java started out as the price of entry for applets. It eventually settled in as a portable platform for enterprisey stuff.
Lisp was the price of entry for AI. It died back when strong AI faltered.
Functional programming may yet become the price of entry for reliable concurrent programming.
Consider the results of an experiment I first saw described in Peopleware (scroll down to "Creative Space"). The researchers compared performance at Fortran programming between people in quiet rooms and people in rooms with music. The good news is that performance was about the same. The bad news was:
There was a hidden wildcard. The specification required an output data stream be formed through a series of manipulations on numbers in the input data stream. Although unspecified, the net effect of all the operations was that each output number was equal to its input number. Of those students who figured this out, the overwhelming majority came from the quiet room.
The part of your brain that listens to music is apparently also the part that notices odd things in your code, and it can't do two things at once.
One could argue that romance novels are the female equivalent of porn - they present an unrealistic view of men that is unlikely to be found in the real world, and women who read a lot of them are setting themselves up for relationship failure.
Stoll was an individual, with few resources and no authority to require information from anyone. DHS is a large well-funded national agency with serious authority.
They should have left that intrusion alone just long enough to get it traced.
As part of a non-profit, Plesko could not comment specifically on CISPA, which would, as currently drafted, allow companies to share much richer and more individualized data directly with the government. “We get network data,” says Plesko. “Not PII (personally identifiable information).”
That means the NCFTA can pass along information, for example, about suspicious servers or IP addresses and content from spear-phishing emails that companies are seeing in their networks, but not the names or addresses of those who appear to be affiliated with the schemes.
If NCFTA is restricting itself to data like that, I have no problem with it. Problem is, without oversight we can't be sure they really are restricting themselves to that.
I'd like to see privacy-by-default become the norm with personal data. Right now the default is usually "we can share your data arbitrarily unless you opt out, and you have to renew the option every time we change our privacy policy or it goes back to share-with-anyone".
Which is wonderful for the businesses, but sucks for users.
If only to give x86 a swift kick in the bits. These days, the x86 standard serves primarily as a barrier to entry - anyone can pick up open-source tools and write sort-of shippable code, but to get the last factor of two in performance, you either have to be Intel, or buy Intel's compiler and know a lot about configuring it and your code to mesh right.
Science is supposed to generate results that are not contradicted by experiment. Kuhn aside, it is not done purely by popular vote - if one guy publishes a repeatable experiment that differs from current consensus, eventually other guys will repeat it, publish as well, and the consensus will change. Science has the luxury of being able to say "we don't know enough now to decide" and "we used to believe X, but experiment contradicts that, so now we believe Y".
Law is supposed to generate results that are acceptable to the majority of citizens. It has to make decisions with evidence that is often questionable or missing. It tries to be long-term internally consistent (adherence to precedent). The gold standard in law is a jury decision - a sample of the population looks at the evidence and makes a choice, which often comes down to whose lawyer's story sounded better to them.
Lawyers and the law in general are only interested in science when it supports their primacy in running society. I will believe otherwise when juries are routinely told that eyewitenss identification is unreliable - in some states in the US, it is illegal to mention this well-established fact in front of a jury.
A dam in the Fukushima prefecture of Japan was breached following the recent earthquake and tsunamis which have devastated the country. According to media reports, the dam broke on Friday, with a wall of water washing away 1800 homes downstream.
An amazing amount of ambiguity and crap in the tax code would go away if the Government were required to publish a program in Java (probably best balance of portability, capability, and specification) and that program WAS the definition of the tax code.
This would have the nice side effect of keeping lawyers who can't think formally (in the mathematical sense) away from tax law.
The only requirements are that a customer's account must be in good standing, their device cannot be associated with a current and active term commitment on an AT&T customer account, and they need to have fulfilled their contract term, upgraded under one of our upgrade policies or paid an early termination fee.
I have an ancient 3G recently superseded by a 4S - I want it unlocked on general principles, and will ask them to do it sometime soon.
Slashdot Mirror
But we'll drop AT&T like a hot rock the moment they cancel our grandfathered $30/month data plan.
Was Apple really the first to place the keys at the top of the open clamshell, and the pointing device on the lower half under your thumbs? I used a couple of PC notebooks before getting access to a Powerbook, and remember thinking "now this makes *&^%$#@! sense".
... for ages.
It is rare for a heat wave in the US to be blamed for more than 20 deaths. The worst one that turned up in a casual Google search was 1936, where 5000+ people died. That was before the wide availability of air conditioning.
In 2003, a heat wave in Europe killed 70,000.
Europeans can complain about US infrastructure when theirs gets within an order of magnitude of ours at preserving citizens' lives.
So employers must do one of two things:
1. Pay more.
2. Train more.
Training and paying market-plus wages have to both be considered a long-term investment in the employee. Businesses are reluctant to train because a trained employee can leave and take that training to the competition; the way to stop that is to compensate trained, valuable employees well enough that they don't want to leave.
and really evil lawyers.
Intel will have to compete without its traditional legacy compatibility advantage?
They'll hate it, but they'll probably suck it up, go do some real chip engineering, and at least catch up on the useful-cycles-per-watt front.
After all, they did basically that to the PowerPC.
Companies that make assistive devices like this know:
* Their customer is actually the government agency that pays them
* The customer will pay an incredible premium for a single-purpose device instead of software on a general-purpose device
That last one is important. In some states, agencies are precluded from giving their clients laptops with, say, scanners for the blind or textspeech software for the deaf, because the client might use the device to do things other than handle their disability.
And I say this as a long-time Mac user. Bumping the Retina Mac Book Pro from 8GB to 16 GB RAM costs $200.
I know the RAM in this machine is very new, but 8GB of standard laptop RAM is currently around $50 - $75.
If adding 8GB cost $100, I would agree with your analysis. $200 is standard highway robbery for Apple RAM.
The Watson model looks a lot like how expert systems were supposed to work back in the 1970's and 80's. Both of them get high-level performance at specific tasks out of a computer system by encoding expert knowledge and drawing inferences from it.
Watson has several big advantages over previous expert systems work:
* It has a lot more data available
* It reasons probabilistically from that data, so its conclusions are less brittle
* The data starts out mostly as raw text, so it's easier to update
* Watson can deliver results via the Internet
Those last two are actually the biggest win for Watson. What killed early expert systems was the maintenance effort required to keep them up to date, both in improving the knowledge base and in distributing it to users. Having Watson-like services delivered via the internet makes that maintenance much easier.
A language/methodology can catch on fast if it is the price of entry to a hot field. Broad, long-term popularity requires finding applications outside the original field.
Objective C is the current price of entry for iOS development. It will stay around as long as Apple allows no alternatives, but is unlikely to grow outside that niche.
OOP was the price of entry for GUI toolkits. It has since proved its worth as a general structure and analysis method.
Java started out as the price of entry for applets. It eventually settled in as a portable platform for enterprisey stuff.
Lisp was the price of entry for AI. It died back when strong AI faltered.
Functional programming may yet become the price of entry for reliable concurrent programming.
Consider the results of an experiment I first saw described in Peopleware (scroll down to "Creative Space"). The researchers compared performance at Fortran programming between people in quiet rooms and people in rooms with music. The good news is that performance was about the same. The bad news was:
There was a hidden wildcard. The specification required an output data stream be formed through a series of manipulations on numbers in the input data stream. Although unspecified, the net effect of all the operations was that each output number was equal to its input number. Of those students who figured this out, the overwhelming majority came from the quiet room.
The part of your brain that listens to music is apparently also the part that notices odd things in your code, and it can't do two things at once.
One could argue that romance novels are the female equivalent of porn - they present an unrealistic view of men that is unlikely to be found in the real world, and women who read a lot of them are setting themselves up for relationship failure.
With a big fossil fuel plant or a hydro dam, you get 90% or more of the rated output, and outages are rare.
With a big solar PV plant, you get 1/3 to 1/4 the rated maximum output, and outages happen nightly
With a big wind farm, you get 30% of rated maximum, if you're lucky, and long periods of calm happen more frequently than you'd like.
For example, England has been working on wind power for the last decade or so. The results have not been good.
... judging from the comments I see there.
One with a GPS jamming detector.
"Oh we're sorry - our missile wandered into your GPS jamming zone and lost its way..."
Stoll was an individual, with few resources and no authority to require information from anyone. DHS is a large well-funded national agency with serious authority.
They should have left that intrusion alone just long enough to get it traced.
Is there potential trouble if a language has a ISO/ANSI/IEEE standards document? Those typically have copyright assigned to the standards body...
From TFA:
As part of a non-profit, Plesko could not comment specifically on CISPA, which would, as currently drafted, allow companies to share much richer and more individualized data directly with the government. “We get network data,” says Plesko. “Not PII (personally identifiable information).”
That means the NCFTA can pass along information, for example, about suspicious servers or IP addresses and content from spear-phishing emails that companies are seeing in their networks, but not the names or addresses of those who appear to be affiliated with the schemes.
If NCFTA is restricting itself to data like that, I have no problem with it. Problem is, without oversight we can't be sure they really are restricting themselves to that.
I'd like to see privacy-by-default become the norm with personal data. Right now the default is usually "we can share your data arbitrarily unless you opt out, and you have to renew the option every time we change our privacy policy or it goes back to share-with-anyone".
Which is wonderful for the businesses, but sucks for users.
If only to give x86 a swift kick in the bits. These days, the x86 standard serves primarily as a barrier to entry - anyone can pick up open-source tools and write sort-of shippable code, but to get the last factor of two in performance, you either have to be Intel, or buy Intel's compiler and know a lot about configuring it and your code to mesh right.
Science is reality. Law is verisimilitude.
Science is supposed to generate results that are not contradicted by experiment. Kuhn aside, it is not done purely by popular vote - if one guy publishes a repeatable experiment that differs from current consensus, eventually other guys will repeat it, publish as well, and the consensus will change. Science has the luxury of being able to say "we don't know enough now to decide" and "we used to believe X, but experiment contradicts that, so now we believe Y".
Law is supposed to generate results that are acceptable to the majority of citizens. It has to make decisions with evidence that is often questionable or missing. It tries to be long-term internally consistent (adherence to precedent). The gold standard in law is a jury decision - a sample of the population looks at the evidence and makes a choice, which often comes down to whose lawyer's story sounded better to them.
Lawyers and the law in general are only interested in science when it supports their primacy in running society. I will believe otherwise when juries are routinely told that eyewitenss identification is unreliable - in some states in the US, it is illegal to mention this well-established fact in front of a jury.
Dam breaches following Japan earthquake
A dam in the Fukushima prefecture of Japan was breached following the recent earthquake and tsunamis which have devastated the country.
According to media reports, the dam broke on Friday, with a wall of water washing away 1800 homes downstream.
Casual web searches don't turn up anything other than exploiting a vulnerability to get onto your machine (bad enough!).
Did anyone successfully command it to do anything?
An amazing amount of ambiguity and crap in the tax code would go away if the Government were required to publish a program in Java (probably best balance of portability, capability, and specification) and that program WAS the definition of the tax code.
This would have the nice side effect of keeping lawyers who can't think formally (in the mathematical sense) away from tax law.
From the AT&T announcement:
The only requirements are that a customer's account must be in good standing, their device cannot be associated with a current and active term commitment on an AT&T customer account, and they need to have fulfilled their contract term, upgraded under one of our upgrade policies or paid an early termination fee.
I have an ancient 3G recently superseded by a 4S - I want it unlocked on general principles, and will ask them to do it sometime soon.