Although this isn't really an exception. There were some much more exciting games that came out later for the Atari 800, some of which included staggeringly good graphics considering the limitations of that machine. The game developers definitely learned how to push that machine to its limits over the product's lifecycle.
One qualification to this otherwise good post: more recent research shows that the distinction between simple and complex carbohydrates is less clear-cut than was previously thought. Some foods containing complex carbohydrates have a much more rapid impact on blood sugar than others. Things that make the food harder to digest (like fiber) tend to slow down the digestion and reduce the sugar surge. The impact of foods on blood sugar is characterized by glycemic index and glycemic load, which have been measured for a wide variety of foods in several research studies. Generally white bread, pasta, and rice cause a much stronger sugar surge in the bloodstream than wholegrain bread, brown rice, etc. Fruit may not be as bad as you think, because the high fiber content slows down the
sugar surge. An apple has a glycemic load of 4, vs. 10 for a piece of white bread, 8 for whole wheat bread, or 17 for a doughnut. (20 on this scale is very high.) Note that many websites use glycemic index rather than glycemic load. The link I give above explains the difference.
The real impact of this on diet and weight is less clear. Some have taken this new research as compelling evidence that carbohydrates are bad and should be avoided. Other nutritionists are skeptical of this position. The truth probably lies somewhere in between--North Americans probably eat too many carbohydrates, and too many of the ones we eat are of the kind that is rapidly processed into blood sugar (e.g. white bread instead of whole wheat).
Note that, by law, the ingredients are listed by quantity, from most to least. (I'm not sure if it's by weight or by volume.) This means that the filling is primarily corn syrup and sugars. There is less strawberry in that filling than each of the three kinds of sugar listed, and less of each of those than corn syrup. Note that while artificial and natural flavors and red dye #40 are way down the list, it doesn't take much of these to give the red color and the nice strawberry flavor. Altogether, the filling is best characterized as synthetically-flavored sugar syrup. They add a tiny amount of strawberry and apple so that they can claim on the packaging that it contains real fruit, without increasing the cost too much.
If you still have it handy, check out the nutrition information box on the package. Does one serving contain a measurable amount of fiber? I have run into products made with 'whole wheat' that somehow managed not to have even a gram of fiber in them. I'm not sure how they manage that.
Nutrigrain bars are basically vitamin-enriched cookies. They are probably better for you than a regular cookie, or a donut, but they don't really qualify as healthy food either.
Re:reason for, reason not for
on
Blank Keyboard
·
· Score: 1
I think that typing should be mandatory. With the prevalence of computers in the workplace, typing is just too useful a skill to allow people to graduate without it. Not everybody needs to be able to type 100 words per minute, but pretty much everyone should be able to bang something out at 20-30 wpm.
I call B.S. That's not how compression works. Yes, the phones are probably optimized for the best possible voice reproduction in a limited bandwidth. No, that does not mean that music is not also transmitted.
Have you ever been put on hold? Did they play music? Case closed.
Actually, I'm not vegetarian, but I often buy various kinds of veggie burgers because I like them. They aren't the same as a 'real' burger (which I also like), but they have their own merits.
There was even at least one that could wipe the BIOS eproms, leaving the computer completely inoperable and difficult to repair if not outright irreparable.
I respect your expertise on the role of rational decision making vs. quick reaction in piloting. I was actually not so interested in the specific comment about an AI pilot, but more interested in the larger issue of whether a purely 'rational' AI would be superior in general to one that has emotions. My impression is that recent psychological research is pointing in the direction that the brain's emotional and rational functions are very much intertwined--that our emotions play a role in our decision making even when we think we are being rational, and that this is beneficial. Emotions are part of that system that allows us to quickly coalesce information from many sources and build patterns. Panic is clearly not good (when flying a plane or otherwise), but it seems to me that our emotions do play a role in general in helping us process and prioritize the information we receive from the world around us.
As an aside, you wrote "that's why when the problem space and information representation is clearly defined, a computer is always a better decision maker than a human." Unfortunately, a computer that depends on a clearly defined problem space and information representation will never be a good AI. Decision-making in the real world very often (maybe almost always) involves making a choice (rational or otherwise) given incomplete and possibly inaccurate information. This has been much studied in AI development, particularly using techniques like Bayesian logic to find the optimum ('rational') decision based on incomplete information.
Hmm...but what if the AI is a thrillseeker? Suicidal? Psychotic? What if it suddenly develops acrophobia? If we're going to have a true AI with emotions, these are issues that need to be addressed, don't you think?
Admittedly, one wouldn't want an overemotional AI. He probably is on the right track, though, in suggesting that a real AI would have emotions. There is some pretty good evidence that emotions are an important part of human intelligence and that we would function much more poorly without them. In particular, emotions help us make reasonable decisions fast. Sometimes it's much better to make a pretty good decision on the spot than to take the time to rationally analyze a situation and make the best decision a minute or two later. Especially if a tiger is about to eat your face or your plane is about to crash.
At some level, though, it becomes just a cost/benefit calculation. The kind of security you describe sounds like it would be expensive to implement. Companies might well decide to stick with the tried and true approach of not implementing such a scheme, taking whatever fines they get when a leak occurs, and arranging {fines|lawsuits|jail} for the employees responsible. While not perfect, this approach may well be almost as effective and cheaper in the long run.
Except that DrunkenBatman is distributing the sourcecode he received under the GPL. He is probably perfectly OK to redistribute that sourcecode as received with the missing copyright notices. The GPL violation occurred at MXS. DrunkenBatman is not liable for that.
Remember that Columbus had a plan to make money before going on his little trip.
Yah, but ironically his plan was complete vaporware. He had no hope of reaching India that way. The Earth's diameter was much larger than he estimated.
Actually, it's not. IANAL, but I believe trademarks are normally only effective within a particular area of business. A company with a trademarked name for a food product cannot necessarily prevent other companies from trademarking the same name for another type of product. The key is that the products need to be different enough that consumers would not be confused into thinking the two companies are related. There is an exception in the U.S. for 'famous' brands. Trademarks like McDonald's or Ford automatically cover all areas of business.
It's not so much a random number seeded by the fingerprint, as it is a hash of the fingerprint. Security of hashed personal data is an issue, the same way that security of a hashed password file is an issue. Yes, you can't reconstruct the original passwords from the hashed values, but if an attacker has the hashed values there are ways to compromise the system's security.
In particular, someone with access to a true fingerprint database (i.e. police/FBI) should be able to apply the same 15-point process to it and generate numbers that can be matched against the library 'bar codes'. The fact that the 'bar codes' do not encode the entire fingerprint does not really do much to increase privacy protection.
I mostly agree with you, but I don't think it's all the VOIP providers' fault. It does appear that in some areas the local phone companies have deliberately made it hard for the VOIP providers to hook into the 911 system. Hopefully the new FCC ruling will remedy that. From the article it does sound like they have ordered the phone companies to cooperate.
Give it time. The grandparent is right. In a decade or two there is not going to be any conventional landline service. Everyone is going to have a big, fat data pipe to their home that has five nines uptime and huge bandwidth, and it will carry the home's internet, voice, and television. Data is data, and there is no more reason to have multiple data lines coming in than there is to have separate water lines for the kitchen and the bathroom.
This 'pipe' may be a fiber connected to what used to be a phone company, or a coax cable connected to what used to be a cable TV company. It won't matter. They will all be the same.
Don't judge tomorrow's technology based on the limitations of today's.
The scene you mentioned between Gandalf and Pippin was good, but there is good reason why it wasn't in the book. It's clear from Tolkien's books that Gandalf has no idea what it is like for a human (or a hobbit) to die. The text for the scene comes from the description in the book of Frodo's journey into the West. Ringbearers aside, humans (and hobbits) do not normally go there when they die. Only elves are allowed to sail to Valinor. The elves and the demi-gods like Gandalf do not know what happens to humans when they die. It is implied that humans move on to some sort of afterlife which is beyond their comprehension, so that the fate of humans to die after a short span on earth is in fact a gift, while the elves' virtual immortality ultimately leads to stagnation.
(Yes, off-topic and irrelevant to your point, but I couldn't resist.)
Unfortunately, many users didn't go find Firefox once. They had someone more technically oriented install it for them.
The fact that Firefox security updates don't automatically install unless you notice and click on that red arrow in the upper right corner pretty much guarantees that a large fraction of copies will remain unpatched. When I've visited people for whom I installed Firefox 1.0 when it came out, I've noticed that none of them have noticed the red update icon or updated Firefox on their own.
If users have to go and get updates, many machines will remain vulnerable to security holes.
The problem is that, fundamentally, there is no difference between the two. My phone is plugged into my cable modem. So is my computer. What's the difference between the two? I can get voice on either.
Of course, you do have to factor in the 15-20MB of Firefox updates to fix security holes. We're up to 1.03 now, with 1.04 coming out in a couple days, and every update requires downloading the whole browser again from scratch. The MSIE updates have better automation too.
I'm guessing you have broadband. The slashdot rendering problem only shows up on dialup connections. The display bug shows up when the page is rerendered multiple times during the download, which only happens on a slower connection.
Yes, I am a Firefox user and hate IE, but nobody gains from having a false picture of the merits of the two browsers.
Slashdot Mirror
No, coffee shops provide coffee so you will give them money for it. Everything else is ancillary.
Although this isn't really an exception. There were some much more exciting games that came out later for the Atari 800, some of which included staggeringly good graphics considering the limitations of that machine. The game developers definitely learned how to push that machine to its limits over the product's lifecycle.
The real impact of this on diet and weight is less clear. Some have taken this new research as compelling evidence that carbohydrates are bad and should be avoided. Other nutritionists are skeptical of this position. The truth probably lies somewhere in between--North Americans probably eat too many carbohydrates, and too many of the ones we eat are of the kind that is rapidly processed into blood sugar (e.g. white bread instead of whole wheat).
If you still have it handy, check out the nutrition information box on the package. Does one serving contain a measurable amount of fiber? I have run into products made with 'whole wheat' that somehow managed not to have even a gram of fiber in them. I'm not sure how they manage that.
Nutrigrain bars are basically vitamin-enriched cookies. They are probably better for you than a regular cookie, or a donut, but they don't really qualify as healthy food either.
I think that typing should be mandatory. With the prevalence of computers in the workplace, typing is just too useful a skill to allow people to graduate without it. Not everybody needs to be able to type 100 words per minute, but pretty much everyone should be able to bang something out at 20-30 wpm.
Have you ever been put on hold? Did they play music? Case closed.
Actually, I'm not vegetarian, but I often buy various kinds of veggie burgers because I like them. They aren't the same as a 'real' burger (which I also like), but they have their own merits.
There was even at least one that could wipe the BIOS eproms, leaving the computer completely inoperable and difficult to repair if not outright irreparable.
Virgin Mobile has had this feature on their cell phones for some time.
As an aside, you wrote "that's why when the problem space and information representation is clearly defined, a computer is always a better decision maker than a human." Unfortunately, a computer that depends on a clearly defined problem space and information representation will never be a good AI. Decision-making in the real world very often (maybe almost always) involves making a choice (rational or otherwise) given incomplete and possibly inaccurate information. This has been much studied in AI development, particularly using techniques like Bayesian logic to find the optimum ('rational') decision based on incomplete information.
Admittedly, one wouldn't want an overemotional AI. He probably is on the right track, though, in suggesting that a real AI would have emotions. There is some pretty good evidence that emotions are an important part of human intelligence and that we would function much more poorly without them. In particular, emotions help us make reasonable decisions fast. Sometimes it's much better to make a pretty good decision on the spot than to take the time to rationally analyze a situation and make the best decision a minute or two later. Especially if a tiger is about to eat your face or your plane is about to crash.
At some level, though, it becomes just a cost/benefit calculation. The kind of security you describe sounds like it would be expensive to implement. Companies might well decide to stick with the tried and true approach of not implementing such a scheme, taking whatever fines they get when a leak occurs, and arranging {fines|lawsuits|jail} for the employees responsible. While not perfect, this approach may well be almost as effective and cheaper in the long run.
Except that DrunkenBatman is distributing the sourcecode he received under the GPL. He is probably perfectly OK to redistribute that sourcecode as received with the missing copyright notices. The GPL violation occurred at MXS. DrunkenBatman is not liable for that.
Yah, but ironically his plan was complete vaporware. He had no hope of reaching India that way. The Earth's diameter was much larger than he estimated.
Actually, it's not. IANAL, but I believe trademarks are normally only effective within a particular area of business. A company with a trademarked name for a food product cannot necessarily prevent other companies from trademarking the same name for another type of product. The key is that the products need to be different enough that consumers would not be confused into thinking the two companies are related. There is an exception in the U.S. for 'famous' brands. Trademarks like McDonald's or Ford automatically cover all areas of business.
It's not so much a random number seeded by the fingerprint, as it is a hash of the fingerprint. Security of hashed personal data is an issue, the same way that security of a hashed password file is an issue. Yes, you can't reconstruct the original passwords from the hashed values, but if an attacker has the hashed values there are ways to compromise the system's security. In particular, someone with access to a true fingerprint database (i.e. police/FBI) should be able to apply the same 15-point process to it and generate numbers that can be matched against the library 'bar codes'. The fact that the 'bar codes' do not encode the entire fingerprint does not really do much to increase privacy protection.
I mostly agree with you, but I don't think it's all the VOIP providers' fault. It does appear that in some areas the local phone companies have deliberately made it hard for the VOIP providers to hook into the 911 system. Hopefully the new FCC ruling will remedy that. From the article it does sound like they have ordered the phone companies to cooperate.
In many places, the police etc. will tell you not to do this.
This 'pipe' may be a fiber connected to what used to be a phone company, or a coax cable connected to what used to be a cable TV company. It won't matter. They will all be the same.
Don't judge tomorrow's technology based on the limitations of today's.
(Yes, off-topic and irrelevant to your point, but I couldn't resist.)
The close gesture does not cause crashes for me. Maybe some quirk with your system?
Unfortunately, not.
The fact that Firefox security updates don't automatically install unless you notice and click on that red arrow in the upper right corner pretty much guarantees that a large fraction of copies will remain unpatched. When I've visited people for whom I installed Firefox 1.0 when it came out, I've noticed that none of them have noticed the red update icon or updated Firefox on their own.
If users have to go and get updates, many machines will remain vulnerable to security holes.
The problem is that, fundamentally, there is no difference between the two. My phone is plugged into my cable modem. So is my computer. What's the difference between the two? I can get voice on either.
I'm guessing you have broadband. The slashdot rendering problem only shows up on dialup connections. The display bug shows up when the page is rerendered multiple times during the download, which only happens on a slower connection.
Yes, I am a Firefox user and hate IE, but nobody gains from having a false picture of the merits of the two browsers.