I sorta meant there could be a shift in engines to diesel engines, which would then replace gasoline engines, and gasoline fuel...
In contemplating the greenhouse gas problem (or our budget deficit for energy), its quite a pickle. Cars burn predominantly gas. Gas requires oil - petroleum oil - for production. The USA doesn't make enough oil, so we necessarily send a LOT of US dollars to OPEC each year, and many OPEC members are openly antagonistic towards the USA.
Do we really want to give our enemies a lot of money?
The ONLY reasonable ways out are 1) electric cars 2) biodiesel
Electric cars don't have enough power or range (and if they gain power and range, they gain cost and weight and lose efficiency). However, electric is more efficient. But the energy would still need to come from somewhere, most of the US power plants that would provide the energy for electric cars still burn hydrocarbons, but more importantly, consumers have not taken to them,
Biodiesel costs $1 more per gallon - a HUGE 37% decrease in efficiency. Unless oil stays over $70/bl, then it comes close to break even. Biodiesel creates a LOT of USA jobs in farming and manufacturing. Biodiesel doesn't increase greenhouse gas emissions.
What does the future hold for energy? The gas car trap is a REAL problem for the USA. Biodiesel is one way out. Electric cars and more nuclear plants is the most probable other solution.
Frankly, I like the biodiesel solution, but it may be cheaper in the short-term to continue overthrowing Middle East governments. I wish there was a sarcastic overtone to that, but it is more like a grudging acceptance of our current reality.
If oil is over $70/barrel it is not so far fetched. Not the yellow grease based biodiesel, which is very limited in scope, but a B100 biodiesel.
If you factor in the cost of emissions...B100 biodiesel has NO net emissions, and petroleum based diesel has significant emissions costs....
Factor in the cost of emissions on environmental destruction at $1/gallon of fuel and keep oil over $70/barrel, and America has a new soybean based economy.
I don't think that is really gonna happen, but there are not so unrealistic contingencies that would bring it to pass. Like if the US government ever really starts to care about global warming and at the same time the oil fields in Iraq and Iran both get incapacitated.
There's scads of papers on biodiesel, its efficiency, and cost.
If petroleum goes up in price a bunch more, biodiesel gets to break even.
The unmapped territory is that although it burns a hydrocarbon, 100% biodiesel doesn't increase atmospheric CO2, because that CO2 was removed from the atmosphere less than a year prior. It is cyclic in the short-term. Biodiesel could be a near drop-in replacement for gasoline in cars and solve greenhouse gas problems from automobiles. Of course, if you use peanuts instead of soybeans, and oil costs stay high....people bet billions on shifts like this, the shift to biodiesel would become reality if regular diesel wholesale prices get too high and we have a strong need to minimize emissions...both of which are very real scenarios. Both factors have shifted a lot since this white paper in 2002.
This seems like the sort of thing that could happen if the brain's speech area's neurons somehow became trained to stop delivering impulses for "normal" speech. In this case, it would be theoretically possible to train the network back to normal levels. Of course, it could be something completely different.
Spasmodic dysphonia is a form of focal dystonia. This neurological condition occurs when sensory inputs, and motor outputs, get 'confused'. There is a necessity for motor programs to operate in opposition - flexors and extensors mutually inhibit. So, when you flex, the extensor muscles are not contracting at the same time, and vice versa. Somehow, in focal dystonia, the motor programs that cannot conflict if one is to have normal motor control, do conflict.
The good news is that at least for some forms of focal dystonia, effective treatments are now available from one of my former colleagues. Arch Phys Med Rehabil. 2003 Oct;84(10):1505-14.
And yes, I've contributed as well Somatosens Mot Res. 2002;19(4):347-57
I'm not familiar with the details of focal dystonia of the vocal tracts, but I suspect that an intelligent otolaryngologist could devise a training strategy that would restore normal speech....
Prior work has shown a very strong relationship between the number of hours of TV averaged per day before age two, and incidence of ADD. In fact, if TV were a causative factor, this would explain the vast majority of ADD cases, because the correlation was quite high.
There is also recent work from Israel showing that increased father age at conception is a huge correlational risk factor for autism.
There is also the recommendation from professional pediatric societies. They have consistently recommended, for years and years and years, that there is no recommended amount of daily TV watching for children, and in fact there should be NO TV watching for children under 3.
The reasons are simple. A child, before age 3, will interact with everything in its environment. This is a learning process. If you watch a 2 year old in front of Barney, they turn into a statue. Motionless, completely non-interactive, they are simply mesmerized. If you do not interact, a huge source for learning is lost. Now, not all two year olds will sit motionless in front of a TV. But a lot will. And if that results in parental feedback, that whenever they need a spare moment to themselves, just turn on Barney, well, it is easy to see how there would be a problem.
The behavioral indicators for autism spectrum disorders are often quite weak before age two. I don't find it much of a stretch to think that paternal age and TV time could explain the vast majority of autism increase since the mid 1980s. Certainly this is a hypothesis much more worth pursuing than the thimerisol issue. But there is also enough evidence of a pre-natal abnormalities in autism to think it is ridiculous that TV would be the only reason for the increase in autism rate since the mid 1980s.
PDF presenters do not auto-load the next slide....this leads to an uncomfortable silence between the time you try to advance the slide, and the time the slide actually advances. This effect depends on how complicated the next slide is.
PDF cannot embed audio/video, so if you want a multimedia presentation you are screwed.
You cannot draw onto a slide to explain your point.
I've used MagicPoint, which works fine. You make each slide into a standalone graphics file, like a JPEG or GIF or PDF. Then, you setup a control file to present them. It can handle all of the PDF glitches.
OpenOffice Impress can also be used, although I was much less impressed with it than with PowerPoint. But that was a while ago, things may be better now.
The real problem is the strength of the Powerpoint monopoly. At mixed presentations, like symposia, the presenters force everyone to load their PowerPoint presentation onto one Windows machine, and anyone using Powerpoint on a Mac, or Keynote, or Linux, without fail has some glitch. On the one hand there is a real advantage to everyone being able to use the same presentation machine, but on the other hand the incompatibilities stink up the presentation.
First of all, TiVO never sold out to advertisers. They recognized, from the beginning, that TV is not free. Advertisters pay for your TV in order to get hits on you. If you remove their hits, they will come after you, big-time, and your cool little dot-com startup will be just another crash and burn. This is TV, the nature of the business is delivering targetting advertising.
Second of all, TiVO's future revenue stream is ALL about advertising revenue. Consider they can allow you to download a longer infomercial for any product you are interested in by having you click. They can and do record EVERY time you rewind a commercial to watch it again. They report this to advertisers for a fee.
TiVO is positioned to provide an additional layer of targetted advertising on TV content, Advertising WITH FEEDBACK on how the use watches it. How many time, where it gets paused, skipped, FFWed, etc. And not only that, TiVO can provide advertisers with cross-correlational analysis of who watches which TV shows in intimate detail with 10000 times hte power that Nielsen can.
TiVO is headed to becoming the advertising powerhouse of 21st century television.
Mac is not dramatically more secure through launchd...
It is simple really. Six years into OS X, growing market share, and no viruses in the wild.
First principle. No ports open by default. Macs ship with a closed box. Plug it into the Internet, wait, and your machine will never get infected simply because it is not listening on any port, and no attacker has any foothold to get into the box. Over the years Windows has shipped with a wide variety of open ports, whether they be for netbios, smbd, messenger, IIS (on NT), or others. Many of these have been launching pads for viruses and worms.
Second principle. Design the OS from the ground up to support privilege descalation. That is, make it so that every action on the machine is executed with User privileges or less, unless you really need more privilege. Launchd is a part of this. On Windows, you still have ActiveX with escalatable privilege, and people get infected from web surfing or opening email.
That is really all it takes. Make it so a user cannot compromise the OS trivially, and there are no open ports, and you made a box as secure as a Mac. Once you start opening ports, you need to know what you are doing or you will be 0wn3d by some script kiddy. Make it secure by default, and force the user to take positive action to do anything that is a potential security problem (like installing executables from random places on the internet).
But at the same time, I don't think that TiVo should have such a broad patent on this technology. In this case, it seems they're using their patent defensively simply to stay alive in a market that can quickly and easily be taken away from them by the satellite and cable companies that provide the content transport.
I think this is EXACTLY what patents were made for.
The cable and satellite companies control the flow of content. TiVO had no say about that. They have an invention that makes the content more valuable.
The obvious solution for the content providers is to strangle TiVO out of existence, reverse engineer their product, and take the advances that a DVR brings to themselves.
Patents were created specifically to protect the inventors from leveraged competitors. TiVO was the first one to make a useful DVR, and they did a very good job expanding their patent portfolio for coverage.
Now, they are only demanding that the patents their invention and market are already using are enforced against competitors. This is not defensive patents - like the Wright bros who tried to take a toll from all airplane makers. This is an inventor with a product and a patent trying to make the most of it.
Exactly the scenario in which patent protect valid inventors against bigger competitors - why patent law exists in the first place.
I haven't seen the actual patent, but it sounds a lot like the patents that simply apply a well understood concept to new technology. Like patenting a wheel made out of some new alloy, rather than just a patent for the alloy itself.
You misunderstand the landscape surrounding video encoding/playback in 1997.
It was ALMOST impossible to use a hard drive to encode video in real-time in 1997.
It required substantial high tech engineering to make a prototype that worked reasonably.
TiVO did this and had a product out at a time no one else could even think about a cheapo consumer product that acted as a DVR and provided the quality that TiVO provided.
It's hard to think of this in 1997 terms. The patent examiner, in considering such a patent, will only use patents or other items with clear dates associated with them to assess if it is non-obvious. The fact that no one had a working prototype or patent application to do real-time video digitization and encoding and playback and time shifting in 1997 speaks to the validity of the patent (under existing patent law).
Just to make things perfectly clear. Continuous playback of encoded video was difficult in 1997 - much less encoding in real-time WHILE reading a recorded video and decoding it.
Today parts of your consumer level computer are 10 to 1000 times faster, and you could do this in most operating systems. That is not relevant to whether it is a good patent. If it was obvious in 1997, someone else would have done it earlier.
1997: big hard drive was 10 GBytes 233 MHz Pentium MMX was the hot processor. MPEG 1/2 was the dominant encoding scheme.
I didn't mean to suggest that reading and writing video from a computer wasn't obvious.
Just that with consumer level equipment, in 1997, it was DERN tough to record one show while playing back another using one hard drive.
The Pentium 233 MHz MMX was the newest chip then...it has to do real-time video encoding without buffer overflow. Or you have to hand off the video encoding to another DSP, and it had to do real-time video encoding without buffer overflow.
The fastest hard drives then had data rates more than 10 times slower than today.
To record one show while playing back another you can to do real-time encoding, buffering, and reading and writing from different parts of the hard drive. Again, not impossible to conceive of, but very difficult to do with equipment available then in a way cost-effective for a consumer device.
I'm always a fan of a company that becomes a verb. TiVO, google, etc...
The patents were neither obvious or easy at the time of the patent application. Hardware was so slow back then that video encoding and playback from hard drives were difficult. Today, everything is 10 times faster, so it is easy to think of it as trivial. But you need to think of it in terms of what was available in 1997.
That brings up somewhat obvious questions about the applicability and utility of our patent system. TiVO patented something in 1997 that was novel and non-obvious. However, it would have been both obvious and easy 5 years later. So, they get 17 years of monopoly for being ahead of their time.
I dig it though, I have friends who work there, and they could use the money...
High frequency hearing loss is a definite trend, but cannot be used to predict individual responses. The hair cells in the cochlea that encode high frequencies are the highest metabolism hair cells, and they are the ones that die first, which leads to high frequency hearing loss.
There was a story a while back about a high frequency ringtone that kids used in classrooms thinking their teacher could not hear it...but some teachers can still hear 17 kHz just fine.
Talking to a machinist in person will usually result in cheaper prices than emachineshop, for single one-off parts or multiple parts. Buying a mill is great, but you will also need a wide range of accessories, and a lathe, bandsaw, drill press, grinder, belt sander, etc. A typical non-CNC machineshop probably costs $100,000 to fully equip. But you need to have some savvy, machinists run from $100+ per hour down to about $35/hr, and some of the $35/hr guys are much better than most of the $100+/hr guys.
The really nice thing about emachineshop is you don't actually have to talk to a machinist.
Under de-regulation, if powerlines go down, the power companies contract an emergency service repair, and charge it to their customers on the next bill.
However, power companies do have to pay out of their pockets for prophylactic tree service. So they stopped doing that, and their quarterly earnings improved dramatically.
This is de-regulation!
If powerlines are above ground, but tree service is kept up regularly, then power doesn't go down in storms.
Cost is about TCO, not just initial. It depends on how far out you extend your costs whether it would be worthwhile to the power companies or not. This, I have no idea.
Not for power companies, it is about paying for it.
Before de-regulation, power companies regularly did prophylactic tree trimming to keep the power on, and power rarely went off.
Now, we are all de-regulated. The trees grow, and power companies don't cut them. WTF?
A storm comes, and the power goes out. The power company contracts an emergency crew at many times the rate to do the minimal tree trimming necessary to get the power back on. The de-regulated power company assesses everyone a surplus fee to cover their costs.
Power companies spend many times as much on tree trimming now, but the consumer takes all of the hit. Tree trimming has fallen off the quarterly balance sheets for the power companies, and the consumer pays more for it.
This is how de-regulated power works. Take a look at your bill.
The rub is how I found this out. An acquaintance did a 2 year re-structuring gig at a power company that had been de-regulated. He was pissed his power went out every time there was a storm and made them cut all the trees covering his power lines. Five years later the power has not gone out even one time.
Not sure what you're saying here. The more water in ice, the lower the water levels will be. Global warming would mean less ice, higher water levels overall.
This is a common misconception.
Take a glass, fill it with ice cubes, and add water until it is just about to spill over. Then wait. As the ice melts, the water level in the glass decreases.
This occurs because the ice is less dense than the water.
There has been an enormous reduction in the Arctic icepack since 1992. Explanations on the 3.2 mm decrease in sea level would seem likely to be explained simply by decreases in icepack which in turn reduce water levels.
They don't talk about the water temp much either, but water changes density with temperature, and 4C is the densest temperature for water. So a slight shift in water temperature could also have impacts on total water volume.
Also, the mean atmospheric pressure over the Arctic could have changed.
In a series of well-directed op-ed style pieces, he claims there is no warming trend since 1998. That's right, global warming ended 8 years ago.
Only a really sincere attempt at misdirection could conclude that from the data from which he extracted it. The secret, of course, is that the annual data is noisy and 1998 is an anomalously warm year. Knowing that, if you start at 1998, there is no significant trend. But that only holds in the one data set he used, annual air temperature averages. Water temperatures act as better barometers, because they have capacity for temperature and inherently average over time, and there is no evidence global warming is slowing using better statistical analysis.
Carter is a shill paid by Exxon Mobil to fight global warming to keep public policy pro-oil.
I'm not going to look into each of them, but I suspect each is a paid shill. Carter receives the bulk of his research funding from petroleum-based energy companies.
As another point, climate change is a HOT topic right now. Top universities are paying top dollar to get good climatologists on board. And look at the universities associated with the profs cited in the article.
The best minds on the matter say global warming is real, and the best evidence available says we contribute to the change in a meaningful way.
I shouldn't have excluded Python. Both Perl and Python are reasonable in memory management, as in a simple "Hello World" program doesn't require 250 MBytes of RAM to execute.
Others, not so much, ESPECIALLY Java.
I do mostly scientific programming, I use Matlab for a lot of quick and dirty stuff, Perl when it requires a little more programming, and C/C++ for any real heavy computing because the others are JUST TOO SLOW. There is no good reason that Matlab or Perl or Python should be 10 times slower and use many times the RAM as C/C++, but they do. And Java is a complete memory hog, off by an order of magnitude.
I'm sorry, you lost me there when I spit out my coffee on my keyboard.
Pascal was a language written to teach people to code, not to actually get work done.
Interpreted languages, or scripting languages as I like to call them, GENERALLY are real crappy with memory management, bloated, pigs of software.
Perl is a lovely exception.
But Java or Matlab or VB seem to thrive in using as much RAM as possible to get a job done. Any serious work I do in any of them I end up re-writing in C/C++ in order to streamline the RAM usage and make the program work reasonably.
And this is the main reason Java will not take over as a dominant desktop theme. People actually like programs to work and not lock up the machine and screech it to a halt.
Real programmers manage their own memory usage, and don't think of it as a nuisance, but as a mission critical objective, because too many scripting engines are REALLY bad at it. Java on linux is a great example. It is real buggy and locks up a lot, and cannot do ANYTHING without loading 250 MBytes of libraries into RAM. That's crap, and it will always be crap. It is possible to write scripting engines that do not suck, and when that happens in something like Java it WILL take over because programmers want their programs to work first and foremost, and secondarily want to code it quickly.
This array was in work in the lab I in which I did my PhD, and has more recently become viable. It can mimic, to the best extent possible, spatial inputs to the skin, in a fairly incredible virtual reality kind of way, and is a great research tool. But it is a HUGE device made to provide good control over a single fingertip of skin.
Slashdot Mirror
I sorta meant there could be a shift in engines to diesel engines, which would then replace
gasoline engines, and gasoline fuel...
In contemplating the greenhouse gas problem (or our budget deficit for energy), its quite a pickle. Cars burn predominantly gas. Gas requires oil - petroleum oil - for production. The USA doesn't make enough oil, so we necessarily send a LOT of US dollars to OPEC each year, and many OPEC members are openly antagonistic towards the USA.
Do we really want to give our enemies a lot of money?
The ONLY reasonable ways out are
1) electric cars
2) biodiesel
Electric cars don't have enough power or range (and if they gain power and range, they gain cost and weight and lose efficiency). However, electric is more efficient. But the energy would still need to come from somewhere, most of the US power plants that would provide the energy for electric cars still burn hydrocarbons, but more importantly, consumers have not taken to them,
Biodiesel costs $1 more per gallon - a HUGE 37% decrease in efficiency. Unless oil stays over $70/bl, then it comes close to break even. Biodiesel creates a LOT of USA jobs in farming and manufacturing. Biodiesel doesn't increase greenhouse gas emissions.
What does the future hold for energy? The gas car trap is a REAL problem for the USA. Biodiesel is one way out. Electric cars and more nuclear plants is the most probable other solution.
Frankly, I like the biodiesel solution, but it may be cheaper in the short-term to continue overthrowing Middle East governments. I wish there was a sarcastic overtone to that, but it is more like a grudging acceptance of our current reality.
If oil is over $70/barrel it is not so far fetched. Not the yellow grease
based biodiesel, which is very limited in scope, but a B100 biodiesel.
If you factor in the cost of emissions...B100 biodiesel has NO net emissions,
and petroleum based diesel has significant emissions costs....
Factor in the cost of emissions on environmental destruction at
$1/gallon of fuel and keep oil over $70/barrel,
and America has a new soybean based economy.
I don't think that is really gonna happen, but there
are not so unrealistic contingencies that
would bring it to pass. Like if the US government ever
really starts to care about global warming and at the
same time the oil fields in Iraq and Iran both get
incapacitated.
There's scads of papers on biodiesel, its efficiency, and cost.
e l/
If petroleum goes up in price a bunch more, biodiesel gets to break even.
The unmapped territory is that although it burns a hydrocarbon, 100% biodiesel
doesn't increase atmospheric CO2, because that CO2 was removed from the atmosphere
less than a year prior. It is cyclic in the short-term. Biodiesel could be a
near drop-in replacement for gasoline in cars and solve greenhouse gas
problems from automobiles. Of course, if you use peanuts instead of soybeans, and
oil costs stay high....people bet billions on shifts like this, the shift
to biodiesel would become reality if regular diesel wholesale prices get too
high and we have a strong need to minimize emissions...both of which are
very real scenarios. Both factors have shifted a lot since this white paper
in 2002.
http://www.eia.doe.gov/oiaf/analysispaper/biodies
This seems like the sort of thing that could happen if the brain's speech area's neurons somehow became trained to stop delivering impulses for "normal" speech. In this case, it would be theoretically possible to train the network back to normal levels. Of course, it could be something completely different.
Spasmodic dysphonia is a form of focal dystonia. This neurological condition occurs when sensory inputs, and motor outputs, get 'confused'. There is a necessity for motor programs to operate in opposition - flexors and extensors mutually inhibit. So, when you flex, the extensor muscles are not contracting at the same time, and vice versa. Somehow, in focal dystonia, the motor programs that cannot conflict if one is to have normal motor control, do conflict.
The good news is that at least for some forms of focal dystonia, effective treatments are now available from one of my former colleagues.
Arch Phys Med Rehabil. 2003 Oct;84(10):1505-14.
And yes, I've contributed as well
Somatosens Mot Res. 2002;19(4):347-57
I'm not familiar with the details of focal dystonia of the vocal tracts, but I suspect that an intelligent otolaryngologist could devise a training strategy that would restore normal speech....
I dunno about that.
Prior work has shown a very strong relationship between the number of hours of TV averaged per day before age two, and incidence of ADD. In fact, if TV were a causative factor, this would explain the vast majority of ADD cases, because the correlation was quite high.
There is also recent work from Israel showing that increased father age at conception is a huge correlational risk factor for autism.
There is also the recommendation from professional pediatric societies. They have consistently recommended, for years and years and years, that there is no recommended amount of daily TV watching for children, and in fact there should be NO TV watching for children under 3.
The reasons are simple. A child, before age 3, will interact with everything in its environment. This is a learning process. If you watch a 2 year old in front of Barney, they turn into a statue. Motionless, completely non-interactive, they are simply mesmerized. If you do not interact, a huge source for learning is lost. Now, not all two year olds will sit motionless in front of a TV. But a lot will. And if that results in parental feedback, that whenever they need a spare moment to themselves, just turn on Barney, well, it is easy to see how there would be a problem.
The behavioral indicators for autism spectrum disorders are often quite weak before age two. I don't find it much of a stretch to think that paternal age and TV time could explain the vast majority of autism increase since the mid 1980s. Certainly this is a hypothesis much more worth pursuing than the thimerisol issue. But there is also enough evidence of a pre-natal abnormalities in autism to think it is ridiculous that TV would be the only reason for the increase in autism rate since the mid 1980s.
Stick to the truthiness, it is what really counts!
(Yes I sometimes watch Colbert too, and btw WOZ is the guest tonite!)
Problem....
PDF presenters do not auto-load the next slide....this leads to an uncomfortable silence between the time you try to advance the slide, and the time the slide actually advances. This effect depends on how complicated the next slide is.
PDF cannot embed audio/video, so if you want a multimedia presentation you are screwed.
You cannot draw onto a slide to explain your point.
I've used MagicPoint, which works fine. You make each slide into a standalone graphics file, like a JPEG or GIF or PDF. Then, you setup a control file to present them. It can handle all of the PDF glitches.
OpenOffice Impress can also be used, although I was much less impressed with it than with PowerPoint. But that was a while ago, things may be better now.
The real problem is the strength of the Powerpoint monopoly. At mixed presentations, like symposia, the presenters force everyone to load their PowerPoint presentation onto one Windows machine, and anyone using Powerpoint on a Mac, or Keynote, or Linux, without fail has some glitch. On the one hand there is a real advantage to everyone being able to use the same presentation machine, but on the other hand the incompatibilities stink up the presentation.
Let's get a few things straight.
First of all, TiVO never sold out to advertisers. They recognized, from the beginning, that TV is not free. Advertisters pay for your TV in order to get hits on you. If you remove their hits, they will come after you, big-time, and your cool little dot-com startup will be just another crash and burn. This is TV, the nature of the business is delivering targetting advertising.
Second of all, TiVO's future revenue stream is ALL about advertising revenue. Consider they can allow you to download a longer infomercial for any product you are interested in by having you click. They can and do record EVERY time you rewind a commercial to watch it again. They report this to advertisers for a fee.
TiVO is positioned to provide an additional layer of targetted advertising on TV content, Advertising WITH FEEDBACK on how the use watches it. How many time, where it gets paused, skipped, FFWed, etc. And not only that, TiVO can provide advertisers with cross-correlational analysis of who watches which TV shows in intimate detail with 10000 times hte power that Nielsen can.
TiVO is headed to becoming the advertising powerhouse of 21st century television.
Elias Zerhouni may be a little miffed at being called Zerhounibut!
Emacs or Vi is really nice for development, but neither of them are an IDE.
One window open with VIM.
One window open with GDB.
Choose one of a dozen source tree browser apps.
La Voila, you have an efficient IDE.
Mac is not dramatically more secure through launchd...
It is simple really. Six years into OS X, growing market share, and no viruses in the wild.
First principle. No ports open by default. Macs ship with a closed box. Plug it into the Internet, wait, and your machine will never get infected simply because it is not listening on any port, and no attacker has any foothold to get into the box. Over the years Windows has shipped with a wide variety of open ports, whether they be for netbios, smbd, messenger, IIS (on NT), or others. Many of these have been launching pads for viruses and worms.
Second principle. Design the OS from the ground up to support privilege descalation. That is, make it so that every action on the machine is executed with User privileges or less, unless you really need more privilege. Launchd is a part of this. On Windows, you still have ActiveX with escalatable privilege, and people get infected from web surfing or opening email.
That is really all it takes. Make it so a user cannot compromise the OS trivially, and there are no open ports, and you made a box as secure as a Mac. Once you start opening ports, you need to know what you are doing or you will be 0wn3d by some script kiddy. Make it secure by default, and force the user to take positive action to do anything that is a potential security problem (like installing executables from random places on the internet).
But at the same time, I don't think that TiVo should have such a broad patent on this technology. In this case, it seems they're using their patent defensively simply to stay alive in a market that can quickly and easily be taken away from them by the satellite and cable companies that provide the content transport.
I think this is EXACTLY what patents were made for.
The cable and satellite companies control the flow of content. TiVO had no say about that. They have an invention that makes the content more valuable.
The obvious solution for the content providers is to strangle TiVO out of existence, reverse engineer their product, and take the advances that a DVR brings to themselves.
Patents were created specifically to protect the inventors from leveraged competitors. TiVO was the first one to make a useful DVR, and they did a very good job expanding their patent portfolio for coverage.
Now, they are only demanding that the patents their invention and market are already using are enforced against competitors. This is not defensive patents - like the Wright bros who tried to take a toll from all airplane makers. This is an inventor with a product and a patent trying to make the most of it.
Exactly the scenario in which patent protect valid inventors against bigger competitors - why patent law exists in the first place.
I haven't seen the actual patent, but it sounds a lot like the patents that simply apply a well understood concept to new technology. Like patenting a wheel made out of some new alloy, rather than just a patent for the alloy itself.
You misunderstand the landscape surrounding video encoding/playback in 1997.
It was ALMOST impossible to use a hard drive to encode video in real-time in 1997.
It required substantial high tech engineering to make a prototype that worked reasonably.
TiVO did this and had a product out at a time no one else could even think about a cheapo
consumer product that acted as a DVR and provided the quality that TiVO provided.
It's hard to think of this in 1997 terms. The patent examiner, in considering such a
patent, will only use patents or other items with clear dates associated with them
to assess if it is non-obvious. The fact that no one had a working prototype or patent
application to do real-time video digitization and encoding and playback and time
shifting in 1997 speaks to the validity of the patent (under existing patent law).
Just to make things perfectly clear. Continuous playback of encoded video was difficult
in 1997 - much less encoding in real-time WHILE reading a recorded video and decoding
it.
Today parts of your consumer level computer are 10 to 1000 times faster, and you could
do this in most operating systems. That is not relevant to whether it is a good patent. If
it was obvious in 1997, someone else would have done it earlier.
1997: big hard drive was 10 GBytes
233 MHz Pentium MMX was the hot processor.
MPEG 1/2 was the dominant encoding scheme.
I didn't mean to suggest that reading and writing video from a computer wasn't obvious.
Just that with consumer level equipment, in 1997, it was DERN tough to record one show while playing back another using one hard drive.
The Pentium 233 MHz MMX was the newest chip then...it has to do real-time video encoding without buffer overflow. Or you have to hand off the video encoding to another DSP, and it had to do real-time video encoding without buffer overflow.
The fastest hard drives then had data rates more than 10 times slower than today.
To record one show while playing back another you can to do real-time encoding, buffering, and reading and writing from different parts of the hard drive. Again, not impossible to conceive of, but very difficult to do with equipment available then in a way cost-effective for a consumer device.
I'm always a fan of a company that becomes a verb. TiVO, google, etc...
The patents were neither obvious or easy at the time of the patent application. Hardware was so slow back then that video encoding and playback from hard drives were difficult. Today, everything is 10 times faster, so it is easy to think of it as trivial. But you need to think of it in terms of what was available in 1997.
That brings up somewhat obvious questions about the applicability and utility of our patent system. TiVO patented something in 1997 that was novel and non-obvious. However, it would have been both obvious and easy 5 years later. So, they get 17 years of monopoly for being ahead of their time.
I dig it though, I have friends who work there, and they could use the money...
I'm 38 and I can hear them just fine.
n g.html
High frequency hearing loss is a definite trend, but cannot be used to predict individual responses. The hair cells in the cochlea that encode high frequencies are the highest metabolism hair cells, and they are the ones that die first, which leads to high frequency hearing loss.
There was a story a while back about a high frequency ringtone that kids used in classrooms thinking their teacher could not hear it...but some teachers can still hear 17 kHz just fine.
http://www.nytimes.com/2006/06/12/technology/12ri
Talking to a machinist in person will usually result in cheaper prices than emachineshop, for single one-off parts or multiple parts. Buying a mill is great, but you will also need a wide range of accessories, and a lathe, bandsaw, drill press, grinder, belt sander, etc. A typical non-CNC machineshop probably costs $100,000 to fully equip. But you need to have some savvy, machinists run from $100+ per hour down to about $35/hr, and some of the $35/hr guys are much better than most of the $100+/hr guys.
The really nice thing about emachineshop is you don't actually have to talk to a machinist.
These arguments are ALL irrelevant.
Under de-regulation, if powerlines go down, the power companies contract an emergency service repair, and charge it to their customers on the next bill.
However, power companies do have to pay out of their pockets for prophylactic tree service. So they stopped doing that, and their quarterly earnings improved dramatically.
This is de-regulation!
If powerlines are above ground, but tree service is kept up regularly, then power doesn't go down in storms.
Cost is about TCO, not just initial. It depends on how far out you extend your costs whether it would be worthwhile to the power companies or not. This, I have no idea.
Not for power companies, it is about paying for it.
Before de-regulation, power companies regularly did prophylactic tree trimming to keep the power on, and power rarely went off.
Now, we are all de-regulated. The trees grow, and power companies don't cut them. WTF?
A storm comes, and the power goes out. The power company contracts an emergency crew at many times the rate to do the minimal tree trimming necessary to get the power back on. The de-regulated power company assesses everyone a surplus fee to cover their costs.
Power companies spend many times as much on tree trimming now, but the consumer takes all of the hit. Tree trimming has fallen off the quarterly balance sheets for the power companies, and the consumer pays more for it.
This is how de-regulated power works. Take a look at your bill.
The rub is how I found this out. An acquaintance did a 2 year re-structuring gig at a power company that had been de-regulated. He was pissed his power went out every time there was a storm and made them cut all the trees covering his power lines. Five years later the power has not gone out even one time.
Not sure what you're saying here. The more water in ice, the lower the water levels will be. Global warming would mean less ice, higher water levels overall.
This is a common misconception.
Take a glass, fill it with ice cubes, and add water until it is just about to spill over. Then wait. As the ice melts, the water level in the glass decreases.
This occurs because the ice is less dense than the water.
There has been an enormous reduction in the Arctic icepack since 1992. Explanations on the 3.2 mm decrease in sea level would seem likely to be explained simply by decreases in icepack which in turn reduce water levels.
They don't talk about the water temp much either, but water changes density with temperature, and 4C is the densest temperature for water. So a slight shift in water temperature could also have impacts on total water volume.
Also, the mean atmospheric pressure over the Arctic could have changed.
Interesting stuff.
In a series of well-directed op-ed style pieces, he claims there is no warming trend since 1998. That's right, global warming ended 8 years ago.
Only a really sincere attempt at misdirection could conclude that from the data from which he extracted it. The secret, of course, is that the annual data is noisy and 1998 is an anomalously warm year. Knowing that, if you start at 1998, there is no significant trend. But that only holds in the one data set he used, annual air temperature averages. Water temperatures act as better barometers, because they have capacity for temperature and inherently average over time, and there is no evidence global warming is slowing using better statistical analysis.
Carter is a shill paid by Exxon Mobil to fight global warming to keep public policy pro-oil.
Carter is a paid shill of the energy companies.
I'm not going to look into each of them, but I suspect each is a paid shill. Carter receives the bulk of his research funding from petroleum-based energy companies.
As another point, climate change is a HOT topic right now. Top universities are paying top dollar to get good climatologists on board. And look at the universities associated with the profs cited in the article.
The best minds on the matter say global warming is real, and the best evidence available says we contribute to the change in a meaningful way.
Perl is very C oriented in its syntax.
I shouldn't have excluded Python. Both Perl and Python are reasonable in memory management, as in a simple "Hello World" program doesn't require 250 MBytes of RAM to execute.
Others, not so much, ESPECIALLY Java.
I do mostly scientific programming, I use Matlab for a lot of quick and dirty stuff, Perl when it requires a little more programming, and C/C++ for any real heavy computing because the others are JUST TOO SLOW. There is no good reason that Matlab or Perl or Python should be 10 times slower and use many times the RAM as C/C++, but they do. And Java is a complete memory hog, off by an order of magnitude.
I am a Pascal programmer from ancient days
I'm sorry, you lost me there when I spit out my coffee on my keyboard.
Pascal was a language written to teach people to code, not to actually
get work done.
Interpreted languages, or scripting languages as I like to call them, GENERALLY are real crappy with memory management, bloated, pigs of software.
Perl is a lovely exception.
But Java or Matlab or VB seem to thrive in using as much RAM as possible to get a job done. Any serious work I do in any of them I end up re-writing in C/C++ in order to streamline the RAM usage and make the program work reasonably.
And this is the main reason Java will not take over as a dominant desktop theme. People actually like programs to work and not lock up the machine and screech it to a halt.
Real programmers manage their own memory usage, and don't think of it as a nuisance, but as a mission critical objective, because too many scripting engines are REALLY bad at it. Java on linux is a great example. It is real buggy and locks up a lot, and cannot do ANYTHING without loading 250 MBytes of libraries into RAM. That's crap, and it will always be crap. It is possible to write scripting engines that do not suck, and when that happens in something like Java it WILL take over because programmers want their programs to work first and foremost, and secondarily want to code it quickly.
Google "400-pin tactile array"
This array was in work in the lab I in which I did my PhD, and has more recently become viable. It can mimic, to the best extent possible, spatial inputs to the skin, in a fairly incredible virtual reality kind of way, and is a great research tool. But it is a HUGE device made to provide good control over a single fingertip of skin.