HD decoding isn't exactly lightweight stuff. Most HDTVs have some sort of operating system installed, along with a CPU and some RAM.
Sure, it might not be a general-purpose computer, although modern TVs do indeed do a fair bit of computation. In fact, my Samsung TV runs Linux to support some of its functionality. (I had a somewhat nerdy moment when I first purchased it, and noticed a printed copy of the GPL in the box)
I have no idea what the cable box runs, although my TV, DVR, and wireless router all run Linux natively (and not as a result of any sort of conscientious purchasing decisions on my part). Blu-Ray players are required by the spec to have a considerable amount of processing power, memory, local storage, and an ethernet connection.
Roku sell a $99 box that will stream Netflix to your TV over 802.11g. If we can create a standardized video on demand protocol, I'm sure we'll start seeing this hardware built into TVs as well.
We've really blurred the line of what a "computer" actually is. Now we just need to wait for software and user interfaces to catch up so that we can finally start seeing some convergance.
How old is that list? Those products are ancient. SONICBlue went bankrupt in 2003!
Also, IIRC, the 3rd-party players could only sync with the Mac version of iTunes. I wouldn't be surprised if the support has been removed entirely, given that I sincerely doubt people are still using those players (some of which hold as little as 32MB!)
I was with you up until the DRM part. The iTunes store is almost entirely DRM-free by now.
If you don't like AAC either, Amazon.com sells MP3 downloads that are cheaper than iTunes downloads, and has about the same selection. iTunes was once a monopoly, although Apple's DRM practices ironically ended that rather quickly.
Cut me a break. 95% of the time, the folks fighting against illegal immigration are racist as hell, and automatically label any hispanic person as a probable 'illegal'
Is it an economic problem? Definitely. Is it as bad as people are claiming it to be? Probably not.
The solutions aren't great either. Immigration is something we're either going to have to put up with, or commit some pretty severe human rights violations to correct. (Also, are there many native-born Americans who are willing to pick fruit for $3/hour? Like it or not, we've had so many illegal immigrants for so long that the illegal labor force has become an integral part of the economy)
Most major experiments shut down for at least a few months out of the year for scheduled maintenance and/or improvements. Additionally, most big projects don't have the funding to operate 24/7/365 -- cryo expenses are particularly staggering.
Given the amount of time it takes to warm/cool the LHC, it makes sense to schedule all of this maintenance all in one go. Once you're in that frame of mind, you can reschedule your operations to reduce electricity costs...and why wouldn't you?
I'm pretty sure that these boats are stripped to bare metal before being sunk. In other words, the hazardous materials part of the problem doesn't really exist.
Still, (safely) chopping a boat up can't be cheap. However, we are hurting for jobs, so it couldn't hurt...
(The deferred dismantling of the Reserve Fleet actually makes a lot of sense from an economic perspective. If the government has a project that it can defer almost indefinitely, it makes sense to wait until there's a recession and/or high unemployment. People get jobs, the government gets a better price, and everybody wins. In this case, it's particularly nice, considering that most of the money goes directly to domestic labor)
Apply the same scientific principles (and some RPG-esque nerdiness) to the problem, and you should come out fine after a few iterations.
Find a recipe of something that sounds good to you (Epicurious is one of many great places to look), and give it a go. Youtube, and a number of other video sites are treasure troves for brief instructional videos on various cooking techniques and methods. Some techniques do involve a bit of skill and dexterity, but these are few, far between, and easy to avoid.
Cook it (in a small quantity) If you don't like the result, figure out what went wrong, and try it again, adjusting the proportions or cooking methods as you go.
Cooking is like a bizarre combination of the scientific method and D&D. Once you do it enough (and know to make adjustments along the way), you'll Level Up, and get better. It's not necessarily a long, drawn-out process either. I taught myself almost everything I know about cooking over about 2 months.
Survivorship Bias: We tend to only remember the good ones.
That all said, the signal-to-noise ratio does seem to have decreased in recent years, with most successful titles being part of already-established franchises.
(Offtopic: I haven't played a good RPG in ages. Any suggestions? Doesn't matter how old (or new) it is.)
I'm not quite sure I agree with you. Proper safety protocols can be implemented on a very short time-scale.
In addition to working in a few research labs, I've also worked in the special events industry (theatre, concerts, tradeshows, etc.). There, our timescales rarely exceed days, let alone hours. Improvisation is often required, and it's extremely important to pay attention to safety requirements throughout the whole course of the job. Obviously, we don't have the time to seek approval for every single thing that we do.
Rather than conform to a strict set of safety regulations, we have standards and risk assessments for all of the basic tasks that we perform and the devices we operate. Some of these are very basic (make sure electrical devices are grounded, inspect all cords on a regular basis, double-check that the ceiling is strong enough to support that enormous light fixture, etc...). However, it's important that they're in place and being enforced. Workers need to have at least a basic understanding of safety protocols and procedures before working at even the lowest levels.
Although the process of writing these risk assessments is inane and boring, you often uncover faults that can easily be corrected. If you've got dangerous voltages running through an exposed lead, you can cross the item off of the risk assessment simply by covering the wire in an insulator, and attaching a warning label.
These processes could easily be adopted to a laboratory setting with a minimal effort.
Of course, the event industry doesn't have the brightest reputation. We're working on it. In the US, the unions have enacted a strict set of rules to protect their workers (as well as other absurdly strict sets of rules that make the industry extremely undesirable to work in IMHO, but that's another topic). In the EU, a strict set of Health & Safety rules (a la OSHA) have recently come into effect, and have had a tremendous impact on worker safety.
At risk of drifting somewhat off-topic, this is actually a big problem. Political discourse has becomes so polarized that we're unable to actually work out solutions to our problems.
You'll never get the conservatives to reform the welfare system, because they don't want it to exist. Unfortunately, however, they also don't have the power to abolish it. In the end, you end up with a lot of petty arguments and underhanded political tactics, and the problem grows larger and larger.
These sort of polarizing arguments occur quite frequently in university politics, and can result in safety standards not being properly implemented. Tenured professors can be very hot-headed.
Thank God this is finally coming up to the surface. The manner in which scientific research is conducted today is outright embarrassing.
I just graduated with an undergraduate physics degree last week, and after working two summers in separate research labs, I have virtually no desire to seek a higher degree in my subject. The manner in which some graduate students are treated is terrible.* There's no way that I can justify subjecting myself to poverty-level wages during the prime years of my life.
As much as I hate to say it, graduate students need some sort of union or organized labor movement. They literally form the backbone of all major academic research labs, and are treated (and paid) like dirt. Postdocs have an even more miserable existence.
Of course, you could also discuss the effects this has on the research itself quite extensively. In short, it can't be good. The most experienced individuals (ie. tenured professors) are almost entirely removed from the research itself.
Unfortunately, I now have a rather worthless undergraduate degree.
*I've seen it both ways. The first lab I worked in treated its students far better than the second, although the pay was still abysmal. However, I don't see many options for students who are paired with abusive research advisors, given dead-end projects, or otherwise treated unfairly.
...you let undergrads lose in a lab. A friend of mine was nearly electrocuted because one of her undergrads took it upon himself to do some wiring, and "grounded" the black wire to the body of a vacuum chamber. Little did he know that the "red is power, black is ground" convention that he learned in his intro to EE course doesn't apply to AC circuits.
Honestly, this sounds 100% like the professor or lab instructor's fault. You can't blame the undergrad for not knowing something that intuitively makes no sense.
When teaching safety standards (such as the color-coding of the wires), proper context should have been provided, ie. that the black/red convention only applies to DC circuits. However, I could see how this could easily be overlooked or misinterpreted.
What I can't see being overlooked is that the students were given a piece of equipment that presumably had mains current flowing through exposed and unlabeled leads.
If they were working directly with AC currents from the wall, they should have received the proper safety instruction before ever setting foot in the lab. There are no excuses for this.
I worked in an industrial research facility for a short time. I spent several days being briefed about the various safety regulations and precautions to be taken at the facility before I was even allowed to enter the building without an escort, let alone operate the equipment in my lab. This training was mandatory for all personnel, regardless of their prior education or work experience, and needed to be repeated periodically. "Letting undergrads loose in a lab" should never be allowed before some sort of basic safety instruction has been provided.
As for russian technical products, this is (or used to be, at least up until 15 years ago, I'm not up to date on their latest trends in production) a very weird mix of excellent quality parts and abysmal quality parts, assembled together with the greatest attention about 50% of the time, but also assembled together with half-arsed nonchalance the rest of the time.
A lot of that dates back to WWII. Russia acquired a great deal of technology from Germany as a spoil of war.
Unfortunately for them, they had to fill in the gaps concerning anything that was produced in the western half of the country.
This is why many optical devices produced in the Soviet Union (and even Russia today) were of extremely high quality and precision, but suffered from certain quality problems that were unusual in such high-end devices.
How many hardware failures are actually characterized by a complete 100% loss of communication (as you'd get by pulling the plug)?
Don't CPU and Memory failures tend to make the computer somewhat unstable before completely bringing it down? How would vSphere handle (or even notice) that?
Even hard disk failures can take a small amount of time before the OS notices anything is awry (although you're an idiot if you care enough about redundancy to worry about this sort of thing, but don't have your disks in a RAID array)
If you *really* want to just throw words onto a page, you could just use LyX or LaTeX (with everything set to the default).
Somebody once told me that LaTeX is supposed to be easy and user-friendly. Although anybody who's ever used it knows otherwise, I suppose you'd end up with a nicely-formatted document if you're willing to have the default stylesheet imposed upon you.
You can't mention office suites on the mac without bringing up iWork. Keynote is so much better than PowerPoint that it's virtually alone in its class, while Pages and Numbers are both quite good.
Pages lets you do proper page layouts in a word processing app, and offers better typography than Word (although it still lags behind LaTeX). Although it's arguably the least mature of the iWork apps, it's come a long way since its initial release, and I rarely use Word anymore.
Numbers is a great lightweight spreadsheet app that's perfectly sufficient for 99% of the world's spreadsheeting needs. In fact, its UI brings out most of Excel's commonly used functions (ie. most of what's in the Format -> Cell dialog) right onto the application's only toolbar. (The remaining 1% use Excel's advanced data acquisition and pivoting features, which, in all honesty, are quite good)
However, I think that they missed the point when applying that mantra to 10GB Ethernet. I've worked on plenty of networks that were saturated with local traffic. Replacing some of the uplinks with 10GB Ethernet connections over the existing copper would be an excellent (and cheap) way to improve capacity.
Every port of every switch doesn't need to operate at 10GB. However, it'd be nice if the backbone did (and was able to do so inexpensively, and over standard ethernet cabling).
However, I think that Ubuntu's a bit too young to call it a 'flop.' The project still has plenty of forward momentum behind it.
That it's the most popular Linux to date is certainly a feat, and major manufacturers have adopted it (albeit in limited circumstances). It may not have changed everything, though it did give things an enormous shove in the right direction. Currently, my eyes are on OpenOffice to clean up its act, or for a new competitor to emerge. The OS itself is no longer the limiting factor.
The closure of Bell Labs is one of the tragedies of the 20th century.
The closure of Bell Labs was the result of one of the greatest farces of the 20th century.
(Yes, the telecommunications act did address some very real problems that needed attention. Unfortunately the solution was in many ways, worse than the problem itself)
On a more serious note, I'd love to see some Math/Post-rock type stuff on Guitar Hero. Stuff by Mogwai, Radiohead, Sigur Ros, or 65daysofstatic would be fun and challenging. It'd be even cooler if you managed to work something like Fuck Buttons into the guitar hero controller scheme.
(If the developers are particularly evil, they'll integrate Rock Band's vocal functions into the Sigur Ros tracks)
True story: My computer randomly picks a song out of my music library to play as my alarm clock in the morning.
I was late for work the day it picked John Cage's 4'33"
(I eventually came to hate most of my music library as a result of this mechanism, and have now relegated the script to limit its selections to a playlist of songs I don't particularly like (and don't also happen to consist only of 4 and a half minutes of silence)
Slashdot Mirror
HD decoding isn't exactly lightweight stuff. Most HDTVs have some sort of operating system installed, along with a CPU and some RAM.
Sure, it might not be a general-purpose computer, although modern TVs do indeed do a fair bit of computation. In fact, my Samsung TV runs Linux to support some of its functionality. (I had a somewhat nerdy moment when I first purchased it, and noticed a printed copy of the GPL in the box)
I have no idea what the cable box runs, although my TV, DVR, and wireless router all run Linux natively (and not as a result of any sort of conscientious purchasing decisions on my part). Blu-Ray players are required by the spec to have a considerable amount of processing power, memory, local storage, and an ethernet connection.
Roku sell a $99 box that will stream Netflix to your TV over 802.11g. If we can create a standardized video on demand protocol, I'm sure we'll start seeing this hardware built into TVs as well.
We've really blurred the line of what a "computer" actually is. Now we just need to wait for software and user interfaces to catch up so that we can finally start seeing some convergance.
How old is that list? Those products are ancient. SONICBlue went bankrupt in 2003!
Also, IIRC, the 3rd-party players could only sync with the Mac version of iTunes. I wouldn't be surprised if the support has been removed entirely, given that I sincerely doubt people are still using those players (some of which hold as little as 32MB!)
I was with you up until the DRM part. The iTunes store is almost entirely DRM-free by now.
If you don't like AAC either, Amazon.com sells MP3 downloads that are cheaper than iTunes downloads, and has about the same selection. iTunes was once a monopoly, although Apple's DRM practices ironically ended that rather quickly.
Cut me a break. 95% of the time, the folks fighting against illegal immigration are racist as hell, and automatically label any hispanic person as a probable 'illegal'
Is it an economic problem? Definitely. Is it as bad as people are claiming it to be? Probably not.
The solutions aren't great either. Immigration is something we're either going to have to put up with, or commit some pretty severe human rights violations to correct. (Also, are there many native-born Americans who are willing to pick fruit for $3/hour? Like it or not, we've had so many illegal immigrants for so long that the illegal labor force has become an integral part of the economy)
Maybe not that obvious.
Most major experiments shut down for at least a few months out of the year for scheduled maintenance and/or improvements. Additionally, most big projects don't have the funding to operate 24/7/365 -- cryo expenses are particularly staggering.
Given the amount of time it takes to warm/cool the LHC, it makes sense to schedule all of this maintenance all in one go. Once you're in that frame of mind, you can reschedule your operations to reduce electricity costs...and why wouldn't you?
It's pretty clear that he considered it a joke at the beginning, and then realized he could make a fortune off of it.
-jcr
Fixed that for you.
I'm pretty sure that these boats are stripped to bare metal before being sunk. In other words, the hazardous materials part of the problem doesn't really exist.
Still, (safely) chopping a boat up can't be cheap. However, we are hurting for jobs, so it couldn't hurt...
(The deferred dismantling of the Reserve Fleet actually makes a lot of sense from an economic perspective. If the government has a project that it can defer almost indefinitely, it makes sense to wait until there's a recession and/or high unemployment. People get jobs, the government gets a better price, and everybody wins. In this case, it's particularly nice, considering that most of the money goes directly to domestic labor)
Apply the same scientific principles (and some RPG-esque nerdiness) to the problem, and you should come out fine after a few iterations.
Find a recipe of something that sounds good to you (Epicurious is one of many great places to look), and give it a go. Youtube, and a number of other video sites are treasure troves for brief instructional videos on various cooking techniques and methods. Some techniques do involve a bit of skill and dexterity, but these are few, far between, and easy to avoid.
Cook it (in a small quantity) If you don't like the result, figure out what went wrong, and try it again, adjusting the proportions or cooking methods as you go.
Cooking is like a bizarre combination of the scientific method and D&D. Once you do it enough (and know to make adjustments along the way), you'll Level Up, and get better. It's not necessarily a long, drawn-out process either. I taught myself almost everything I know about cooking over about 2 months.
Wouldn't you want to use an IR Spectrometer?
I'm actually somewhat surprised that we've never sent one up to Mars, given that you can find one in most research facilities today.
Survivorship Bias: We tend to only remember the good ones.
That all said, the signal-to-noise ratio does seem to have decreased in recent years, with most successful titles being part of already-established franchises.
(Offtopic: I haven't played a good RPG in ages. Any suggestions? Doesn't matter how old (or new) it is.)
I'm not quite sure I agree with you. Proper safety protocols can be implemented on a very short time-scale.
In addition to working in a few research labs, I've also worked in the special events industry (theatre, concerts, tradeshows, etc.). There, our timescales rarely exceed days, let alone hours. Improvisation is often required, and it's extremely important to pay attention to safety requirements throughout the whole course of the job. Obviously, we don't have the time to seek approval for every single thing that we do.
Rather than conform to a strict set of safety regulations, we have standards and risk assessments for all of the basic tasks that we perform and the devices we operate. Some of these are very basic (make sure electrical devices are grounded, inspect all cords on a regular basis, double-check that the ceiling is strong enough to support that enormous light fixture, etc...). However, it's important that they're in place and being enforced. Workers need to have at least a basic understanding of safety protocols and procedures before working at even the lowest levels.
Although the process of writing these risk assessments is inane and boring, you often uncover faults that can easily be corrected. If you've got dangerous voltages running through an exposed lead, you can cross the item off of the risk assessment simply by covering the wire in an insulator, and attaching a warning label.
These processes could easily be adopted to a laboratory setting with a minimal effort.
Of course, the event industry doesn't have the brightest reputation. We're working on it. In the US, the unions have enacted a strict set of rules to protect their workers (as well as other absurdly strict sets of rules that make the industry extremely undesirable to work in IMHO, but that's another topic). In the EU, a strict set of Health & Safety rules (a la OSHA) have recently come into effect, and have had a tremendous impact on worker safety.
At risk of drifting somewhat off-topic, this is actually a big problem. Political discourse has becomes so polarized that we're unable to actually work out solutions to our problems.
You'll never get the conservatives to reform the welfare system, because they don't want it to exist. Unfortunately, however, they also don't have the power to abolish it. In the end, you end up with a lot of petty arguments and underhanded political tactics, and the problem grows larger and larger.
These sort of polarizing arguments occur quite frequently in university politics, and can result in safety standards not being properly implemented. Tenured professors can be very hot-headed.
Thank God this is finally coming up to the surface. The manner in which scientific research is conducted today is outright embarrassing.
I just graduated with an undergraduate physics degree last week, and after working two summers in separate research labs, I have virtually no desire to seek a higher degree in my subject. The manner in which some graduate students are treated is terrible.* There's no way that I can justify subjecting myself to poverty-level wages during the prime years of my life.
As much as I hate to say it, graduate students need some sort of union or organized labor movement. They literally form the backbone of all major academic research labs, and are treated (and paid) like dirt. Postdocs have an even more miserable existence.
(It's worth mentioning here that is a depressingly accurate portrayal of academia in general)
Of course, you could also discuss the effects this has on the research itself quite extensively. In short, it can't be good. The most experienced individuals (ie. tenured professors) are almost entirely removed from the research itself.
Unfortunately, I now have a rather worthless undergraduate degree.
*I've seen it both ways. The first lab I worked in treated its students far better than the second, although the pay was still abysmal. However, I don't see many options for students who are paired with abusive research advisors, given dead-end projects, or otherwise treated unfairly.
...you let undergrads lose in a lab. A friend of mine was nearly electrocuted because one of her undergrads took it upon himself to do some wiring, and "grounded" the black wire to the body of a vacuum chamber. Little did he know that the "red is power, black is ground" convention that he learned in his intro to EE course doesn't apply to AC circuits.
Honestly, this sounds 100% like the professor or lab instructor's fault. You can't blame the undergrad for not knowing something that intuitively makes no sense.
When teaching safety standards (such as the color-coding of the wires), proper context should have been provided, ie. that the black/red convention only applies to DC circuits. However, I could see how this could easily be overlooked or misinterpreted.
What I can't see being overlooked is that the students were given a piece of equipment that presumably had mains current flowing through exposed and unlabeled leads.
If they were working directly with AC currents from the wall, they should have received the proper safety instruction before ever setting foot in the lab. There are no excuses for this.
I worked in an industrial research facility for a short time. I spent several days being briefed about the various safety regulations and precautions to be taken at the facility before I was even allowed to enter the building without an escort, let alone operate the equipment in my lab. This training was mandatory for all personnel, regardless of their prior education or work experience, and needed to be repeated periodically. "Letting undergrads loose in a lab" should never be allowed before some sort of basic safety instruction has been provided.
As for russian technical products, this is (or used to be, at least up until 15 years ago, I'm not up to date on their latest trends in production) a very weird mix of excellent quality parts and abysmal quality parts, assembled together with the greatest attention about 50% of the time, but also assembled together with half-arsed nonchalance the rest of the time.
A lot of that dates back to WWII. Russia acquired a great deal of technology from Germany as a spoil of war.
Unfortunately for them, they had to fill in the gaps concerning anything that was produced in the western half of the country.
This is why many optical devices produced in the Soviet Union (and even Russia today) were of extremely high quality and precision, but suffered from certain quality problems that were unusual in such high-end devices.
How many hardware failures are actually characterized by a complete 100% loss of communication (as you'd get by pulling the plug)?
Don't CPU and Memory failures tend to make the computer somewhat unstable before completely bringing it down? How would vSphere handle (or even notice) that?
Even hard disk failures can take a small amount of time before the OS notices anything is awry (although you're an idiot if you care enough about redundancy to worry about this sort of thing, but don't have your disks in a RAID array)
Good point.
Put the "back room" hundreds of miles away. Shouldn't be too hard with modern technology.
If you *really* want to just throw words onto a page, you could just use LyX or LaTeX (with everything set to the default).
Somebody once told me that LaTeX is supposed to be easy and user-friendly. Although anybody who's ever used it knows otherwise, I suppose you'd end up with a nicely-formatted document if you're willing to have the default stylesheet imposed upon you.
Wouldn't radiation be a concern?
You can't mention office suites on the mac without bringing up iWork. Keynote is so much better than PowerPoint that it's virtually alone in its class, while Pages and Numbers are both quite good.
Pages lets you do proper page layouts in a word processing app, and offers better typography than Word (although it still lags behind LaTeX). Although it's arguably the least mature of the iWork apps, it's come a long way since its initial release, and I rarely use Word anymore.
Numbers is a great lightweight spreadsheet app that's perfectly sufficient for 99% of the world's spreadsheeting needs. In fact, its UI brings out most of Excel's commonly used functions (ie. most of what's in the Format -> Cell dialog) right onto the application's only toolbar. (The remaining 1% use Excel's advanced data acquisition and pivoting features, which, in all honesty, are quite good)
I agree.
However, I think that they missed the point when applying that mantra to 10GB Ethernet. I've worked on plenty of networks that were saturated with local traffic. Replacing some of the uplinks with 10GB Ethernet connections over the existing copper would be an excellent (and cheap) way to improve capacity.
Every port of every switch doesn't need to operate at 10GB. However, it'd be nice if the backbone did (and was able to do so inexpensively, and over standard ethernet cabling).
Maybe, maybe not.
However, I think that Ubuntu's a bit too young to call it a 'flop.' The project still has plenty of forward momentum behind it.
That it's the most popular Linux to date is certainly a feat, and major manufacturers have adopted it (albeit in limited circumstances). It may not have changed everything, though it did give things an enormous shove in the right direction. Currently, my eyes are on OpenOffice to clean up its act, or for a new competitor to emerge. The OS itself is no longer the limiting factor.
The closure of Bell Labs is one of the tragedies of the 20th century.
The closure of Bell Labs was the result of one of the greatest farces of the 20th century.
(Yes, the telecommunications act did address some very real problems that needed attention. Unfortunately the solution was in many ways, worse than the problem itself)
Other options:
- Anything by Sunn O))))
- Saint of the Pit by Diamanda Galas
- Any song without guitars (or traditional instruments for that matter. La lettre d'explicationby Yann Tiersen comes to mind).
On a more serious note, I'd love to see some Math/Post-rock type stuff on Guitar Hero. Stuff by Mogwai, Radiohead, Sigur Ros, or 65daysofstatic would be fun and challenging. It'd be even cooler if you managed to work something like Fuck Buttons into the guitar hero controller scheme.
(If the developers are particularly evil, they'll integrate Rock Band's vocal functions into the Sigur Ros tracks)
True story: My computer randomly picks a song out of my music library to play as my alarm clock in the morning.
I was late for work the day it picked John Cage's 4'33"
(I eventually came to hate most of my music library as a result of this mechanism, and have now relegated the script to limit its selections to a playlist of songs I don't particularly like (and don't also happen to consist only of 4 and a half minutes of silence)