I publish an open source package. Here in Europe. Too bad you wouldn't be able to get it. (You expect me to snail-mail a CD to someone in the other regions? HAHAHA!)
Similarly, I get (open source) software/updates from all over the place, and mirroring again isn't an option.
Mars' orbit is at around 1.5 AU: 1.5 times further from the sun than earth.
When earth and Mars are closest, their distance will be about 0.5 AU, when they are furthest (on opposite sides of the sun), about 2.5 AU.
Light travels about 8 minutes per AU, so it's 4 to 20 minutes one way. Ping times would vary between 8 and 40 minutes.
As Mars runs a full circle about every two years, earth only catches up with it (i.e. the "excellent" ping times of just 8 minutes) once every two years.
Instead of about google, the story should now turn to the other search engines: They apparently turned over "personal data" to government people who didn't have just cause!!
I'd be seriously upset if my search engine would give my personal data to just anybody who doesn't have the right to such data.
Microsoft sells software with a very restrictive EULA. The fact that this EULA is illegal in most places doesn't prevent microsoft from pointing out (e.g. towards Ebay) that resale of an otherwise legit copy of windows is against the EULA.
As I understand things, if I happen to buy an extra copy of Windows by accident, the law says I get to sell it (provided I haven't used it or made a copy). Microsoft considers this piracy.
Software is something "soft". If you buy it, the microsof reasoning is that the soft stuff sticks to you. YOU are licenced to use the software and nobody else. If someone else wants to use the licence they have to pay again. From a microsoft-financial standpoint this viewpoint is the preferred standpoint.
Some Ebay sellers may be selling real pirated copies. (i.e. with the activation keys deactivated or something like that). Catching one or two of these provides food for pulling thousands of auctions for legal, extraneous copies of MS Windows.
[i]ask the banks to stop sending you the applications in the first place:[/i]
Try it one day. I have 3 company accounts with one bank. So they send me three fliers for loans in three separate envelopes. With a: "If you don't want to recieve this anymore, send us a note at:... ".
So, I send them a note: I have three accounts, Please stop sending the materials for accounts XXX and YYY, I'll still be notified of interesting things because I still get the junk for account ZZZ.
Two months later, again a whole pile of junk. So, printed the note again, added a note: told you before, and send the stuff back.
And last week, AGAIN some 50 duplcate pages of junkmail... I haven't gotten around to finding my old letter. (ah. Found it. Printing... )
One of the design flaws of the I386 was that it was not possible to fully virtualize. Has this been fixed in recent processors? Otherwise it must be possible for an OS to detect wether or not it's running inside a VM or not....
What many people fail to realize is that it takes "evolution" on the order of tens to hundreds thousands of years to "invent" a gene.
Random mutations have to encode for a new protein that activates in the right cells and "does the right thing". From then on, this is likely to become a "gene": Almost any random mutation will invalidate the protein, and disable the "feature".
Suppose such a new "invention" is not always advantageous. Say, only during an ice age. During ice ages, those carrying the intact encoding for the protein (we say they "have the gene"), will survive best, those that don't have it will drop in numbers. Once such a condition is over (say ice age stops), natural selection suddely starts to favor those that "do not have the gene". Still, as they decend from a population where most had the gene to survive, they remain "genetically close", and the gene will easily activate and proliferate during the next ice age.
A real world example is Sicle Cell Anemia. It is a genetic disease: You're born with or without it. Advantage of HAVING the disease? You don't die of Malaria (you do die of the disease, but most have had children by then).
So depending on the amount of malaria mosquitos around, the percentage of people with the Sicle Cell Anemia gene varies a lot. Natural selection at work!
Now, if you look at 10000 to 15000 years, it is unlikely that "evolution" has "invented" a lot of new genes. That however genes have activated and deactivated is however very likely.
If the "running fast" gene was "mostly essential" 10000 years ago in africa, but now not any more, then natural selection would have ensured that 90-95% of the population had that gene 10000 years ago. Nowadays, there is no longer a selection for-or-against this gene. So, the percentage of the people having the gene will slowly drop (I don't work in the field, I have no idea how fast this goes).
Did you ever notice that different children "don't like" different foods? This is a genetic safeguard to preserve the species. Evolution apparently "invented" that a long time ago.
If five percent of your tribe "Simply doesn't like to eat chicken", and the H5N1 Chicken flue comes around, about 5% of the tribe is likely to survive to pass on a much elevated "don't like chicken" gene.
Most likely the "common knowledge" about what to eat and what not to eat has leveled out the "taste" genes: They no longer significantly influence survival.
I'm not sure if write caching in the flash chips is a good idea...
Flash chips write at around 1Mb per second. Tops. Modern harddrives write around 50Mb per second. You'd need quite a lot of flash chips in parallel to top that.
Of course, you can cache acouple of writes on the laptop drive to prevent having to spin up the disk. But it is not going to be a speed issue.
I'd say that it's possible that this research will show that people who work like that on a permanent basis have less capability for learning new stuff.
The brain in "survival mode" is a wonderous thing. It can do things that you wouldn't hold possible before. Some people might be taking advantage of that.
On the other hand, there must also be some disadvantages: There is no such thing as a free lunch. So Evolution didn't make it the 'default'. Maybe less (or none at all?) sex.
First (mostly at slower speeds) is the grip of the wheels on the street. If the grip-factor is 0.8 you can accellerate at 0.8G, and reach 26.6 m/s in 3.4 seconds.
Secondly, (mostly at higher speeds) the power output of the motor becomes the limiting factor. To continue to accellerate a 0.5 ton car at 8m/s after reaching 25m/s requires 80 kw, 113 Hp.
When you're in the Ferrari range (sub 4 second) the "at slow speeds" applies to most of the 0-60. (i.e. something like 0 - 50mph). (for those cars they sometimes also quote the 0-120mph time).
And in those cases the engine simply has to be able to ramp up to max power in the 0-50 range (while keeping up the max-accelleration the wheels allow).
I estimated the grip-factor at 0.8 because my father's car was once measured at being able to BRAKE 60-0 in 4 seconds (grip factor of 0.7). But that would have been on 4 wheels!
If you ask me: 50 MPG sounds reasonable. (just doing a bit of math, that's 80 km on 4.something liters, right? Any modern diesel car will get that!). You need to get a lot of things EXACTLY RIGHT and perfectly tuned to get 0-60 in 4 seconds.
FYI, the bio-diesel here in Europe sells for 80% of the price of normal diesel. It costs about twice as much to produce, but taxes on normal diesel are close to 150%.
The energy content should be about the same as normal diesel. I found a petrol station in the south of Germany that had the stuff. I just poured it in, paid less than I would normally had, and continued on my trip (to Italy). We ran a bit better mileage on that tank than the others, but this could be because we had to drive very slowly during a significant part of that part of the trip.
So I estimate that your mileage (compared to normal diesel) will be between -5 and +10% according to my (single tank!) measurement.
The reason an auto-placer is usually bad is because it is incapable of seeing the "big picture". I would think that placing a few key components like the big chips would allow an auto-placer to do a very reasonable job in putting the resistors and caps in mostly the right places.
When I place a design manually, the small things get in the way, and sometimes I put a few of them in the design before I'm ready for them. And then I forget to move them, sometimes leading to long traces all over the place when I'm done routing.
There are several ways to find a security bug. One that I think is common is: You run a program, and it crashes. If you've got the right security mindset, that immediately triggers: Hey, the program crashes, maybe it has a flaw that can be exploited.
If something goes wrong, modern programs should display a popup: "can't display page" or something like that. A crash usually means it started scribbling on itself, and that is usually exploitable with the right data-input.
Attaching a debugger and finding out where the crash happens is the next step.
Finding a real exploit from there on is not nobel-prize-level work, but also it is more complicated than "stick a scredriver in the moving parts under the bonnet".
Simply allow them their audit: Here is a room, we've prepared for your audit. It doesn't have any communications links, leave your cell phone at the door. There are two computers with the source code. Feel free to audit it all you want.
It is realtively easy to "open up" an appartment building by using the existing coax cable (for cable television) to transport say 1gbps to that building. It is then relatively easy to transport lots of bandwidth (up to say 1gbps) to the appartments in that building.
I suspect that that's what they are providing. 100 people in an appartment building and one coax cable going to an cable-tv-office a couple of miles away and transporting 100*1Gbps both ways is unsolved, and it likely stays that way the coming years.
After reading the article, they are up to 100mbps now, but think the technology might expand up to 1Gbs in the future.
For CDs there is a load of competition to make the CDs at $0.10 a piece. As you note, if you pay $20 in the shop, most of it is for the content. Someone selling content can probably charge $40 for 50Gb worth of higher-res content instead of the older $20. If the production costs for that 50Gb DVD is $1.00 higher (a tenfold increase) then that will still be a good deal.
Suppose you know a thing or two about Chemistry, and you are teaching a (basic) course in the second year of university. Nothing speculative, just the basics: 2 H2 + O2 -> 2 H2O stuff like that.
Suppose the school board decides that you have to point out to the students that what you're teaching is just a theory, and that you have to point out that there are other theories. (Nobody has SEEN the molecules react!). Someone is appointed to teach the "earth, water, air and fire" theory. How would that make you feel? How's that for science?
The most recent noteworthy news from Rome is that they have found a reasonable way to make science and christianity work together. If a super-being set off the universe a long, long time ago, setting the rules of physics, evolution and chemistry, then they don't see anything inconsistent with the bible.
Now suddenly the "war" between creationists and evolutionists is defused. Good. Science is not inconsistent with the bible. Good.
... still under 10%. Of course they can't sell them profitably at that rate...
Oh? Why not? The cost of producing a pentium chip, (exlcuding R&D writeoff and Yield factorization) is very similar to producing any other chip at the most about $10. But they sell for up to $1000. The ones that sell for $1000 are the ones that come from the 10% yield line, at $10 per chip (bad or good) produced, and a 10% yield, you still get to write off about $900 on R&D for that chip.....
Do you burn DVDs at a rate close to or higher than the maximum printed on the media? Do you use media from the cheapest half of the market?
Plextor media works 100%, the other stuff is great for creating stuff that you will then read back on the drive that wrote it. Or on a drive that you know reads the media from your writer.
If you seal the drive with plexiglass, you have a window of a couple of hours that is tricky. During those hours, the dust that got in will have been caught in the filter, and from then on, chances are the drive will continue to work.
The problem is if you leave an open connection to the environment: If you allow air to go in and out unfiltered, you continue to risk crashing the head. e.g. your friend who removed the rubber.
Some drives even control the ability of the heads to move with a wind-driven interlock mechanism
Most drives do this. Quantums have the wind-driven interlock. Most others do this by having a magnet hold the head-arm in the "parked" position. IBM/Hitachi park the heads OFF the platters.
This last solution allows the heads to sometimes get parked ON the platter, and that results in the motor sometimes not being strong enough to get the drive to spin at all after that.
Some older drives have a solenoid that releases the head assembly.
Re:Too much free time and money.
on
Hard Drive Window
·
· Score: 2, Interesting
Even with a cleanroom it's not trivial: I've ruined two drives trying to do this, using the cleanroom. Twice I managed to get the window too close to a running drive, so that the hub would scrape off some acrylic. From then on, it's game over.
I should take the time to create a sufficient spacer, so that this doesn't happen again.
Slashdot Mirror
I publish an open source package. Here in Europe. Too bad you wouldn't be able to get it. (You expect me to snail-mail a CD to someone in the other regions? HAHAHA!)
Similarly, I get (open source) software/updates from all over the place, and mirroring again isn't an option.
Mars' orbit is at around 1.5 AU: 1.5 times further from the sun than earth.
When earth and Mars are closest, their distance will be about 0.5 AU, when they are furthest (on opposite sides of the sun), about 2.5 AU.
Light travels about 8 minutes per AU, so it's 4 to 20 minutes one way. Ping times would vary between 8 and 40 minutes.
As Mars runs a full circle about every two years, earth only catches up with it (i.e. the "excellent" ping times of just 8 minutes) once every two years.
Instead of about google, the story should now turn to the other search engines: They apparently turned over "personal data" to government people who didn't have just cause!!
I'd be seriously upset if my search engine would give my personal data to just anybody who doesn't have the right to such data.
Microsoft sells software with a very restrictive EULA. The fact that this EULA is illegal in most places doesn't prevent microsoft from pointing out (e.g. towards Ebay) that resale of an otherwise legit copy of windows is against the EULA.
As I understand things, if I happen to buy an extra copy of Windows by accident, the law says I get to sell it (provided I haven't used it or made a copy). Microsoft considers this piracy.
Software is something "soft". If you buy it, the microsof reasoning is that the soft stuff sticks to you. YOU are licenced to use the software and nobody else. If someone else wants to use the licence they have to pay again. From a microsoft-financial standpoint this viewpoint is the preferred standpoint.
Some Ebay sellers may be selling real pirated copies. (i.e. with the activation keys deactivated or something like that). Catching one or two of these provides food for pulling thousands of auctions for legal, extraneous copies of MS Windows.
[i]ask the banks to stop sending you the applications in the first place:[/i]
... )
Try it one day. I have 3 company accounts with one bank. So they send me three fliers for loans in three separate envelopes. With a: "If you don't want to recieve this anymore, send us a note at:... ".
So, I send them a note: I have three accounts, Please stop sending the materials for accounts XXX and YYY, I'll still be notified of interesting things because I still get the junk for account ZZZ.
Two months later, again a whole pile of junk. So, printed the note again, added a note: told you before, and send the stuff back.
And last week, AGAIN some 50 duplcate pages of junkmail... I haven't gotten around to finding my old letter. (ah. Found it. Printing
One of the design flaws of the I386 was that it was not possible to fully virtualize. Has this been fixed in recent processors? Otherwise it must be possible for an OS to detect wether or not it's running inside a VM or not....
What many people fail to realize is that it takes "evolution" on the order of tens to hundreds thousands of years to "invent" a gene.
Random mutations have to encode for a new protein that activates in the right cells and "does the right thing". From then on, this is likely to become a "gene": Almost any random mutation will invalidate the protein, and disable the "feature".
Suppose such a new "invention" is not always advantageous. Say, only during an ice age. During ice ages, those carrying the intact encoding for the protein (we say they "have the gene"), will survive best, those that don't have it will drop in numbers. Once such a condition is over (say ice age stops), natural selection suddely starts to favor those that "do not have the gene". Still, as they decend from a population where most had the gene to survive, they remain "genetically close", and the gene will easily activate and proliferate during the next ice age.
A real world example is Sicle Cell Anemia. It is a genetic disease: You're born with or without it. Advantage of HAVING the disease? You don't die of Malaria (you do die of the disease, but most have had children by then).
So depending on the amount of malaria mosquitos around, the percentage of people with the Sicle Cell Anemia gene varies a lot. Natural selection at work!
Now, if you look at 10000 to 15000 years, it is unlikely that "evolution" has "invented" a lot of new genes. That however genes have activated and deactivated is however very likely.
If the "running fast" gene was "mostly essential" 10000 years ago in africa, but now not any more, then natural selection would have ensured that 90-95% of the population had that gene 10000 years ago. Nowadays, there is no longer a selection for-or-against this gene. So, the percentage of the people having the gene will slowly drop (I don't work in the field, I have no idea how fast this goes).
Did you ever notice that different children "don't like" different foods? This is a genetic safeguard to preserve the species. Evolution apparently "invented" that a long time ago.
If five percent of your tribe "Simply doesn't like to eat chicken", and the H5N1 Chicken flue comes around, about 5% of the tribe is likely to survive to pass on a much elevated "don't like chicken" gene.
Most likely the "common knowledge" about what to eat and what not to eat has leveled out the "taste" genes: They no longer significantly influence survival.
If your OS is WRITING to swap in a "small random blocks" it should reevaluate its swap allocation strategy.
I'm not sure if write caching in the flash chips is a good idea...
Flash chips write at around 1Mb per second. Tops. Modern harddrives write around 50Mb per second. You'd need quite a lot of flash chips in parallel to top that.
Of course, you can cache acouple of writes on the laptop drive to prevent having to spin up the disk. But it is not going to be a speed issue.
I'd say that it's possible that this research will show that people who work like that on a permanent basis have less capability for learning new stuff.
The brain in "survival mode" is a wonderous thing. It can do things that you wouldn't hold possible before. Some people might be taking advantage of that.
On the other hand, there must also be some disadvantages: There is no such thing as a free lunch. So Evolution didn't make it the 'default'. Maybe less (or none at all?) sex.
There are two things limiting "0-60" speed.
First (mostly at slower speeds) is the grip of the wheels on the street. If the grip-factor is 0.8 you can accellerate at 0.8G, and reach 26.6 m/s in 3.4 seconds.
Secondly, (mostly at higher speeds) the power output of the motor becomes the limiting factor. To continue to accellerate a 0.5 ton car at 8m/s after reaching 25m/s requires 80 kw, 113 Hp.
When you're in the Ferrari range (sub 4 second) the "at slow speeds" applies to most of the 0-60. (i.e. something like 0 - 50mph). (for those cars they sometimes also quote the 0-120mph time).
And in those cases the engine simply has to be able to ramp up to max power in the 0-50 range (while keeping up the max-accelleration the wheels allow).
I estimated the grip-factor at 0.8 because my father's car was once measured at being able to BRAKE 60-0 in 4 seconds (grip factor of 0.7). But that would have been on 4 wheels!
If you ask me: 50 MPG sounds reasonable. (just doing a bit of math, that's 80 km on 4.something liters, right? Any modern diesel car will get that!). You need to get a lot of things EXACTLY RIGHT and perfectly tuned to get 0-60 in 4 seconds.
FYI, the bio-diesel here in Europe sells for 80% of the price of normal diesel. It costs about twice as much to produce, but taxes on normal diesel are close to 150%.
The energy content should be about the same as normal diesel. I found a petrol station in the south of Germany that had the stuff. I just poured it in, paid less than I would normally had, and continued on my trip (to Italy). We ran a bit better mileage on that tank than the others, but this could be because we had to drive very slowly during a significant part of that part of the trip.
So I estimate that your mileage (compared to normal diesel) will be between -5 and +10% according to my (single tank!) measurement.
The reason an auto-placer is usually bad is because it is incapable of seeing the "big picture". I would think that placing a few key components like the big chips would allow an auto-placer to do a very reasonable job in putting the resistors and caps in mostly the right places.
When I place a design manually, the small things get in the way, and sometimes I put a few of them in the design before I'm ready for them. And then I forget to move them, sometimes leading to long traces all over the place when I'm done routing.
There are several ways to find a security bug. One that I think is common is: You run a program, and it crashes. If you've got the right security mindset, that immediately triggers: Hey, the program crashes, maybe it has a flaw that can be exploited.
If something goes wrong, modern programs should display a popup: "can't display page" or something like that. A crash usually means it started scribbling on itself, and that is usually exploitable with the right data-input.
Attaching a debugger and finding out where the crash happens is the next step.
Finding a real exploit from there on is not nobel-prize-level work, but also it is more complicated than "stick a scredriver in the moving parts under the bonnet".
Simply allow them their audit: Here is a room, we've prepared for your audit. It doesn't have any communications links, leave your cell phone at the door. There are two computers with the source code. Feel free to audit it all you want.
Apparently there are systems around where users can't look in others' homedirs. Fine.
/home/public; chmod 777 /home/public
for files that others on the system can be trusted with:
# mkdir
for files that require "for your eyes only":
jmorrison% uuencode myfile myfile | mail -s "here is the file" bsmith
problem solved.
It is realtively easy to "open up" an appartment building by using the existing coax cable (for cable television) to transport say 1gbps to that building. It is then relatively easy to transport lots of bandwidth (up to say 1gbps) to the appartments in that building.
I suspect that that's what they are providing. 100 people in an appartment building and one coax cable going to an cable-tv-office a couple of miles away and transporting 100*1Gbps both ways is unsolved, and it likely stays that way the coming years.
After reading the article, they are up to 100mbps now, but think the technology might expand up to 1Gbs in the future.
For CDs there is a load of competition to make the CDs at $0.10 a piece. As you note, if you pay $20 in the shop, most of it is for the content. Someone selling content can probably charge $40 for 50Gb worth of higher-res content instead of the older $20. If the production costs for that 50Gb DVD is $1.00 higher (a tenfold increase) then that will still be a good deal.
Suppose you know a thing or two about Chemistry, and you are teaching a (basic) course in the second year of university. Nothing speculative, just the basics: 2 H2 + O2 -> 2 H2O stuff like that.
Suppose the school board decides that you have to point out to the students that what you're teaching is just a theory, and that you have to point out that there are other theories. (Nobody has SEEN the molecules react!). Someone is appointed to teach the "earth, water, air and fire" theory. How would that make you feel? How's that for science?
The most recent noteworthy news from Rome is that they have found a reasonable way to make science and christianity work together. If a super-being set off the universe a long, long time ago, setting the rules of physics, evolution and chemistry, then they don't see anything inconsistent with the bible.
Now suddenly the "war" between creationists and evolutionists is defused. Good. Science is not inconsistent with the bible. Good.
Oh? Why not? The cost of producing a pentium chip, (exlcuding R&D writeoff and Yield factorization) is very similar to producing any other chip at the most about $10. But they sell for up to $1000. The ones that sell for $1000 are the ones that come from the 10% yield line, at $10 per chip (bad or good) produced, and a 10% yield, you still get to write off about $900 on R&D for that chip.....
Didn't anybody notice that they are producing ROM, i.e. factory mastered disks?
Do you burn DVDs at a rate close to or higher than the maximum printed on the media? Do you use media from the cheapest half of the market?
Plextor media works 100%, the other stuff is great for creating stuff that you will then read back on the drive that wrote it. Or on a drive that you know reads the media from your writer.
If you seal the drive with plexiglass, you have a window of a couple of hours that is tricky. During those hours, the dust that got in will have been caught in the filter, and from then on, chances are the drive will continue to work.
The problem is if you leave an open connection to the environment: If you allow air to go in and out unfiltered, you continue to risk crashing the head. e.g. your friend who removed the rubber.
Some drives even control the ability of the heads to move with a wind-driven interlock mechanism
Most drives do this. Quantums have the wind-driven interlock. Most others do this by having a magnet hold the head-arm in the "parked" position. IBM/Hitachi park the heads OFF the platters.
This last solution allows the heads to sometimes get parked ON the platter, and that results in the motor sometimes not being strong enough to get the drive to spin at all after that.
Some older drives have a solenoid that releases the head assembly.
Even with a cleanroom it's not trivial: I've ruined two drives trying to do this, using the cleanroom. Twice I managed to get the window too close to a running drive, so that the hub would scrape off some acrylic. From then on, it's game over.
I should take the time to create a sufficient spacer, so that this doesn't happen again.