"Which is more efficacious? To take up as much simultaneous processor time as possible in order to finish faster, or to leave extra cores open for other processes to run simultaneously."
This is up to the OS to decide. The programmer's job is to provide something for the processors to do. If doing it serially is wasteful, doing it in parallel (or, at least, asynchronously) is the way to go.
Of course, when you think parallel and threads rely on each other's data, you need a suite of tests that can test it properly. Either that, or you risk losing endless nights tracking nasty real-time bugs.
Concurrent is clearly the way to go, as we are increasingly close to a performance wall with single-threaded sequential stuff.
I don't think it's about the time it takes to transfer a single bit but the amount of bits that can be transmitted at once with light rather than wires. If we can talk line-of-sight transmission between boards, it's easy to line up an array of about a million emitters with an array of a million detectors and send back and forth the same amount of data you would need a couple thousand wires (taking translation times into account) to do.
Sun is a very entertaining company to watch. Even when their gizmos never end up in products, they are always cool.
It's not about the video definition - it's about storage capacity and transfer rate. You can't fit a huge dataset with detailed models and textures that really use the raw processing power these consoles have in a DVD disk. More than that - game developers love to be able to pack more data within their games. BD beat HD-DVD both on capacity and transfer rate. The DVD media the stock Xbox is capable of reading isn't even in the same league.
As for video, even the HD offerings that exist now on Apple TV don't come close to the video quality of a HD title on a plastic carrier. Expecting them to send you up to 20 GB of data is ludicrous for now and for the near future. MS may go that route with the next generation consoles but for this one, it's game over. Xbox will have only DVD titles.
I am not sure. Can you point to the full text? What are the exact provisions specified in the chapter where 2202 and 3314 are?
And, BTW, what kind of punishment is reserved for people who violate such provisions and jeopardize the proper storage of presidential records?
Most probably, the current administration would get away making such records classified and blocking any investigation on such issues, on grounds they would compromise national security.
"Real ID is an attempt to eliminate the Mexican Government from "assisting" in getting Driver's Licenses to illegals"
You know that's a pretty serious thing to say. Do you have any corroborating evidence? If so, you should contact the the federal authorities as soon as possible.
I see no connection between the existence or not of a unique identity validation system and the results of years of incompetent foreign policy that allowed such extremist groups to thrive. IIRC, the hijackers didn't need to lie about who they were to board and hijack those planes and their real names were there in the passenger list.
That hijacking subverted the commonly accepted rules of the time. Before 9/11, the planes and its passengers were considered valuable hostages that would be used for bargaining and the possibility of using jetliners as missiles was not considered a real threat except for the government officials who were aware of the possibility. I sincerely doubt a couple guys with knives and a fake bomb could hijack a plane these days. The more likely outcome would be that the hijacker would be subdued or killed by the passengers.
You can and should refuse to follow orders that are unethical or illegal. Not to say this is the case. Failure to maintain proper records is one thing, but I think that any storage device that hosted sensitive information is necessarily destroyed when it's EOL'ed.
Perhaps the law should be changed and make it mandatory to store the devices (or a properly audited image) for a fixed period just like proper backups, for the purpose of keeping proper records of all government activity. Failure to do so should carry hefty penalties and could be interpreted as an attempt to obstruct justice.
"Maybe out planet, with it's skin lain bare to the cosmos, is an exception for a life-harboring world."
That sounds very likely. Life on Earth depends on many things that seem to be rare - a strong magnetic field that protects us from our own sun, just right temperature - so that there is liquid water - just right atmosphere - so that there is no runaway greenhouse effect like Venus - and so on. With all the mass extinctions that happened here before we came, we could consider ourselves to be an extremely lucky race.
Because if we weren't so lucky, there would be nobody wondering how lucky we are.
Not entirely new. The biggest problem seems to be, by far, the power source (for both general operations and communications).
Landing is easier with a thick atmosphere. Temperature seems stabler than on Mars - a thick atmosphere supposedly helps with that too. The fact it rains could help with dust if there is any (but it could pose a problem to any rubber seals). The communications problem could be partly solved with an orbiting relay that could, in itself, do some studies from orbit - with it, the rover could have a lower power high-bandwidth link with the orbiter. It would be useful to be able to communicate directly with Earth in case something goes wrong with the orbiter part, even if with lower bandwidth. More smarts in the rover itself could make up for the added comm delay.
Other kinds of challenge arise from the environment being far more dynamic than Mars. When mountains move a couple miles in months, that's because the ground is less grounded than usual. I suspect the life expectancy of any surface structures (that could include rovers) in such environment is much shorter than on less interesting places. I would not expect a nuclear-powered Huygens-like probe to last a couple months there before being crushed by a very quick mountain.
Perhaps, instead of a rover, a long-life balloon could be designed to float just above the surface but, then, that would be your entirely new design.
I would love if the Calc could support an arbitrarily large number of rows or columns. There are times 1024 columns won't cut it.
Besides, Excell 2007 has the bragging rights here - up to 65535 columns, IIRC.
In order to achieve more market penetration, OOo needs to do things Office can't.
Can't Sun hire a bunch of financial analysts, scientists, statisticians or anything like that to ask them to write a wish list of what a dream spreadsheet should do, implement it before Microsoft and gather some good PR for OOo?
And, while we are at it, I would love to see numbers with error margins as valid numeric data. That would be the single largest paradigm shift on spreadsheets since Visicalc and it's one badly needed. If it's done right, it will be a gateway to uses spreadsheets have never been able to touch. I suspect many financial and business analysts would love to see this one.
Presenter also could use some fancy 3D transitions.
Not all of the partnerships gave results as bad as the Rama sequels. I loved Richter 10. Sadly, the other guy (can't remember his name right now) died shortly after writing the book.
the editors in heaven are probably expecting to publish some seriously good books next year;-)
Now, seriously, I can't imagine how this got modded insightful. Funny, perhaps.
Before someone pretends to know many different programming languages (something that is a sign of a good programmer), the person should at least know how to spell their names. Haskell is called, well, "Haskell", not "HASCAL"
And a string in early BASIC (MS-BASIC, Applesoft, circa 1980) is named something like X$, not $X (would be a scalar type in Perl, but I can't imagine what you had in mind). And, BTW, you wouldn't compare it to an integer.
And, before you start comparing the most rudimentary constructs in 5 different languages (4, really and 4 somewhat similar ones) and bragging you can figure them all out, perhaps you should check a little deeper under their hood.
"Sometimes it's nice to see the bean counters in suits being totally ignored and just seeing something done because it's there."
Not only because it's there, but because it's beautiful and inspiring. Machines like the Apollo, the Saturn V and the Concorde are very effective statements of what Man can achieve.
At 100Km and 3,500 Km/h you are hardly out of the gravity well. Accelerating to LEO or escape velocity (out of the gravity well) is one thing - you just need a bigger, more (a whole lot more) powerful rocket. Painfully hard to do, but possible. De-accelerating back to a more or less normal landing is a bitch - using atmosphere and friction is, to say the least, dangerous. Carrying fuel is prohibitively expensive. There are no easy solutions for that one.
Hopefully this technology will bring high-altitude high-speed passenger flight into the realm of possibility. Only when people figure out how to build space planes that can safely get to space and back once a day sub-orbital travel will be viable.
And we don't even need to go that fast, just being high above the atmosphere on a lazy ballistic trajectory to my destination is cool enough for me.
the Cassini sensors and computers are not that much like the embedded computer in your car engine. Perhaps the situation that triggers the bug only happened during the flyby and never before during the trip. Exploring Saturn's icy moons is hardly routine operations.
But, to some extent, I must agree. Events like this present an opportunity to improve testing and simulation. Perhaps when they get what went wrong, processes will be improved and things like this one do not happen again.
The problem here is not wanting to stick with currently proven designs, but the hideous cost of certifying a new design. It is so expensive to re-certify a project after a design change people really don't want to do it often.
The certification process probably makes the design safer, but it also disincentives innovation in ways that would horrify someone used to the rapid pace of consumer electronics.
On the other hand, the kind of reliability standards we see on consumer electronics would horrify me if they ever happened be applied to a nuclear facility or an airplane.
I always find the use of the term "non-weapon-grade" as "perfectly safe to be left unattended" disturbing. Just because it can't be used to create a fission bomb doesn't mean it can't be used as a weapon. A whole sort of dirty bombs come to my mind.
Slashdot Mirror
"Which is more efficacious? To take up as much simultaneous processor time as possible in order to finish faster, or to leave extra cores open for other processes to run simultaneously."
This is up to the OS to decide. The programmer's job is to provide something for the processors to do. If doing it serially is wasteful, doing it in parallel (or, at least, asynchronously) is the way to go.
Of course, when you think parallel and threads rely on each other's data, you need a suite of tests that can test it properly. Either that, or you risk losing endless nights tracking nasty real-time bugs.
Concurrent is clearly the way to go, as we are increasingly close to a performance wall with single-threaded sequential stuff.
I don't think it's about the time it takes to transfer a single bit but the amount of bits that can be transmitted at once with light rather than wires. If we can talk line-of-sight transmission between boards, it's easy to line up an array of about a million emitters with an array of a million detectors and send back and forth the same amount of data you would need a couple thousand wires (taking translation times into account) to do.
Sun is a very entertaining company to watch. Even when their gizmos never end up in products, they are always cool.
It's not about the video definition - it's about storage capacity and transfer rate. You can't fit a huge dataset with detailed models and textures that really use the raw processing power these consoles have in a DVD disk. More than that - game developers love to be able to pack more data within their games. BD beat HD-DVD both on capacity and transfer rate. The DVD media the stock Xbox is capable of reading isn't even in the same league.
As for video, even the HD offerings that exist now on Apple TV don't come close to the video quality of a HD title on a plastic carrier. Expecting them to send you up to 20 GB of data is ludicrous for now and for the near future. MS may go that route with the next generation consoles but for this one, it's game over. Xbox will have only DVD titles.
I am not sure. Can you point to the full text? What are the exact provisions specified in the chapter where 2202 and 3314 are?
And, BTW, what kind of punishment is reserved for people who violate such provisions and jeopardize the proper storage of presidential records?
Most probably, the current administration would get away making such records classified and blocking any investigation on such issues, on grounds they would compromise national security.
That's really sad.
"Real ID is an attempt to eliminate the Mexican Government from "assisting" in getting Driver's Licenses to illegals"
You know that's a pretty serious thing to say. Do you have any corroborating evidence? If so, you should contact the the federal authorities as soon as possible.
I see no connection between the existence or not of a unique identity validation system and the results of years of incompetent foreign policy that allowed such extremist groups to thrive. IIRC, the hijackers didn't need to lie about who they were to board and hijack those planes and their real names were there in the passenger list.
That hijacking subverted the commonly accepted rules of the time. Before 9/11, the planes and its passengers were considered valuable hostages that would be used for bargaining and the possibility of using jetliners as missiles was not considered a real threat except for the government officials who were aware of the possibility. I sincerely doubt a couple guys with knives and a fake bomb could hijack a plane these days. The more likely outcome would be that the hijacker would be subdued or killed by the passengers.
You can and should refuse to follow orders that are unethical or illegal. Not to say this is the case. Failure to maintain proper records is one thing, but I think that any storage device that hosted sensitive information is necessarily destroyed when it's EOL'ed.
Perhaps the law should be changed and make it mandatory to store the devices (or a properly audited image) for a fixed period just like proper backups, for the purpose of keeping proper records of all government activity. Failure to do so should carry hefty penalties and could be interpreted as an attempt to obstruct justice.
"how does JavaScript 2.0 protect against this?"
I would guess the first implementations will not allow code to be downloaded from a site/tree other than the one the page belongs to.
"The main problem is that given MicroSoft's history, I'm not sure I trust it."
Having known Microsoft since the early 80s, I can tell you I won't.
"Maybe out planet, with it's skin lain bare to the cosmos, is an exception for a life-harboring world."
That sounds very likely. Life on Earth depends on many things that seem to be rare - a strong magnetic field that protects us from our own sun, just right temperature - so that there is liquid water - just right atmosphere - so that there is no runaway greenhouse effect like Venus - and so on. With all the mass extinctions that happened here before we came, we could consider ourselves to be an extremely lucky race.
Because if we weren't so lucky, there would be nobody wondering how lucky we are.
"will take an entirely new design."
Not entirely new. The biggest problem seems to be, by far, the power source (for both general operations and communications).
Landing is easier with a thick atmosphere. Temperature seems stabler than on Mars - a thick atmosphere supposedly helps with that too. The fact it rains could help with dust if there is any (but it could pose a problem to any rubber seals). The communications problem could be partly solved with an orbiting relay that could, in itself, do some studies from orbit - with it, the rover could have a lower power high-bandwidth link with the orbiter. It would be useful to be able to communicate directly with Earth in case something goes wrong with the orbiter part, even if with lower bandwidth. More smarts in the rover itself could make up for the added comm delay.
Other kinds of challenge arise from the environment being far more dynamic than Mars. When mountains move a couple miles in months, that's because the ground is less grounded than usual. I suspect the life expectancy of any surface structures (that could include rovers) in such environment is much shorter than on less interesting places. I would not expect a nuclear-powered Huygens-like probe to last a couple months there before being crushed by a very quick mountain.
Perhaps, instead of a rover, a long-life balloon could be designed to float just above the surface but, then, that would be your entirely new design.
It's a nice OS. Too bad it lacks a decent text editor ;-) /me ducks
Turn in your geek card when you leave.
What did you mean in line 30? Shell-style variable substitution? You can't do that in BASIC!
At least not in any BASIC I know of.
"(Now comes the obligatory "you should be using a database!" post.)"
Spreadsheet users can't care less about how it's implemented as long as they can manipulate their data the way they would like to.
Being able to summon such immense data sets within a familiar tool ought to be really cool.
I would love if the Calc could support an arbitrarily large number of rows or columns. There are times 1024 columns won't cut it.
Besides, Excell 2007 has the bragging rights here - up to 65535 columns, IIRC.
In order to achieve more market penetration, OOo needs to do things Office can't.
Can't Sun hire a bunch of financial analysts, scientists, statisticians or anything like that to ask them to write a wish list of what a dream spreadsheet should do, implement it before Microsoft and gather some good PR for OOo?
And, while we are at it, I would love to see numbers with error margins as valid numeric data. That would be the single largest paradigm shift on spreadsheets since Visicalc and it's one badly needed. If it's done right, it will be a gateway to uses spreadsheets have never been able to touch. I suspect many financial and business analysts would love to see this one.
Presenter also could use some fancy 3D transitions.
Seriously, it's in the April issue.
Not all of the partnerships gave results as bad as the Rama sequels. I loved Richter 10. Sadly, the other guy (can't remember his name right now) died shortly after writing the book.
;-)
the editors in heaven are probably expecting to publish some seriously good books next year
"Methodoligy", "widly used", "very tight", "Sciencetific", "favorates", "deveation"?!
Why not English as your next language?
Now, seriously, I can't imagine how this got modded insightful. Funny, perhaps.
Before someone pretends to know many different programming languages (something that is a sign of a good programmer), the person should at least know how to spell their names. Haskell is called, well, "Haskell", not "HASCAL"
And a string in early BASIC (MS-BASIC, Applesoft, circa 1980) is named something like X$, not $X (would be a scalar type in Perl, but I can't imagine what you had in mind). And, BTW, you wouldn't compare it to an integer.
And, before you start comparing the most rudimentary constructs in 5 different languages (4, really and 4 somewhat similar ones) and bragging you can figure them all out, perhaps you should check a little deeper under their hood.
Even Slashdot has some Digg days...
"Sometimes it's nice to see the bean counters in suits being totally ignored and just seeing something done because it's there."
Not only because it's there, but because it's beautiful and inspiring. Machines like the Apollo, the Saturn V and the Concorde are very effective statements of what Man can achieve.
At 100Km and 3,500 Km/h you are hardly out of the gravity well. Accelerating to LEO or escape velocity (out of the gravity well) is one thing - you just need a bigger, more (a whole lot more) powerful rocket. Painfully hard to do, but possible. De-accelerating back to a more or less normal landing is a bitch - using atmosphere and friction is, to say the least, dangerous. Carrying fuel is prohibitively expensive. There are no easy solutions for that one.
Hopefully this technology will bring high-altitude high-speed passenger flight into the realm of possibility. Only when people figure out how to build space planes that can safely get to space and back once a day sub-orbital travel will be viable.
And we don't even need to go that fast, just being high above the atmosphere on a lazy ballistic trajectory to my destination is cool enough for me.
How exactly does a more-or-less desktop PC with more-or-less mundane parts qualify as a supercomputer?
Is it about the transparent parts?
the Cassini sensors and computers are not that much like the embedded computer in your car engine. Perhaps the situation that triggers the bug only happened during the flyby and never before during the trip. Exploring Saturn's icy moons is hardly routine operations.
But, to some extent, I must agree. Events like this present an opportunity to improve testing and simulation. Perhaps when they get what went wrong, processes will be improved and things like this one do not happen again.
The problem here is not wanting to stick with currently proven designs, but the hideous cost of certifying a new design. It is so expensive to re-certify a project after a design change people really don't want to do it often.
The certification process probably makes the design safer, but it also disincentives innovation in ways that would horrify someone used to the rapid pace of consumer electronics.
On the other hand, the kind of reliability standards we see on consumer electronics would horrify me if they ever happened be applied to a nuclear facility or an airplane.
I always find the use of the term "non-weapon-grade" as "perfectly safe to be left unattended" disturbing. Just because it can't be used to create a fission bomb doesn't mean it can't be used as a weapon. A whole sort of dirty bombs come to my mind.
You know... There must be an upper limit on the size of the crater you create. For rocky planets, a couple thousand miles seems to be it. ;-)