The system is named after a Danish king Harald Blåtand (Harold I of Denmark in English, kong Harald Blåtann in Norwegian), King of Denmark and Norway from 935 and 936 respectively, to 940 known for his unification of previously warring tribes from Denmark (including Scania, present-day Sweden, where the Bluetooth technology was invented) and Norway. Bluetooth likewise was intended to unify different technologies like computers and mobile phones. The Bluetooth logo merges the Nordic runes analogous to the modern Latin H and B: and . The name may have been inspired less by the historical Harald than the loose interpretation of him in The Long Ships by Frans Gunnar Bengtsson, a Swedish best-selling Viking-inspired novel.
This logo is similar to an older logo for Beauknit Textiles, a division of Beauknit Corporation. That logo, using the obvious connection of a reversed K and B for Beauknit, is wider and has rounded corners, but is otherwise the same.
The name was originally only a code-name for the project, but ended up sticking.
I work as the head of visualization at an HPC center, and I have to say that you are right in many cases. However, there are plenty of times when such a facility, or a Powerwall, is warranted. Let me try to outline some of the use cases:
1. Public relations, presentation. These facilities are used all the time to present scientific results to program managers, collaborators, funding agencies, and the like. Don't underestimate the power of these types of presentations. Though it's not "real science," they can often convince people with the purse strings of the important/quality of the "real science."
2. Collaboration. Displaying and discussing scientific results while crowded around someone's laptop sucks. Even a 30" Apple display can only accomodate so many people. These large facilities allow for entire working groups to discuss results and calculations. That's fairly valuable for scientific collaboration.
3. High-resolution data. There are some simulation domains that have such high spatial resolution that you simply can't see a single time step in full detail without a large display. Imagine a hydrodynamics instability problem calculated on a 2k^3 grid. Your monitor can't display that. Or a molecular dynamics calculation with a billion atoms. Map/GIS data also fits into this category easily.
4. Data aggregation. Even for low-resolution data, it can be useful to lay out multiple timesteps on one display to see them all at once. Or if you have a simulation with a large numbers of degrees of freedom, you might want to see, say, 20 of those variables at once. You can't do that on an office display. You need a high-resolution display to do that data aggregation and correlation.
Those are some of the basic use cases that I've seen in my work. I hope this discussion is useful.
I've said this in a number of other comments, but I'll highlight it here. Here is the best line that I've ever put in my ~/.vimrc file:
map ;:
That globally maps the semicolon key to do what the colon key normally does. I lose the "search for last character within the current line" feature that semicolon used to do, but I've never missed it.
Do you have access to his IP address? I know/. does, but that's not public information. To put it very specifically - I am an employer, and use the net to do a bit of research on the people I consider hiring. But I have no clue what IP address that AC is posting from, so I have no way of correlating his/. posts, should I have any desire to. Seems to fit the 90%, if not the 99%, solution.
A deficiency in mathematics skills is "innumeracy," a counterpart to "illiteracy." The scary part is that people nurture innumeracy as if it were a thing to be proud of. Imagine if people took innumeracy as seriously as they did illiteracy. The literacy rate is well trumpeted as a measure of a society's success. Imagine if the numeracy rate were as widely reported and remarked upon.
One thing that I don't get... Why is "pocketing" money a bad thing? No one ever actually "pockets" money. They just about always spend it on something else. Even if it's just putting it in the bank so they can build up interest for later, it's actually giving it to the bank to spend on something else. Money never sits in once place, unless you actually secret it away in your mattress.
And it's that movement of money that helps everyone. The increase in the movement of money, the "acceleration" of the economy, as it were, helps increase the economy and provide the resources for better wages, better technology, all of the "better" things that people have been mentioning in this discussion.
One comment that many people are making here is that it would be very expensive and difficult for Apple to start a cellular service, that it would be a capital expense that just isn't up their alley. The fine article address this to an extent:
Getting into the cellular services business is easier these days, thanks to the likes of Sprint, T-Mobile, Verizon, Cingular, and their support for Mobile Virtual Network Operators (MVNO). An MVNO is a company that leases bandwidth, infrastructure, and time from a carrier and then rebrands and resells it. The best-known MVNO in the US is arguably Virgin Mobile, which also operates on Sprint's network.
The idea being, of course, that Apple could become an MVNO and rebrand the service that another company provides. I can easily see them doing this. To an extent, it's what they're already providing with iDisk. The disks and bandwidth are actually leased from another company, but rebranded and repackaged as an "Apple" experience.
It was a GUI/workflow app, not a hardware control system. Diagnostic Radiology is when a radiologist is looking at your images (think of the digital equivalent of a lightbox on the wall). Our system could bring up the patient's historical exams (useful for cancer patients), cross-reference exams in different orientations (coronal, saggital, and axial), make annotations, and dictate a diagnosis into an automated dictation system.
That makes sense. With this in mind, Java works perfectly in this situation. I would probably have chosen PyQt, but I'm quirky that way. Java's probably an even better solution. Congratulations on a successful project!
Radiation treatment is the kind of system where there's a gun to turn on/off. This is a system some friends of mine have built that is truly amazing. If you have a lung tumor near the spine, this thing takes into account your lungs' expansion while you breathe (in realtime) as it administers the dose. The math is, umm, a bit stunning.
That's impressive. I've worked with people who do the tomographic reconstruction from the raw data from MRI/CT machines, and I've had them explain some of the math to me. Impressive stuff.
Am I correct in assuming that Java wasn't the system which actually interfaced to the hardware control systems? It sounds like it was primarily the GUI at the user end, and other software (possibly embedded) handled the actual hardware control mechanisms that turned the radiation beam on and off.
The groups mentioned include every organization capable of doing the proposed "impartial research". This is a false argument, allowing the author to simply discount any group which arrives at the inconvenient conclusion that action is necessary.
I agree. And even if the conclusion reached is contrary to the views of the author, it should be able to be debated on scientific merits alone.
If there is any valid doubt that the activity of people is contributing to the warming trend (as there almost certainly is), then surely there are articles containing discussions of the actual data and methodology used to arrive at the questionable conclusions by people versed in the relevant sciences. This is simply not such an article.
I think you're right. I know for a fact that there are less inflammatory articles that question the source of global warming, since I have read many myself. Honestly, the article I liked to was the first one I googled up as a counterexample of the "prevailing wisdom." I wanted to give a rapid response, so I turned quickly to google.
[T]his article does not at all dispute the observations of global warming.
And, in fact, it says, "Greens say, rightly, that the best scientific assessment today is that global warming is occurring."
More below.
Otherwise, this is not an article that discusses specific scientific observations
You only found one instance of a specific scientific observation? The first sentence gives a specific instance of low temperature observations. The second paragraph discusses a span of recent large-scale global cooling. The sixth paragraph makes a general observation about the accuracy of global climate models. There are still more direct observations in the article.
Let's step back a minute. The point of bringing this article to light was to illustrate that the recent attention about global warming, and its proposed anthropogenic source, may be a bit precipitous, given the accuracy of the predictive modelling of global climate simulations. While I've seen results of global climate simulations that extend out 150 years, I've yet to see any good data that give me confidence that our accuracy is high enough to say more than "global warming is not much than a mildly promising theory." And specifically "global warming" = "anthropogenic causation of global warming". This was the claim that I was attempting to bolster.
Please, global warming is a fact, the man-made greenhouse effect is the theory that is being questioned. Please keep them separate. When you question global warming you have to back it up with proof that the temperature measurements from the past century are wrong.
In scientific visualisation (biochemical in my case) people don't use ATI's or NVidia's "g4m3rz" cards (GeForce, Radeon), people use Quadro/FireGL cards instead.
Ah, I believe I understand you (and the GGP post) now. The distinction is consumer level cards vs. "professional". I'm not sure I see a lot of distinction. I can't speak to the ATI cards, but the differences in a desktop display for the nVidia Quadro vs. their consumer card aren't going to be as large as they used to be. The architecture of the 7800 GTX is amazingly powerful in relation to a Quadro 4400, for instance. Now, when I build powerwalls and such, I definitely go Quadro for unified frame buffer and synchronization ability (genlock, swaplock, framelock). But for just desktop performance, I don't often sink $2-3K into an individual's machine. It happens, but not often. A consumer level card (top of the line consumer level, mind you), often does the trick, while also saving me some bucks.
As for Linux, some desktops (workstations) here actually do run that, but we mostly use Solaris and Win2k. There's the odd Irix box hanging around as well. We only run these things on workstations, there's no big iron Linux clusters or anything like that involved in pure visualisation work. Maybe in advanced molecular simulation, but that's not happening at my dept, and they don't need their power to be in any frigging graphics card.
For visualization that can fit into the disk/RAM/GPU of a desktop workstation, you're absolutely right. Windows, Solaris, Linux, heck, even Macintosh to some level do the job. There are enough cross-platform tools around that it doesn't really matter.
As for big iron, it depends on how much data you have to visualize. If your I/O rates can only be acheived by parallel I/O, clusters are your game. Same thing with raw processing power throughput. And if you're going for large frame buffer sizes, there's no substitute for a Linux cluster with fast PCIe-attached GPUs and Infiniband to composite your image. It doesn't matter whether the parallel requirements come from molecular dynamics, climate simulation, or astrophysics calculations. It's just data processing, at some level.
I have a customer here who wants to process/visualize 100,000,000 particles at once. Not entirely sure how I'm going to do it, but it's likely going to need our visualization cluster to make it happen. I just don't see it happening on the desktop.
Sadly, only for members of the ACM. You can't even register for access unless you're a member.
What are you talking about? I'm not a member of the ACM, yet I was able to get to the direct link here. I dunno whether that link has a session identifier in it, but it could work.
Short answer: click the little "Pdf" link right next to "Full text".
Slashdot Mirror
Your wish is the market's command: http://www.esato.com/board/viewtopic.php?topic=753 95
Catholic priests have 3x the prostate cancer rate of normal men.
= Retrieve&db=PubMed&list_uids=7242091&dopt=Abstract
Please provide a citation. The first study I found claims the opposite: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd
I work as the head of visualization at an HPC center, and I have to say that you are right in many cases. However, there are plenty of times when such a facility, or a Powerwall, is warranted. Let me try to outline some of the use cases:
1. Public relations, presentation. These facilities are used all the time to present scientific results to program managers, collaborators, funding agencies, and the like. Don't underestimate the power of these types of presentations. Though it's not "real science," they can often convince people with the purse strings of the important/quality of the "real science."
2. Collaboration. Displaying and discussing scientific results while crowded around someone's laptop sucks. Even a 30" Apple display can only accomodate so many people. These large facilities allow for entire working groups to discuss results and calculations. That's fairly valuable for scientific collaboration.
3. High-resolution data. There are some simulation domains that have such high spatial resolution that you simply can't see a single time step in full detail without a large display. Imagine a hydrodynamics instability problem calculated on a 2k^3 grid. Your monitor can't display that. Or a molecular dynamics calculation with a billion atoms. Map/GIS data also fits into this category easily.
4. Data aggregation. Even for low-resolution data, it can be useful to lay out multiple timesteps on one display to see them all at once. Or if you have a simulation with a large numbers of degrees of freedom, you might want to see, say, 20 of those variables at once. You can't do that on an office display. You need a high-resolution display to do that data aggregation and correlation.
Those are some of the basic use cases that I've seen in my work. I hope this discussion is useful.
Nope, doesn't affect it at all. I've programmed in C, C++, perl, and shell with that mapping for years.
Try it and see.
Please do tell me how you get the ":" without hitting two keys at the same time.
.vimrc.
map ; :
The best line I have in my
With this, doing
Please explain how you manage to type ":" without pressing two buttons at once.
.vimrc line ever.
map ; :
The best
Do you have access to his IP address? I know /. does, but that's not public information. To put it very specifically - I am an employer, and use the net to do a bit of research on the people I consider hiring. But I have no clue what IP address that AC is posting from, so I have no way of correlating his /. posts, should I have any desire to. Seems to fit the 90%, if not the 99%, solution.
Then can you tell me why significant useful science (understanding about solidification at the atomic level) was found with this even larger MD simulation: http://pda.physorg.com/lofi-news-simulations-natio nal-nuclear_8651.html
Amen, brother.
A deficiency in mathematics skills is "innumeracy," a counterpart to "illiteracy." The scary part is that people nurture innumeracy as if it were a thing to be proud of. Imagine if people took innumeracy as seriously as they did illiteracy. The literacy rate is well trumpeted as a measure of a society's success. Imagine if the numeracy rate were as widely reported and remarked upon.
One thing that I don't get... Why is "pocketing" money a bad thing? No one ever actually "pockets" money. They just about always spend it on something else. Even if it's just putting it in the bank so they can build up interest for later, it's actually giving it to the bank to spend on something else. Money never sits in once place, unless you actually secret it away in your mattress.
And it's that movement of money that helps everyone. The increase in the movement of money, the "acceleration" of the economy, as it were, helps increase the economy and provide the resources for better wages, better technology, all of the "better" things that people have been mentioning in this discussion.
What, are you in a Rush?
It was a GUI/workflow app, not a hardware control system. Diagnostic Radiology is when a radiologist is looking at your images (think of the digital equivalent of a lightbox on the wall). Our system could bring up the patient's historical exams (useful for cancer patients), cross-reference exams in different orientations (coronal, saggital, and axial), make annotations, and dictate a diagnosis into an automated dictation system.
That makes sense. With this in mind, Java works perfectly in this situation. I would probably have chosen PyQt, but I'm quirky that way. Java's probably an even better solution. Congratulations on a successful project!
Radiation treatment is the kind of system where there's a gun to turn on/off. This is a system some friends of mine have built that is truly amazing. If you have a lung tumor near the spine, this thing takes into account your lungs' expansion while you breathe (in realtime) as it administers the dose. The math is, umm, a bit stunning.
That's impressive. I've worked with people who do the tomographic reconstruction from the raw data from MRI/CT machines, and I've had them explain some of the math to me. Impressive stuff.
Am I correct in assuming that Java wasn't the system which actually interfaced to the hardware control systems? It sounds like it was primarily the GUI at the user end, and other software (possibly embedded) handled the actual hardware control mechanisms that turned the radiation beam on and off.
The king knows Kung Fu.
Shouldn't that be King Fu?
-ba-dum-dum!-
The groups mentioned include every organization capable of doing the proposed "impartial research". This is a false argument, allowing the author to simply discount any group which arrives at the inconvenient conclusion that action is necessary.
I agree. And even if the conclusion reached is contrary to the views of the author, it should be able to be debated on scientific merits alone.
If there is any valid doubt that the activity of people is contributing to the warming trend (as there almost certainly is), then surely there are articles containing discussions of the actual data and methodology used to arrive at the questionable conclusions by people versed in the relevant sciences. This is simply not such an article.
I think you're right. I know for a fact that there are less inflammatory articles that question the source of global warming, since I have read many myself. Honestly, the article I liked to was the first one I googled up as a counterexample of the "prevailing wisdom." I wanted to give a rapid response, so I turned quickly to google.
[T]his article does not at all dispute the observations of global warming.
And, in fact, it says, "Greens say, rightly, that the best scientific assessment today is that global warming is occurring."
More below.
Otherwise, this is not an article that discusses specific scientific observations
You only found one instance of a specific scientific observation? The first sentence gives a specific instance of low temperature observations. The second paragraph discusses a span of recent large-scale global cooling. The sixth paragraph makes a general observation about the accuracy of global climate models. There are still more direct observations in the article.
Let's step back a minute. The point of bringing this article to light was to illustrate that the recent attention about global warming, and its proposed anthropogenic source, may be a bit precipitous, given the accuracy of the predictive modelling of global climate simulations. While I've seen results of global climate simulations that extend out 150 years, I've yet to see any good data that give me confidence that our accuracy is high enough to say more than "global warming is not much than a mildly promising theory." And specifically "global warming" = "anthropogenic causation of global warming". This was the claim that I was attempting to bolster.
Please, global warming is a fact, the man-made greenhouse effect is the theory that is being questioned. Please keep them separate. When you question global warming you have to back it up with proof that the temperature measurements from the past century are wrong.
Okay. How about this?
Can someone explain to me whether or not time actually exists?
Yes. Read this.
Brilliant!
Arg. Make that 100,000,000,000. 100 billion. (Or 100 thousand million if you're British.)
In scientific visualisation (biochemical in my case) people don't use ATI's or NVidia's "g4m3rz" cards (GeForce, Radeon), people use Quadro/FireGL cards instead.
Ah, I believe I understand you (and the GGP post) now. The distinction is consumer level cards vs. "professional". I'm not sure I see a lot of distinction. I can't speak to the ATI cards, but the differences in a desktop display for the nVidia Quadro vs. their consumer card aren't going to be as large as they used to be. The architecture of the 7800 GTX is amazingly powerful in relation to a Quadro 4400, for instance. Now, when I build powerwalls and such, I definitely go Quadro for unified frame buffer and synchronization ability (genlock, swaplock, framelock). But for just desktop performance, I don't often sink $2-3K into an individual's machine. It happens, but not often. A consumer level card (top of the line consumer level, mind you), often does the trick, while also saving me some bucks.
As for Linux, some desktops (workstations) here actually do run that, but we mostly use Solaris and Win2k. There's the odd Irix box hanging around as well.
We only run these things on workstations, there's no big iron Linux clusters or anything like that involved in pure visualisation work. Maybe in advanced molecular simulation, but that's not happening at my dept, and they don't need their power to be in any frigging graphics card.
For visualization that can fit into the disk/RAM/GPU of a desktop workstation, you're absolutely right. Windows, Solaris, Linux, heck, even Macintosh to some level do the job. There are enough cross-platform tools around that it doesn't really matter.
As for big iron, it depends on how much data you have to visualize. If your I/O rates can only be acheived by parallel I/O, clusters are your game. Same thing with raw processing power throughput. And if you're going for large frame buffer sizes, there's no substitute for a Linux cluster with fast PCIe-attached GPUs and Infiniband to composite your image. It doesn't matter whether the parallel requirements come from molecular dynamics, climate simulation, or astrophysics calculations. It's just data processing, at some level.
I have a customer here who wants to process/visualize 100,000,000 particles at once. Not entirely sure how I'm going to do it, but it's likely going to need our visualization cluster to make it happen. I just don't see it happening on the desktop.
Sadly, only for members of the ACM. You can't even register for access unless you're a member.
What are you talking about? I'm not a member of the ACM, yet I was able to get to the direct link here. I dunno whether that link has a session identifier in it, but it could work.
Short answer: click the little "Pdf" link right next to "Full text".