Power consumption will indeed be interesting -- it's not immediately obvious that it will be awful. Remember, the dual-core Athlons are clocked slightly slower than the comparable single-core ones, and therefore actually use less power than their sing-core counterparts. The leakage current goes up mighty fast as the switch speed increases...
Several interesting Near-Earth asteroids were recently discovered (2004) in a possibly Earth-crossing orbit. Spectroscopic analysis showed they were made of titanium dioxide, making them very unique objects -- the first titanium asteroids ever found. Then someone realized that they were painted. (Titanium dioxide is the whitener used in most paints). They were, in fact, several of the Apollo Saturn V boosters, which are still tooling around in orbits that bring them close to Earth. One day one of them might hit us.
My point: simply hurling into space is not a good way to get rid of anything. Unless you know what you are about, it is likely to hang around for a very long time and maybe eventually hit you.
I must admit I'm sort of puzzled by the position that you (and a lot of people) have that Linux distros are "only a version or two away from functionality" -- I use MacOS, Linux, and Microsoft Windows on an almost daily basis, and both MacOS and the KDE Linux GUI are much more usable to me than Microsoft. Within XP I'm always wondering where the heck my files got put, or how to do (where is something trivial in either of the other to OS'es), or why my.doc file isn't opening correctly, or why this PowerPoint presentation doesn't display the right fonts, or how to change all the file names in such-and-such folder in some trivial way that doesn't involve clicking and typing a hundred things, or why my workstation seems to think it's in battery mode...
Would you mind, terribly, pointing out, say, five specific things that work better in Microsoft Windows than in X Windows/Ubuntu? (I say five because at least two -- DVD playback and WMA playback -- are obvious DRM issues).
To my knowledge, the U.K. is the only country in which this practice is illegal. In the U.S. the courts have repeatedly held up (1) the doctrine of first sale -- you can sell a physical artifact you bought and (2) the doctrine that incidental and temporary copies made in the course of using software do not activate copyright law.
Provided you don't agree to the "EULA" or such rot, you aren't bound by it.
Hmm... Actually, total n00bs would do better in a modern user-friendly Linux than in Microsoft Windows (which has stayed pretty byzantine, while Linux keeps getting smoother) -- the problem/advantage (depending on whether you advocate open source or Microsoft) is that there are hardly any complete OS virgins left anymore -- enough people use Microsoft Windows that n00bs (rightly or wrongly) feel more comfortable using it.
Certainly there are more specialized biz apps written for Microsoft Windows than for anything else, and for collaborative doc prep Microsoft Office (in a homogeneous Microsoft environment) still wins -- the markup and collaborative features just aren't available anywhere else. On the other hand, for tooling around at home, the Ubuntu environment has long been superior -- full-featured, clean, largely proof against malware, easy to back up, and posessed of a journaling file system. I'd rather set up my grandmother with Ubuntu than Windows any day -- if only to spare her the "ROCK HARD C0KZZ" popups my other family members complain about.
That is sort of a reflexive answer here on/. but it's definitely something to consider.
Their cheapest course is to buy a new hard drive and install Ubuntu on it (mounting the old Windows drive of course), then buy CrossoverOffice to run their old apps. CrossoverOffice costs $35 -- it's a beefed-up WINE that runs a lot more than naked WINE and has a nice GUI installation procedure. I put "dual" in quotes because they probably won't want to boot into the Windows environment again.
I just had to install a new copy of XP on a recycled box, and frankly the installation procedure is tedious, hard to understand, and slow compared to modern Linuxes. It also doesn't come with practically anything useful -- I had to go get basics like a PDF viewer and a non-Internet Exploder browser.
The parent anonymous coward had an excellent point -- the FCC rules under which your wireless access point operates include a clause that you (the owner of the access point) must accept any interference, even interference that causes undesired operation. To my mind, that includes interference that causes child-porn to be downloaded. (After all, that certainly counts as "undesired operation" of the router for most people).
In short, owner/operator beware -- whatever EM radiation hits your router, it's your problem. If you don't intend your network to be shared, you must take active measures to prevent that.
There are two main problems with Ruby as a replacement for "a very ugly language called Perl":
* It's not sufficiently prettier than Perl
* It's not Perl
Perl may look ugly but it is to most programming languages as English is to most other languages. Perl is a brawling, sprawling mess of borrowed, Hamming-optimized idioms that is extremely ugly from the POV of a syntax engineer and extremely expressive from the POV of a fluent speaker.
Ruby is more like Esperanto - elegant, clean, and spoken by practically no-one because it isn't very expressive.
... as did Intel with their previous round of patched-up dual-core machines. The reason AMD's multicore is so much better than Intel's is because AMD provided a much better caching architecture. Intel's 64-bit multicores could be compared to a large V-8 engine stuck behind a tiny VW carburetor -- totally starved for data. AMD's multicores effectively shared one anothers' L2 caches (a big win), and achieved lower latency on RAM fetches (another big win).
If the two giants start to compete on core count, you can bet your family farm that there will be fudging going on over cross-communication, latency, and RAM bandwidth.
The advantage of WP isn't that it's right all the time, it's that it is (through the tireless effort of zillions of people on five-minute breaks) self-correcting. When the AP screwed up their Ken Lay story, it took overnight before a retraction was posted. WP's story is screwed up for 5-20 minutes at a time.
The mainstream media are almost equally susceptible to being hacked -- even if you don't follow wingnuts like Rush Limbaugh or the insane propaganda and political fart-lighting on Fox News, it's not hard to spot gross errors or oversights in news reporting. "Unbiased" news doesn't exist, investigative reporting isn't anymore, and the media circus is just that -- a circus. Wikipedia may be raw, uncensored, or wrong, but at least it tends to correct itself rapidly.
For what it's worth, the science articles are rapidly becoming the most comprehensive archive of science knowledge ever aimed at the general public. (Of course the refereed literature is larger, but it's not a reference work for the layperson).
... when Einstein first derived it. Remember, "E=mc^2" is just the first term in the Taylor expansion for relativistic mass of a moving body:
E = m c^2 sqrt( 1 / (1 - v^2/c^2) )
which expands to the approximation
E ~ m c^2 + m v^2/2 +...
and recovers the classical kinetic energy equation (that second term) from the Lorenz contraction formulae.
Einstein is reputed to have worked for a while to try to explain away the mc^2 constant term on the front (which doesn't affect classical motion since it is constant), but it was not measurable until nuclear decay was characterized. Chemical reactions don't release enough energy for the binding-energy mass loss to be measurable, but nuclear reactions due. Every (non-failing) chemistry student is familiar with the mass deficit in bound nuclei (the atomic mass of hydrogen is more than 1/12 the atomic mass of C-12, because the C-12 nucleus is tightly bound and lost some mass/energy when it stuck together).
My point is that the mere fact that something is not measurable today does not make it completely senseless. The fact that nuclear mass deficits and corresponding energy loss during radioactive decay agreed with Einstein's relation was a major early win for Einsteinian relativity.
Yes we care -- surprise package delivery...
on
Shuttle Launch Success
·
· Score: 2, Interesting
I had a conversation with Pete Worden about exactly this issue, back when he was head of the USAF Space Command. He pointed out that the big issue is "surprise package delivery". If anyone with $50M can own his very own reusable manned vehicle, then anyone with $50M can put pretty much whatever he wants wherever he wants with just 45 minutes' notice.
On reflection, that's pretty scary: a nav system capable of a rendezvous on-orbit is also capable of rendezvous with other similarly sized objects such as the White House.
The problem is that NOAA doesn't actually own the main satellites that it is using -- ACE (for solar wind sampling) and SOHO (for solar imaging) are both NASA satellites that are intended for research. SOHO and ACE deliver real-time data on an as-available basis. They don't have the same level of reliability and systems redundancy that a weather satellite would have.
Perhaps more importantly, both ACE and SOHO are aging (SOHO is nearly 11 years old, compared to its original 2-year mission) and there is no currently planned mission to replace the space-weather-relevant instruments (the coronagraph on SOHO and the solar wind samplers on ACE) when those instruments ultimately fail. (the Solar Dynamics Observatory has surface imaging but no coronagraph).
If registers start policing spam on their sites, they will have stepped onto a steep, slippery slope that leads to policing content.
Spam is a problem, but handing even more power over to the registrars is not the answer to that problem.
Registrars, ISPs, politicians, and diapers need to be changed frequently -- for approximately the same reasons(*). If I had any accounts with GoDaddy I'd be switching to Dotster or one of thousands of other registrars right now.
What you are missing is that the interpreted, vectorized languages all vectorize to avoid having to live within the interpreter all the time: if you say (in PDL) "$a=7*$b+2" and $b contains a million values, essentially all the processor's time is spent inside a fast loop written in C, iterating over those values. For light-to-medium tasks, you win big because you get the flexibility of an interpreted language and vector notation, and you don't have to wait for the interpreter to run (as you would with loops written in the JIT language).
But that strategy works best only when the variables actually fit in cache or when the CPU runs at the same speed as the memory bus.
I do a lot of interactive data processing. I use PDL a variant of Perl (which, recall, is JIT-compiled) that is designed for array handling. For most of what I do PDL is great -- the CPU spends most of its time waiting for me to make up my mind what I want to do, and moving my ponderously slow fingers to type the command at 110 baud. But some of the stuff I do (magnetohydrodynamic simulation) is exremely CPU-bound, and that stuff I write in C.
A lot of folks use languages like PDL, IDL, MatLab, Octave, or even NumPy to do array processing, and tout the fact that for large arrays those languages run "essentially as fast as C". But that's bullshit. All those languages vectorize their operations in exactly the wrong order - if you have a hundred million datapoints and you want to do six operations on each one, each of those vectorized languages will dutifully swap each of your hundred million datapoints out of RAM into the processor, multiply it by seven (or whatever), and push it back out to RAM before pulling them all back in to add six to each one. What you really want is to vectorize in pipeline order, doing all the operations you plan to on each data point once and for all so that you can take advantage of your processor's nice, fast cache. Nobody has (to my knowledge) figured out a way to do that, that is robust enough for an interactive/JIT language, so just writing it in "C" and getting the loops nested in the right order can speed you up by a factor of more than 10 on a modern AMD or Intel CPU.
What we really want in the Microsoft OS, is access to some of the user tools and easy administration that comes with the GUI, while having the ability to control application separation, get better resource utilization, be hardware agnostic and stop rebuilding installs all the time...
Oh, I get it. So what we really want from the Microsoft OS is Ubuntu.
...and manage the system as well as other enterprise OS's have in the past
Ye Gods, I hope not. UWB is absolutely terrific so long as a limited number of people use it -- but it's one of those solutions that sound great until you multiply by 10,000,000 installed devices -- then everyone's radio noise floor goes up, stealing bandwidth (range, really) from things like FM music, shortwave, air traffic control, and emergency services. By that time it's too late, because you can't track down and eliminate 10^7 devices -- short of nuking the city centers.
Slashdot Mirror
Power consumption will indeed be interesting -- it's not immediately obvious that it will be awful. Remember, the dual-core Athlons are clocked slightly slower than the comparable single-core ones, and therefore actually use less power than their sing-core counterparts. The leakage current goes up mighty fast as the switch speed increases...
Since Hoboken wasn't making copies of the software, it is difficult to see how they are in violation of the authors' copyright.
Several interesting Near-Earth asteroids were recently discovered (2004) in a possibly Earth-crossing orbit. Spectroscopic analysis showed they were made of titanium dioxide, making them very unique objects -- the first titanium asteroids ever found. Then someone realized that they were painted. (Titanium dioxide is the whitener used in most paints). They were, in fact, several of the Apollo Saturn V boosters, which are still tooling around in orbits that bring them close to Earth. One day one of them might hit us.
My point: simply hurling into space is not a good way to get rid of anything. Unless you know what you are about, it is likely to hang around for a very long time and maybe eventually hit you.
Thanks! That helps me understand a lot better what people are concerned about. Specifics good, flames bad. :-)
Cheers,
Craig
I must admit I'm sort of puzzled by the position that you (and a lot of people) have that Linux distros are "only a version or two away from functionality" -- I use MacOS, Linux, and Microsoft Windows on an almost daily basis, and both MacOS and the KDE Linux GUI are much more usable to me than Microsoft. Within XP I'm always wondering where the heck my files got put, or how to do (where is something trivial in either of the other to OS'es), or why my .doc file isn't opening correctly, or why this PowerPoint presentation doesn't display the right fonts, or how to change all the file names in such-and-such folder in some trivial way that doesn't involve clicking and typing a hundred things, or why my workstation seems to think it's in battery mode...
Would you mind, terribly, pointing out, say, five specific things that work better in Microsoft Windows than in X Windows/Ubuntu? (I say five because at least two -- DVD playback and WMA playback -- are obvious DRM issues).
Cheers,
Craig
To my knowledge, the U.K. is the only country in which this practice is illegal. In the U.S. the courts have repeatedly held up (1) the doctrine of first sale -- you can sell a physical artifact you bought and (2) the doctrine that incidental and temporary copies made in the course of using software do not activate copyright law.
Provided you don't agree to the "EULA" or such rot, you aren't bound by it.
IANAL.
Hmm... Actually, total n00bs would do better in a modern user-friendly Linux than in Microsoft Windows (which has stayed pretty byzantine, while Linux keeps getting smoother) -- the problem/advantage (depending on whether you advocate open source or Microsoft) is that there are hardly any complete OS virgins left anymore -- enough people use Microsoft Windows that n00bs (rightly or wrongly) feel more comfortable using it.
Certainly there are more specialized biz apps written for Microsoft Windows than for anything else, and for collaborative doc prep Microsoft Office (in a homogeneous Microsoft environment) still wins -- the markup and collaborative features just aren't available anywhere else. On the other hand, for tooling around at home, the Ubuntu environment has long been superior -- full-featured, clean, largely proof against malware, easy to back up, and posessed of a journaling file system. I'd rather set up my grandmother with Ubuntu than Windows any day -- if only to spare her the "ROCK HARD C0KZZ" popups my other family members complain about.
That is sort of a reflexive answer here on /. but it's definitely something to consider.
Their cheapest course is to buy a new hard drive and install Ubuntu on it (mounting the old Windows drive of course), then buy CrossoverOffice to run their old apps. CrossoverOffice costs $35 -- it's a beefed-up WINE that runs a lot more than naked WINE and has a nice GUI installation procedure. I put "dual" in quotes because they probably won't want to boot into the Windows environment again.
I just had to install a new copy of XP on a recycled box, and frankly the installation procedure is tedious, hard to understand, and slow compared to modern Linuxes. It also doesn't come with practically anything useful -- I had to go get basics like a PDF viewer and a non-Internet Exploder browser.
The parent anonymous coward had an excellent point -- the FCC rules under which your wireless access point operates include a clause that you (the owner of the access point) must accept any interference, even interference that causes undesired operation. To my mind, that includes interference that causes child-porn to be downloaded. (After all, that certainly counts as "undesired operation" of the router for most people).
In short, owner/operator beware -- whatever EM radiation hits your router, it's your problem. If you don't intend your network to be shared, you must take active measures to prevent that.
Therefore you should encrypt or MAC-filter your connections, to satisfy your TOS with your ISP.
There are two main problems with Ruby as a replacement for "a very ugly language called Perl":
* It's not sufficiently prettier than Perl
* It's not Perl
Perl may look ugly but it is to most programming languages as English is to most other languages. Perl is a brawling, sprawling mess of borrowed, Hamming-optimized idioms that is extremely ugly from the POV of a syntax engineer and extremely expressive from the POV of a fluent speaker.
Ruby is more like Esperanto - elegant, clean, and spoken by practically no-one because it isn't very expressive.
... as did Intel with their previous round of patched-up dual-core machines. The reason AMD's multicore is so much better than Intel's is because AMD provided a much better caching architecture. Intel's 64-bit multicores could be compared to a large V-8 engine stuck behind a tiny VW carburetor -- totally starved for data. AMD's multicores effectively shared one anothers' L2 caches (a big win), and achieved lower latency on RAM fetches (another big win).
If the two giants start to compete on core count, you can bet your family farm that there will be fudging going on over cross-communication, latency, and RAM bandwidth.
The advantage of WP isn't that it's right all the time, it's that it is (through the tireless effort of zillions of people on five-minute breaks) self-correcting. When the AP screwed up their Ken Lay story, it took overnight before a retraction was posted. WP's story is screwed up for 5-20 minutes at a time.
The mainstream media are almost equally susceptible to being hacked -- even if you don't follow wingnuts like Rush Limbaugh or the insane propaganda and political fart-lighting on Fox News, it's not hard to spot gross errors or oversights in news reporting. "Unbiased" news doesn't exist, investigative reporting isn't anymore, and the media circus is just that -- a circus. Wikipedia may be raw, uncensored, or wrong, but at least it tends to correct itself rapidly.
For what it's worth, the science articles are rapidly becoming the most comprehensive archive of science knowledge ever aimed at the general public. (Of course the refereed literature is larger, but it's not a reference work for the layperson).
which expands to the approximation
and recovers the classical kinetic energy equation (that second term) from the Lorenz contraction formulae.
Einstein is reputed to have worked for a while to try to explain away the mc^2 constant term on the front (which doesn't affect classical motion since it is constant), but it was not measurable until nuclear decay was characterized. Chemical reactions don't release enough energy for the binding-energy mass loss to be measurable, but nuclear reactions due. Every (non-failing) chemistry student is familiar with the mass deficit in bound nuclei (the atomic mass of hydrogen is more than 1/12 the atomic mass of C-12, because the C-12 nucleus is tightly bound and lost some mass/energy when it stuck together).
My point is that the mere fact that something is not measurable today does not make it completely senseless. The fact that nuclear mass deficits and corresponding energy loss during radioactive decay agreed with Einstein's relation was a major early win for Einsteinian relativity.
I had a conversation with Pete Worden about exactly this issue, back when he was head of the USAF Space Command. He pointed out that the big issue is "surprise package delivery". If anyone with $50M can own his very own reusable manned vehicle, then anyone with $50M can put pretty much whatever he wants wherever he wants with just 45 minutes' notice.
On reflection, that's pretty scary: a nav system capable of a rendezvous on-orbit is also capable of rendezvous with other similarly sized objects such as the White House.
Yes.
Perhaps more importantly, both ACE and SOHO are aging (SOHO is nearly 11 years old, compared to its original 2-year mission) and there is no currently planned mission to replace the space-weather-relevant instruments (the coronagraph on SOHO and the solar wind samplers on ACE) when those instruments ultimately fail. (the Solar Dynamics Observatory has surface imaging but no coronagraph).
If registers start policing spam on their sites, they will have stepped onto a steep, slippery slope that leads to policing content.
Spam is a problem, but handing even more power over to the registrars is not the answer to that problem.
Registrars, ISPs, politicians, and diapers need to be changed frequently -- for approximately the same reasons(*).
If I had any accounts with GoDaddy I'd be switching to Dotster or one of thousands of other registrars right now.
(*)apologies to Heinlein
Denver today had a record high of 101F for any recorded June 14 ever. It was also
the earliest-in-the-year 100F temperature ever recorded here.
What you are missing is that the interpreted, vectorized languages all vectorize to avoid having to live within the interpreter all the time: if you say (in PDL) "$a=7*$b+2" and $b contains a million values, essentially all the processor's time is spent inside a fast loop written in C, iterating over those values. For light-to-medium tasks, you win big because you get the flexibility of an interpreted language and vector notation, and you don't have to wait for the interpreter to run (as you would with loops written in the JIT language).
But that strategy works best only when the variables actually fit in cache or when the CPU runs at the same speed as the memory bus.
A lot of folks use languages like PDL, IDL, MatLab, Octave, or even NumPy to do array processing, and tout the fact that for large arrays those languages run "essentially as fast as C". But that's bullshit. All those languages vectorize their operations in exactly the wrong order - if you have a hundred million datapoints and you want to do six operations on each one, each of those vectorized languages will dutifully swap each of your hundred million datapoints out of RAM into the processor, multiply it by seven (or whatever), and push it back out to RAM before pulling them all back in to add six to each one. What you really want is to vectorize in pipeline order, doing all the operations you plan to on each data point once and for all so that you can take advantage of your processor's nice, fast cache. Nobody has (to my knowledge) figured out a way to do that, that is robust enough for an interactive/JIT language, so just writing it in "C" and getting the loops nested in the right order can speed you up by a factor of more than 10 on a modern AMD or Intel CPU.
Oh, I get it. So what we really want from the Microsoft OS is Ubuntu.
or, rather RHEL.
Ye Gods, I hope not. UWB is absolutely terrific so long as a limited number of people use it -- but it's one of those solutions that sound great until you multiply by 10,000,000 installed devices -- then everyone's radio noise floor goes up, stealing bandwidth (range, really) from things like FM music, shortwave, air traffic control, and emergency services. By that time it's too late, because you can't track down and eliminate 10^7 devices -- short of nuking the city centers.
... which took a lot less time to load despite being 100% as functional.
I object to your using my login name in your post.
--Dr. Zowie(TM)