That would be nice, but until that change to Wikipedia happens, Veropedia should continue its fork. Merging the two projects back together again in the future will be trivial, since Veropedia will have a database of good wikipedia article version numbers. (If Veropedia actually forked the page, then remerging could be difficult.)
That's allowing the perfect to interfere with the good. Just because you aren't going to do a full variance study doesn't mean you can't eyeball the variation in the meter while doing the measurement and tell us what you think the spread is.
So that's one piece of the uncertainty on the measurement. The next piece we need is how much the power reading fluctuates while the computer is in a "steady" state. Using my Kill-A-Watt, I've seen short time variations of a few watts on a computer (though it drew more power than a laptop).
You do realize that a dual core Athlon64 is many times faster than an Athlon Thunderbird? You can easily underclock an Athlon 64 until it uses less than 74W, or grab the laptop version with much lower power consumption, and it would still outperform a T-bird. (I grabbed the first power measurements I could find and assumed people could do the scaling in their head.) Modern CPUs also do the power throttling you describe already.
By going to smaller transistors, lower voltages, and more clever power management schemes, they have managed to get more work done per watt than before. A new 3 GHz Athlon64 X2 requires 89W of power, whereas the old 1.4 GHz Athlon Thunderbird used 74W.
One could roughly estimate the variance by looking at the meter fluctuations while taking the reading, or checking the design accuracy of the meter in the manufacturer's data sheet. You need some kind of estimate if you are going to draw any conclusions (which the authors of TFA were attempting to do).
Some simple upgrades are still relatively easy to do on the laptop. In particular, you could upgrade the memory in every laptop I've owned using a screwdriver and removing 2 or 4 screws. The hard drive is also replaceable in many laptops by removing a few screws (major exception being the iBook, and now the MacBook Pro, so I hear). The only relatively common upgrade you can't do is replace the video "card."
Less common upgrades, like swapping CPUs, or stripping the case to install a new motherboard, are impossible, of course.
It seems a number of programs, like WinXP Genuine Advantage (talk about newspeak...) and Mathematica, like to use the MAC address of the network interface as part of their computer fingerprinting algorithm. While debugging a networking problem in a guest OS, I removed and re-added the virtual ethernet device in the VM configuration. Parallels generates a new MAC address for every ethernet device you create, so the new device was distinguishable from the old one. XP immediately complained that something had changed, and I would need to reactivate my installation. Thankfully it worked, but now I write down the MAC address of my virtual machines so I don't accidentally do that again.
One thing that was drilled into us in an Engineering Communications class was to assume your audience (often management) was impatient, had limited reading comprehension, and generally ignorant of your subject matter. At the time we thought this was amusing, as we imagined the standard Dilbert stereotype of a manager.
Looking back now, I see this was more of a mental exercise than a statement about our future bosses' intellectual abilities. Engineers tend to be detail-oriented, especially about their particular work. This is generally good, because details matter in implementation, but bad for communication if it clutters up the main points you are trying to convey. By telling engineers to write like their audience is stupid and lazy, you might end up with something that is almost understandable.:)
In reality, your boss might not be an expert in the field, and they also have lots of information flying at them from all directions. Making prose simple and compact speeds comprehension for busy people. Unfortunately, people who are predisposed to have a negative attitude toward management (bad previous managers, overly large nerd egos, social insecurity, etc) just remember this advice as "Write simply because my boss is dumb."
At the moment, data is easy to create, but it is also easy to destroy, especially by accident. The constant churn in storage technologies and file formats ensures that anything which entropy does not destroy might become effectively unreadable in 10-20 years anyway. As it stands now, our digital short term memory lasts maybe decades without well considered, active maintenance.
Think about all digital photos that will certainly be gone in 50 years. (Not that this will be entirely a bad thing. The future probably doesn't want photos of people drinking beer while wearing pirate hats.)
I'd like to see more places quote availability numbers as a percentage AND the longest downtime interval in the previous year. 12 minutes of downtime every night at midnight sucks, but 3 solid days of downtime is a disaster.
Yes, but operating on 2 doubles or 4 floats at a time is only barely "vector" processing. Hopefully the merger of ATI and AMD will bring some wider vector units closer to the CPU.
I currently manage my non-science (physics grad student here) PDFs with Yep, but it is clearly aimed more generically, and includes scanner support for doing things like scanning receipts and other paper documents. Papers looks good, though I wish there was integration with the arXiv, which is the "PubMed" of physics. Hopefully the Papers authors will make good on their intention to add a plugin API for importers, exporters and search engines.
I think another likely answer is that there is no hook to allow Google to replace just the Spotlight index and search algorithm while preserving the Apple UI.
I'm watching it run right now, and Google didn't reinvent the wheel, exactly. Google Desktop is running mdimport (the program that invokes the Spotlight plugins to convert files to collections of terms) in the background. What Google is providing is a replacement/supplement for the Spotlight search interface, but not all of the Spotlight software stack. This is how Google Desktop takes advantage of all your existing Spotlight Importer plugins. (Which are damn easy to write. Props to Apple for that.)
Spotlight's indexing could use some improvement, so I'm looking forward to seeing how Google Desktop performs on my large collection of PDF and Postscript files. Spotlight doesn't seem to do very intelligent ranking of the documents it returns, so unless the search terms are fairly unique, the results can be impossible to sift through. Hopefully Google (or maybe 10.5) will improve that.
The recent Google paper about disk failures that pointed out that SMART is not as reliable an indicator of impending disk death as you would hope. One third of the failed disks had no SMART errors reported at all. (Two thirds is not bad, of course, but you really want something a bit more deterministic.)
Every page has to start with some small, intrinsic amount of karma, otherwise there would be none to pass around. By creating enough bogus pages, you can aggregate some amount of link karma to bestow on the site of your choosing. In principle, I guess this would devalue everyone's PageRank too (kind of like printing money), but for a while it could be profitable.
The second hole is the popularity of websites with user-generated content. Lots of highly ranked websites (like/. in fact) allow anyone, or almost anyone, to add arbitrary links to pages, thereby redirecting some small fraction of the sites possibly large link karma to any place they want. This can also be used to gather karma from insignificant websites (like the thousands and thousands of semi-dead blogs with comments enabled) in mass quantities. It's like the urban myth of the bank scam where someone gets rich stealing all the fractional cents left over in interest calculations.
Of course, these are only problems for the original PageRank algorithm. It's pretty clear that Google has modified in several ways to fight these problems, such as through the introduction of the "nofollow" link attribute.
Oh, that is disappointing. All of Apple's computers for the past year at least have gigabit, so you would have expected the router to support it as well. I guess they figure the average use is not computer to computer, but computer to internet via broadband...
Slashdot Mirror
The iFlea is even better: http://www.youtube.com/watch?v=UvPj22jANDw
That would be nice, but until that change to Wikipedia happens, Veropedia should continue its fork. Merging the two projects back together again in the future will be trivial, since Veropedia will have a database of good wikipedia article version numbers. (If Veropedia actually forked the page, then remerging could be difficult.)
That's allowing the perfect to interfere with the good. Just because you aren't going to do a full variance study doesn't mean you can't eyeball the variation in the meter while doing the measurement and tell us what you think the spread is.
So that's one piece of the uncertainty on the measurement. The next piece we need is how much the power reading fluctuates while the computer is in a "steady" state. Using my Kill-A-Watt, I've seen short time variations of a few watts on a computer (though it drew more power than a laptop).
You do realize that a dual core Athlon64 is many times faster than an Athlon Thunderbird? You can easily underclock an Athlon 64 until it uses less than 74W, or grab the laptop version with much lower power consumption, and it would still outperform a T-bird. (I grabbed the first power measurements I could find and assumed people could do the scaling in their head.) Modern CPUs also do the power throttling you describe already.
By going to smaller transistors, lower voltages, and more clever power management schemes, they have managed to get more work done per watt than before. A new 3 GHz Athlon64 X2 requires 89W of power, whereas the old 1.4 GHz Athlon Thunderbird used 74W.
One could roughly estimate the variance by looking at the meter fluctuations while taking the reading, or checking the design accuracy of the meter in the manufacturer's data sheet. You need some kind of estimate if you are going to draw any conclusions (which the authors of TFA were attempting to do).
Wish (mostly) granted: Sam & Max
Nope, some of us still lurk. :)
Yes, and they are quite expensive. 32 GB will run you about $14k: HyperDrive 4
One word: Pandora
Best thing ever. I've found and purchased 4 albums from artists I never would have heard of otherwise.
I think ODF is an ISO standard now: http://www.iso.org/iso/iso_catalogue/catalogue_tc/catalogue_detail.htm?csnumber=43485
Some simple upgrades are still relatively easy to do on the laptop. In particular, you could upgrade the memory in every laptop I've owned using a screwdriver and removing 2 or 4 screws. The hard drive is also replaceable in many laptops by removing a few screws (major exception being the iBook, and now the MacBook Pro, so I hear). The only relatively common upgrade you can't do is replace the video "card."
Less common upgrades, like swapping CPUs, or stripping the case to install a new motherboard, are impossible, of course.
It seems a number of programs, like WinXP Genuine Advantage (talk about newspeak...) and Mathematica, like to use the MAC address of the network interface as part of their computer fingerprinting algorithm. While debugging a networking problem in a guest OS, I removed and re-added the virtual ethernet device in the VM configuration. Parallels generates a new MAC address for every ethernet device you create, so the new device was distinguishable from the old one. XP immediately complained that something had changed, and I would need to reactivate my installation. Thankfully it worked, but now I write down the MAC address of my virtual machines so I don't accidentally do that again.
One thing that was drilled into us in an Engineering Communications class was to assume your audience (often management) was impatient, had limited reading comprehension, and generally ignorant of your subject matter. At the time we thought this was amusing, as we imagined the standard Dilbert stereotype of a manager.
Looking back now, I see this was more of a mental exercise than a statement about our future bosses' intellectual abilities. Engineers tend to be detail-oriented, especially about their particular work. This is generally good, because details matter in implementation, but bad for communication if it clutters up the main points you are trying to convey. By telling engineers to write like their audience is stupid and lazy, you might end up with something that is almost understandable. :)
In reality, your boss might not be an expert in the field, and they also have lots of information flying at them from all directions. Making prose simple and compact speeds comprehension for busy people. Unfortunately, people who are predisposed to have a negative attitude toward management (bad previous managers, overly large nerd egos, social insecurity, etc) just remember this advice as "Write simply because my boss is dumb."
At the moment, data is easy to create, but it is also easy to destroy, especially by accident. The constant churn in storage technologies and file formats ensures that anything which entropy does not destroy might become effectively unreadable in 10-20 years anyway. As it stands now, our digital short term memory lasts maybe decades without well considered, active maintenance.
Think about all digital photos that will certainly be gone in 50 years. (Not that this will be entirely a bad thing. The future probably doesn't want photos of people drinking beer while wearing pirate hats.)
I'd like to see more places quote availability numbers as a percentage AND the longest downtime interval in the previous year. 12 minutes of downtime every night at midnight sucks, but 3 solid days of downtime is a disaster.
Yes, but operating on 2 doubles or 4 floats at a time is only barely "vector" processing. Hopefully the merger of ATI and AMD will bring some wider vector units closer to the CPU.
The vector processors never went away. They just became your graphics card: 128 floating point units at your command
BTW, here is a real article on TRIPS.
I currently manage my non-science (physics grad student here) PDFs with Yep, but it is clearly aimed more generically, and includes scanner support for doing things like scanning receipts and other paper documents. Papers looks good, though I wish there was integration with the arXiv, which is the "PubMed" of physics. Hopefully the Papers authors will make good on their intention to add a plugin API for importers, exporters and search engines.
I think another likely answer is that there is no hook to allow Google to replace just the Spotlight index and search algorithm while preserving the Apple UI.
I'm watching it run right now, and Google didn't reinvent the wheel, exactly. Google Desktop is running mdimport (the program that invokes the Spotlight plugins to convert files to collections of terms) in the background. What Google is providing is a replacement/supplement for the Spotlight search interface, but not all of the Spotlight software stack. This is how Google Desktop takes advantage of all your existing Spotlight Importer plugins. (Which are damn easy to write. Props to Apple for that.)
Spotlight's indexing could use some improvement, so I'm looking forward to seeing how Google Desktop performs on my large collection of PDF and Postscript files. Spotlight doesn't seem to do very intelligent ranking of the documents it returns, so unless the search terms are fairly unique, the results can be impossible to sift through. Hopefully Google (or maybe 10.5) will improve that.
The recent Google paper about disk failures that pointed out that SMART is not as reliable an indicator of impending disk death as you would hope. One third of the failed disks had no SMART errors reported at all. (Two thirds is not bad, of course, but you really want something a bit more deterministic.)
Every page has to start with some small, intrinsic amount of karma, otherwise there would be none to pass around. By creating enough bogus pages, you can aggregate some amount of link karma to bestow on the site of your choosing. In principle, I guess this would devalue everyone's PageRank too (kind of like printing money), but for a while it could be profitable.
The second hole is the popularity of websites with user-generated content. Lots of highly ranked websites (like /. in fact) allow anyone, or almost anyone, to add arbitrary links to pages, thereby redirecting some small fraction of the sites possibly large link karma to any place they want. This can also be used to gather karma from insignificant websites (like the thousands and thousands of semi-dead blogs with comments enabled) in mass quantities. It's like the urban myth of the bank scam where someone gets rich stealing all the fractional cents left over in interest calculations.
Of course, these are only problems for the original PageRank algorithm. It's pretty clear that Google has modified in several ways to fight these problems, such as through the introduction of the "nofollow" link attribute.
Oh, that is disappointing. All of Apple's computers for the past year at least have gigabit, so you would have expected the router to support it as well. I guess they figure the average use is not computer to computer, but computer to internet via broadband...