Google essentially fails to correct for the number of links a page author tends to put in, meaning that authors who just put in a lot of links have a larger effect of google than authors with a sparser linking style.
How do you know this?
The last whitepaper Google published on their design (cool though it is) is *years* upon *years* out of date, and AFAIK they're not handing out info on the specific metrics, and haven't for a long time. They have definitely revised their system since then to avoid various attacks people have tried on it.
They could be doing precisely what you're predicting.
Hardware tweaking is a waste. In almost any real-world situation, maybe you'll get up to 10% more performance.
A software tweak can buy you much, much more. Your system is *definitely* not running twice as slow as your friends because of a lack of "tweaking" the hardware (assuming you aren't doing something bizarre like running the processor at only half the rated frequency or something silly like that).
You'll have to mention the OS and the programs you're using.
Abstract benchmark scores pretty much mean nothing. Say "Apache maxes out at 150 simultaneous connections" or "I only get 20 FPS in foggy parts of Max Payne".
The point of the article isn't about competing "memes", it's about flaws in Google's PageRank system. In this case, the weighting of site importance on the basis of the number and authoritativeness is being thrown by the incestuous linking behavior of weblogs.
First, I don't see it as a flaw.
Second, I'm not sure what one would do that would be considered "better".
Google is a mechanism, but it's not the drive. People picking up new phrases is the drive. The Register itself is part of it -- a popular, respected site, it will now drive the meme further and further.
Google is a good, efficient mechanism for a meme to spread, but it certainly isn't dangerous or evil.
Unlike Napster or Kazaa, which helped create a network of computers that would not have existed otherwise, "Phynd" and the others search a network that already exists.
Okay. Phynd is a straightforward SMB indexing server. As per comments here from one of the RPI students, one of the persons charged wrote some of the Phynd software, and the other person admined a Phynd server for RPI. The RIAA is *not* going after the people who are serving infringing data, but after the CS students who wrote indexing software...because it's more convenient for the RIAA.
When file indexing services become illegal because one of the servers that they index contains potentially infringing information (as just happened), the world has turned completely upside down. Google indexes copyright-infringing images and text every day, and in *far* larger quantities than these SMB indexers. Should *they* be served with a lawsuit and ordered to shut down? How about Yahoo? AllTheWeb has an FTP search engine, not that far from an SMB search engine...is *that* illegal as well? Hell, if you have a multi-user system, a user stores infringing information in his account, and your cron daemon runs updatedb, you're in the same boat as the students that got charged.
I'm very, very uncomfortable with this, and I feel that the RIAA has gone too far.
Probably they will continue to operate their fab (if the lenders will permit that) on a small scale for a few months while they look for a buyer for the Opteron business which will probably go to Intel who will then shut it down.
I doubt Intel will be interested in buying anything but patents.
So typical of shareholders nowadays to think that the board of a company owes you personally something because you bought 3rd hand stock equivilent to like.01% of their initial public funding. Unless you got in on the IPO, you are a liability to the company, not an investor.
Umm...no. The company *does* owe you, personally, the value of your stock. Sorry, but that's just how capitalism works.
And presumably AMD officials feel that Opteron will not do well from a market standpoint. You want to do something like this when your stock is as high as possible, and since the market will have anticipated profit from the release, now is the best time to do somethin like this.
I'm a bit more pessimistic about AMD's CPU division's chances.
I can't figure out where the benefit is for them, unless AMD's processor divison is expected to lose money (spinoffs are great for this -- you can saddle them with debts from the other company. Coca-Cola used to regularly do this.)
Of course, he doesn't know how to *secure* either of them, and IT isn't asked to set up the server, which is why corporate computer security frequently sucks.
Slashdot Mirror
It fills me with pride to know that even if an asteroid pastes me and half the people on earth, my TPS cover sheet templates will still be readable.
There is nothing wrong with the safety of a wooden case.
You are right that the wood probably won't catch fire.
However, wood is a good thermal insulator, and metal a good thermal conductor.
It's likely to raise case temperature quite a bit.
Too bad you're using Windows.
If you were using, say, Linux, you could just mount a NFS or SMB share and make your bookmarks file on the client a symlink to the one on the host.
Google essentially fails to correct for the number of links a page author tends to put in, meaning that authors who just put in a lot of links have a larger effect of google than authors with a sparser linking style.
How do you know this?
The last whitepaper Google published on their design (cool though it is) is *years* upon *years* out of date, and AFAIK they're not handing out info on the specific metrics, and haven't for a long time. They have definitely revised their system since then to avoid various attacks people have tried on it.
They could be doing precisely what you're predicting.
Mention the OS and the programs you use.
Hardware tweaking is a waste. In almost any real-world situation, maybe you'll get up to 10% more performance.
A software tweak can buy you much, much more. Your system is *definitely* not running twice as slow as your friends because of a lack of "tweaking" the hardware (assuming you aren't doing something bizarre like running the processor at only half the rated frequency or something silly like that).
You'll have to mention the OS and the programs you're using.
Abstract benchmark scores pretty much mean nothing. Say "Apache maxes out at 150 simultaneous connections" or "I only get 20 FPS in foggy parts of Max Payne".
The point of the article isn't about competing "memes", it's about flaws in Google's PageRank system. In this case, the weighting of site importance on the basis of the number and authoritativeness is being thrown by the incestuous linking behavior of weblogs.
First, I don't see it as a flaw.
Second, I'm not sure what one would do that would be considered "better".
Google is a mechanism, but it's not the drive. People picking up new phrases is the drive. The Register itself is part of it -- a popular, respected site, it will now drive the meme further and further.
Google is a good, efficient mechanism for a meme to spread, but it certainly isn't dangerous or evil.
clear that Bush didn't give a flying fuck about public opinion.
The GOP will, come elections.
Googlewanking - One-handed Googling, usu. when performing Google searches for pr0n or special interest advocacy.
? There are people that jerk off to "Focus on the Family" or similar?
Call me cynical, but you wouldn't happen to be a webmaster who didn't get ranked highly, would you?
Or Canadian.
Unlike Napster or Kazaa, which helped create a network of computers that would not have existed otherwise, "Phynd" and the others search a network that already exists.
Okay. Phynd is a straightforward SMB indexing server. As per comments here from one of the RPI students, one of the persons charged wrote some of the Phynd software, and the other person admined a Phynd server for RPI. The RIAA is *not* going after the people who are serving infringing data, but after the CS students who wrote indexing software...because it's more convenient for the RIAA.
When file indexing services become illegal because one of the servers that they index contains potentially infringing information (as just happened), the world has turned completely upside down. Google indexes copyright-infringing images and text every day, and in *far* larger quantities than these SMB indexers. Should *they* be served with a lawsuit and ordered to shut down? How about Yahoo? AllTheWeb has an FTP search engine, not that far from an SMB search engine...is *that* illegal as well? Hell, if you have a multi-user system, a user stores infringing information in his account, and your cron daemon runs updatedb, you're in the same boat as the students that got charged.
I'm very, very uncomfortable with this, and I feel that the RIAA has gone too far.
I mean really, its not the RIAA's job to be our parents. Its should be left up to the college.
Umm...I'd say it isn't the college's "job to be our parents" either.
Anyone living in Canberra who would like to take a shot at setting something like this up can feel free to contact me."
Translation: Anyone who wants to do lots of unpaid work for me is free to do so.
You're confusing bonds and stocks.
Bonds are loans with an obligation to pay back money.
Stocks are a certificate of ownership.
Probably they will continue to operate their fab (if the lenders will permit that) on a small scale for a few months while they look for a buyer for the Opteron business which will probably go to Intel who will then shut it down.
I doubt Intel will be interested in buying anything but patents.
So typical of shareholders nowadays to think that the board of a company owes you personally something because you bought 3rd hand stock equivilent to like .01% of their initial public funding. Unless you got in on the IPO, you are a liability to the company, not an investor.
Umm...no. The company *does* owe you, personally, the value of your stock. Sorry, but that's just how capitalism works.
And presumably AMD officials feel that Opteron will not do well from a market standpoint. You want to do something like this when your stock is as high as possible, and since the market will have anticipated profit from the release, now is the best time to do somethin like this.
I'm a bit more pessimistic about AMD's CPU division's chances.
The Flash business is the (relative) cash cow that's keeping the rest of the company afloat. No way it's going to split.
No, they split the CPU divison away, assign it all the debts, and let it die, which gives the more lucrative flash division a big jump on the books.
Coca Cola used to do this with spinoffs.
I can't figure out where the benefit is for them, unless AMD's processor divison is expected to lose money (spinoffs are great for this -- you can saddle them with debts from the other company. Coca-Cola used to regularly do this.)
Of course, he doesn't know how to *secure* either of them, and IT isn't asked to set up the server, which is why corporate computer security frequently sucks.
I believe Apache 2.0 has a Windows GUI, if you're a GUI kinda guy.
You know, I think I'd be willing to be a system administrator if a company was willing to pay $150k for setting up simple web servers.
You have anything to back this up other than a claim that you talked to some people at Hotmail?
because you know what is going on without thinking, because that "eye candry" is a useful part of the UI...
Um.
"You are now scrolling."
Gee, thanks.