Flash mobs are a threat to security as it exists. When you (and I too) think about security as dealing with static, predictable events. Indeed, when we talk about risk management at minimizing risk, we think about how we can predict events.
But the unpredictable and the formerly low-percentage events are beginning to be...not so low-percentage. Exhibit A: the slashdot effect. Exhibit B: DDoS. Exhibit C: flash mobs.
How can we deal with such fluid events? One idea is that there can only be so much total "disruption" in a whole system. So if people could share security just as they share risk in an insurance plan, larger attacks could successfully be fended off. I think we see a lot of that in current techniques in dealing with server uptime, etc.
The serious question is: how good is the quality of information in the typical wikipedia article?
How good? In many cases better than proprietary solutions. Some cases not as good. You're right, many many Wikipedia articles suck. I've seen many of them through stupid google searches.
Yet I think we all share some kind of open-source optimism that it's getting there.
But how much do we REALLY need to know about Klingon or memetics?
I'd venture to guess that there's more interest in Star Trek and meme theory in slashdot than your typical "important" things like how the king of england gets coronated. don't you think encyclopedias should have information on what you want to know?
the great part, of course, is that both slashdot geeks and history geeks profit because wikipedia covers so much--in sufficient detail.
It's fairly easy toinsert misleading and false information into Wiki. Don't use it like as a replacement for an encyclopedia, or a properly vetted secondary source, unless you're an idiot.
It's also fairly easy to make corrections or readjust bias in wikipedia. I quote:
"Recent research by a team from IBM found that most vandalism suffered by Wikipedia had been repaired within five minutes. That's fast: 'We were surprised at how often we found vandalism, and then surprised again at how fast it was fixed,' says Martin]Wattenberg, a researcher in the IBM TJ Watson Research Center, in Cambridge, Mass." [7]
Is Wikipedia an encyclopedia? Let's go math-class style with the definition:
A comprehensive reference work containing articles on a wide range of subjects or on numerous aspects of a particular field, usually arranged alphabetically.
before you ask how wikipedia will *ever* work by allowing everybody to write, take a look at this quote:
According to a Wall Street Journal article from February 2004, researchers have found that there are frequent instances of vandalism at Wikipedia, but that these are often quickly resolved:
"Recent research by a team from IBM found that most vandalism suffered by Wikipedia had been repaired within five minutes. That's fast: 'We were surprised at how often we found vandalism, and then surprised again at how fast it was fixed,' says Martin]Wattenberg, a researcher in the IBM TJ Watson Research Center, in Cambridge, Mass." [7]
Selling shares on the public market is NOT the only way to cash out. A large number of companies--internet notwithstanding--are acquired privately. In fact it might make sense as a private acquisition for a company that would be able to leverage craigslist's user-base profitably, perhaps more so than others. Thus craigslist would be *particularly* valuable to them, making them willing to pay more than perhaps the going market price--because often companies are afraid that their competitors will get there first if they don't.
right, because real candidates raise enough campaign money to advertise, which makes them real candidates.
maybe one day we will vote for candidates based upon their qualifications, not how famous they are. Arnold comes to mind. and these will be the real candidates you speak of.
the perfect targeted advertising is already happening. on the berkeley campus kids walk around with their pearly white colored headphones and people stare dreamily. not exactly the richest campus, either, but the gears are certainly turning.
often advertisers don't necessarily lie but rather manipulate through sub-conscious phrases, incomplete truths, and meaningless words. think of many famous advertising campaigns (soda comes to mind) and it's just kind of empty propaganda.
there's a website called adbusters which is very interesting.
Here's an idea. Don't charge per click but per sale generated. The advertizer is happy, because he gets what he pays for. Google is happy, because the customer pays for what they get. There wouldn't be any idea in boosting up the click rate, and fraud would be virtually impossible.
good luck auditing that. great idea in principle, but not very doable on a large scale right now.
on the other hand, there is "ebates" and "fatcash" that do something similar to what you suggested. plus they give you x% back (the customer)
Ironically, Colin, I e-mailed you about your entry a few months ago. I'm not sure you took my response well. E-mailed you again, which should refresh your memory.
I've made at least a dozen real-world friends that I regularly go out with, and several more that are online only at this stage, and that's only after three months' membership.
Interesting. A lot better than my experience with Friendster, Myspace, and Thefacebook. I'd love to hear more from you on this point.
I use FeedDemon from Nick Bradbury, the small developer behind the original Homesite. I never bought Homesite a while back so I was happy to support FeedDemon. This guy puts out superior software without bloat and with excellent UI. I think it's worth supporting developers like that.
Slashdot Mirror
Flash mobs are a threat to security as it exists. When you (and I too) think about security as dealing with static, predictable events. Indeed, when we talk about risk management at minimizing risk, we think about how we can predict events.
But the unpredictable and the formerly low-percentage events are beginning to be...not so low-percentage. Exhibit A: the slashdot effect. Exhibit B: DDoS. Exhibit C: flash mobs.
How can we deal with such fluid events? One idea is that there can only be so much total "disruption" in a whole system. So if people could share security just as they share risk in an insurance plan, larger attacks could successfully be fended off. I think we see a lot of that in current techniques in dealing with server uptime, etc.
Hey dude. you still can in orange county:
The Irvine Ranch Land Reserve - more than 50,000 acres of permanently protected open space in the heart of Orange County, California offers a rich menu of recreational opportunities, from hiking and horseback riding to bird-watching and picnicking. The land is diverse; stretching 22 miles from the mountains to the sea; comprising more than half of the historic 93,000-acre Irvine Ranch and is home to hundreds of species of native plants and animals.
peace from newport beach ca
what are you talking about dude? his link doesn't even work.
How good? In many cases better than proprietary solutions. Some cases not as good. You're right, many many Wikipedia articles suck. I've seen many of them through stupid google searches.
Yet I think we all share some kind of open-source optimism that it's getting there.
they already exist on wikipedia.
want more? it takes a while to get to a million. maybe you can help...
But how much do we REALLY need to know about Klingon or memetics?
I'd venture to guess that there's more interest in Star Trek and meme theory in slashdot than your typical "important" things like how the king of england gets coronated. don't you think encyclopedias should have information on what you want to know?
the great part, of course, is that both slashdot geeks and history geeks profit because wikipedia covers so much--in sufficient detail.
It's also fairly easy to make corrections or readjust bias in wikipedia. I quote:
"Recent research by a team from IBM found that most vandalism suffered by Wikipedia had been repaired within five minutes. That's fast: 'We were surprised at how often we found vandalism, and then surprised again at how fast it was fixed,' says Martin]Wattenberg, a researcher in the IBM TJ Watson Research Center, in Cambridge, Mass." [7]
Is Wikipedia an encyclopedia? Let's go math-class style with the definition:
A comprehensive reference work containing articles on a wide range of subjects or on numerous aspects of a particular field, usually arranged alphabetically.
Apparently Wikipedia is an encyclopedia.
-source
Congrats to Wikipedia for the 1 millionth entry...and (less easily measured) even more interesting, deep, and thoughtful articles.
Selling shares on the public market is NOT the only way to cash out. A large number of companies--internet notwithstanding--are acquired privately. In fact it might make sense as a private acquisition for a company that would be able to leverage craigslist's user-base profitably, perhaps more so than others. Thus craigslist would be *particularly* valuable to them, making them willing to pay more than perhaps the going market price--because often companies are afraid that their competitors will get there first if they don't.
Actually I think the most impressive leaps have already been taken in a previous age.
check out this graph of the population of the world. the sudden explosion during the Industrial Revolution is staggering, dwarfing even the exponential increases in computer power.
graph 2
graph 3
you're right. but then doing almost anything civilized in life would require a small research project.
for example, take slashdot's TOS.
Fame = Votes.
Welcome to The USA.
Democracy and populism isn't all fun and games, huh?
peoples' minds are getting lazy..
right, because real candidates raise enough campaign money to advertise, which makes them real candidates.
maybe one day we will vote for candidates based upon their qualifications, not how famous they are. Arnold comes to mind. and these will be the real candidates you speak of.
This has happened before. See how Ken Ribet's discoveries led eventually to a crucial part of Andrew Wiles's proof of Fermat's Last Theorem.
also, it's harder to argue on DNA research than, say, about Bush v. Kerry :)
the perfect targeted advertising is already happening. on the berkeley campus kids walk around with their pearly white colored headphones and people stare dreamily. not exactly the richest campus, either, but the gears are certainly turning.
often advertisers don't necessarily lie but rather manipulate through sub-conscious phrases, incomplete truths, and meaningless words. think of many famous advertising campaigns (soda comes to mind) and it's just kind of empty propaganda.
there's a website called adbusters which is very interesting.
Here's an idea. Don't charge per click but per sale generated. The advertizer is happy, because he gets what he pays for. Google is happy, because the customer pays for what they get. There wouldn't be any idea in boosting up the click rate, and fraud would be virtually impossible.
good luck auditing that. great idea in principle, but not very doable on a large scale right now.
on the other hand, there is "ebates" and "fatcash" that do something similar to what you suggested. plus they give you x% back (the customer)
several of the universities mentioned (university of southern california) are NOT public universities but rather private.
I do think any details will leak out, though. that's the nature of academia, to share information..
Ironically, Colin, I e-mailed you about your entry a few months ago. I'm not sure you took my response well. E-mailed you again, which should refresh your memory.
you must be new here :)
I've made at least a dozen real-world friends that I regularly go out with, and several more that are online only at this stage, and that's only after three months' membership.
Interesting. A lot better than my experience with Friendster, Myspace, and Thefacebook. I'd love to hear more from you on this point.
looking at the technical reports from 2002, mpg and cost numbers look really good. where are the more recent ones, though?
you like juxtaposing those words. can you explain further?
I use FeedDemon from Nick Bradbury, the small developer behind the original Homesite. I never bought Homesite a while back so I was happy to support FeedDemon. This guy puts out superior software without bloat and with excellent UI. I think it's worth supporting developers like that.