What I think they really need is a "Web of Credibility" in the same form as PGP and Thawte have a "Web of Trust". People on the net could rate your comments/diatribes/rants on how factual they seem to be. How each person's rating of you figures into your overall credibility score would depend upon their credibility score. So, some 1337-dUdE who just flames everyone can go ahead and claim that you don't know what you're talking about, but it'll hardly affect your score. Meanwhile, if Stephen Hawking rates you as highly-credible, that would count much more.
Also, knowing that everything you put out there on the net, potentially, can affect your score (or "go onto your permanent record", as our high-school teachers used to threaten), it could make people a little more careful
We already have (or are on the cusp of having) all of the pieces necessary for this. Slashdot is only one of many sites that have things like karma, so we're already used to the paradigm of critiquing other people's comments. We also need a way of having universal identities so that people can't just create new ones after they burn one up by making asinine comments everywhere. Things like OpenID provide an entry into this... but we probably, ultimately, need something like Thawte's web-of-trust personal certificates.
Finally, there has to be a perceived need. I've been waiting over a decade for a web-of-credibility. Now, maybe this latest Wikipedia thing will provide the impetus to get it rolling. They're probably big enough to make it actually catch on.
The new technology includes reusable paper which can be printed and erased dozens of times...
Coming soon to your local office-supply store... 8.5x11 and A4 paper will now be labeled "8.5x11-R" and "A4-R" in preparation for the arrival of "8.5x11-RW" and "A4-RW". In a related story, Sony announced today their competing "8.5x11+RW" and "A4+RW" formats.
So if the article summary is correct does it even matter if the consumer desktop pc has SMART enabled or not?
Well, I was a little disappointed by the article. They looked at a lot of different SMART categories and they looked at the different ages of the drives, but they didn't delve into the different types of failures. I get about 1 "I think my drive crashed and I was hoping you could recover it" call per month and I see a variety of failure types. Probably the most common ones I see now are ones where something has gone wrong with the control circuits/mechanism and not the media itself. For example, something can go wrong with the motor that spins the platters, or you can seize the bearings for the head traversal, etc. I've even seen some where a chip on the controller board literally popped when it got too hot. These aren't going to be detected by SMART... I don't know what would predict failures like that.
The article states that, in about half of the failures, there were no SMART warnings at all. Okay, but what was the breakdown in the kinds of failures of these unpredicted ones? If they were all spindle motor and head traversal failures, then you can't blame SMART for that. If it turns out that SMART gave warnings for 95% of all failures that were media-degradation related (like bad sectors, etc... where the drive still talks to your machine properly, and just can't get the data you want), then I'd say SMART is pretty darn useful.
But, alas, I didn't see any breakdown for failure type....
Dan Gillmor has a piece up on his Center for Citizen Media blog about the coming decline in the venerable professions of photojournalism and videography...
But the good news is, there's expected to be a huge increase in the professions in the editing room, including: image stabilization, contrast color adjustment, sound retouching (to remove all of the "Holy crap, dude, check out this plane that's on fire. Swwweeeeeeet!!!! I'm getting it all with my camera!").
Part of why the pros are pros is not because they have the fancy equipment... it's also because the know how to zoom without making the viewer puke, how to get angles with good lighting, and how to move the camera without jerking the camera too much. All that goes out the window when you've got some drunk dude with a RAZR capturing a street brawl in a dark alley.
Maybe it would be possible to OCR every image as it comes through but then you'll just have spammers sending you CAPTCHA'd messages
As another poster pointed out, the FuzzyOCR plugin for SpamAssassin already does the OCR stuff, and that's just the free solution. I'm sure that many proprietary anti-spam vendors have already rolled something like this out, too.
What I totally didn't expect (at least not as quickly) was the "response" from the spammers of making the images captcha's. Within about a month, more than 75% of my image spam is captcha'd and sprinkled with confetti.
Now, what actually brings a smile to my face, however, is this: Several of the captcha images they've sent me are almost indecipherable by *me*. In order to make it past the OCR engine, the spammers seem to be avoiding "normal" fonts like Arial, Times, Courier, etc. and opting solely for those novelty fonts that look like the letters have snow on them, or that the letters are melting/bleeding. Combine that with random choices for the font and background colors and you often get an image where you *really* have to look hard to see what stock they're trying to get you to buy.
In other words, in an effort to be illegible to the spam filters, they're approaching making it illegible to their target audience. Nice!
I still say that we'll never get the upper-hand until we start crypo-signing all outgoing mail (preferably at the user level, but even doing it at the SMTP server level would be a big step) and start being more critical of any non-signed incoming mail.
Regarding the image spam that's on the rise, some spam filters are actually using OCR to turn the images into text and then scan them. There's a plugin for SpamAssassin called FuzzyOCR which does this. I'm testing it out and it actually succeeds on about half of the image spams I get (the other times, it crashes due to bugs in the various image converters that it relies upon).
It does jack the server loads up, as you'd expect. Fortunately, one of the features that it uses is that it keeps a hash value (and the spam score it got) for all of the images that it OCR's, so it only has to do each image once.
It is pretty surprising to see it work. With FuzzyOCR turned off, my test messages get scores of 2 or 3. With it turned on, the scores jump up to 20-30.
If you can't add signal, you can still improve the signal-to-noise ratio by not adding noise.
I voted today, but only for about 15% of the positions/initiatives on the ballot. The ones I didn't vote on fell into two categories:
Issues where I didn't know enough about the candidates/initiative. In this case, I see it as pointless to just add coin-flip or party-line votes. I'll make it easier for the hand-recounters (if a recount is needed) by giving them one less thing to count.
Issues where I knew the issue, but where the outcome really didn't affect me. In this case, casting a vote would dilute the votes of people who have a much larger stake in the issue. In a perfect world, maybe we could vote for someone/something along with a weight of 1-10 for how strongly we favored our choice. But, today, the weighting can only be binary (0 or 1), so I weighted most of my votes as 0 this time by not voting.
Besides... it's not like it matters anyway. I think of the polling place as more of a hands-on "democracy museum", where you get to experience what it was like when the people got to select their leaders.
There's been so much focus on how awful the Diebold machines are, do you think there's the danger that many people will feel that "If we just ditch the Diebold machines, everything's okay", when Diebold is actually just the worse of a large group of bad e-voting providers?
Although it doesn't pertain so much to electronic voting machines as it does to allowing individual voters to play more of a role in verifying the integrity of the process, could you share your thoughts about PunchScan?
Even though the reports (opening voting machines with mini-bar keys, untraceable viruses that steal votes, etc.) keep getting more and more jaw-dropping, they still haven't caused any significant outrage on the part of the electorate. I tell my friends that the only thing that will do it will be for several nerds hack some voting machines to hand the election to some obscure candidate... by a margin that exceeds the number of registered voters in that district. Until then, I fear that people will just keep assuming that, as long as they're allowed to press a button next to their candidate, that their vote is assured to make it into that big tally they see on CNN that night. It's almost as though they think that all of these reports are just academic or theoretical until it is demonstrated that a *real* election can *really* be stolen.
Can you think of any other way to snap the electorate out of their apathy?
It would be interesting if all email server admins suddenly opened the flood gates for a day or two. Maybe then the general population will gain a better appreciate of the scale of the matter.
Which is why I'm surprised Spamhaus doesn't just "simulate" what life would be like without them... before we're without them. Dispense with the predictions of how much spam will increase and what fate will befall the servers. Just shut off your service for a bit and wait for everyone to offer you their firstborn. Enron did it with California's electricity and it worked like a charm.
Yeah, it's impressive on some small level I guess, but what is the point?
Exactly. The point is to do something that you can hold the world-record for. Personally, I think it would have been easier to recite the digits of "pi + 1". The record probably isn't for as many digits on that one. Or "2 * pi"... or "pi * pi". Heck! They're going to have to have a whole *chapter* set aside for just *my* records.
Blocking a phishing Web site earned you twice as many points as just warning about it in this test
This reminds me of when the "Quarterback Rating" came out back in the 80's. Back then, there were people arguing that Joe Montana was the greatest QB in history. Around that time, a "Quarterback Rating" scheme emerged with some esoteric weighting of various performance stats (completion percentage, TD's per game, etc. etc). Although nobody seemed to understand the rationale for the particular weighting... lo and behold, Joe Montana came out with the highest rating of current QB's and, IIRC, all QB's in history.
The lesson: Beware of any "combined" or "aggregate" score of competing products when the person doing the aggregating: A) has an interest in one product doing better than the others, and B) knows, beforehand, what the strengths of that product are.
Actually (IIRC) it's that the story is wrong. It's not a database of the budget but a database of "earmarks".
Actually, I believe there are *two* different (yet related) things going on. One is the budget database (mentioned in the article), and the other is a change in the House rules eliminating the ability of reps to insert earmarks *anonymously*. The database is a law and can be expected to persist. The rules change on the other hand, although it holds much more promise to curtail budget abuse, is only for this year! So, in light of that, I'd say that the rules change is a purely political move... "Hey, how about we Republicans show the people that we're the party that stands for open-ness and disclosure... at least until it gets us through the mid-term elections!".
The public doesn't get it....
on
Brave New Ballot
·
· Score: 2, Interesting
I recently had a discussion with a friend about this whole Diebold mess, and how the public just has no clue about the vulnerabilities. I'm beginning to feel that an election version of "9/11" is the only way to do it. Have a bunch of operatives volunteer to work an election in some district and then, on election day, have the machines report twice as many votes as there are registered voters... and have them all be for the Green party or something. Until then, the sheeple will think that it's all ivory-tower theoretical stuff. They need to be shown that a real election can really be stolen.
Of course, this will mean jail time for the perps, but I have two things to say about that. 1) Some would consider that a small price to pay for preserving democracy. 2) You might be able to get a light sentence if you mailed, before the election, some letters (saying "Do not open until after the election" on them) to several news agencies declaring that you are going to rig the election in order to expose the dangers to our democracy.
I wonder if the programmers wrote in some cute "back doors" into the sentencing software... where, you could kill someone and, provided it was through bludgeoning, with a brown rock, on a Tuesday, with your left hand.... you only end up serving a week in prison.
I know people that i think are absolutely credible sources about technology subjects, but who I wouldn't consider credible in discussing say politics (although i'm sure they feel they were qualified).
Well then, if they often expounded about technology topics *and* political topics, then other people would be giving them a blend of high and low credibility scores... yielding a score of "so so". However, if they knew to keep their trap shut when the topic turned to politics, then they wouldn't get as many low scores and their resulting score would be higher.
So, it's not a score about how much you know... it would be a score about if you are *aware* of how much you know *and* if you are conscientious enough to stay quiet when you realize that there are more-knowledgeable folk in the discussion. For the *consumer* of this data (ie, somebody viewing your credibility score) it all boils down to "How likely is it that the information they're giving me is accurate?". If you spout off about stuff you have no clue about, you might not be aware that you haven't any clue... or... you might be perfectly aware that you have no clue and just not care. That doesn't matter to the consumer of your information. All that matters is the probability that the information you offer is accurate.
Sure, it would be nice if we had credibility scores for the millions of topics out there... but there are millions of them. What I think we'd have to settle for is a measure of how well each person "self-governs" the information they offer. And the information they offer doesn't necessarily need to be accurate, so long as they don't *portray* or qualify it as such. For example, how often do you seen Slashdotters start with "IANAL, but...."? So... maybe it would be a measure of how accurately *I* represent the accuracy of what I'm saying.
So, in that sense, it doesn't matter if I am an expert in that subject... as long as I realize and admit that I'm not.
About 10 years ago, when I learned of PGP's "Web of Trust" system (where I could choose to trust everybody that you trust), I turned to a colleague and said "What we need next is a 'Web of Credibility'..." where individuals in the community can bestow credibility points to others... and the points I can bestow to you would depend upon how credible the community thinks *I* am, and so on. In other words, if Noam Chomsky or Lester Thurow vouch that you're highly-credible, then that would boost your credibility more than a few glowing scores from your cable guy and the kid at the local sip-n-go. Ultimately, it would be a measure of how likely (or unlikely) you were to spout off on something that you had no clue about.
Having not yet RTFA, I'd just like to say that I agree, wholeheartedly, with the general notion... and I look forward to the day when our credibilities are incoporated into our digital signatures (that I hope we're also all using someday).
- Joe
And if the price has risen recently, sell it short. It's gonna go back down to its true value eventually (probably after the spammers sell their shares).
I actually looked into this earlier this year... figuring that the only *safe* bet in this world is to bet on most Americans being stupid.
As I got each stock-pump spam, I'd record the ticker symbol and the date I first saw a spam for it. Then, I'd look at the stock's price after 1 week, 2 week, 4, 8, 12, and 16 weeks out. I was surprised at how consistent the results were. They'd be down to about 80% after a week or two, and then down to about 40%-60% of their value by weeks 12-16.
So, the prescribed course of action for a pattern like this is to "short" the stock... or buy "put" options on it. But, there are two problems:
First, the volume on these stocks is fairly low... so low that, in order to make more than a couple hundred bucks, the volume *you* add is going to be on par with the (otherwise) "normal" volume. At that point, you're going to start affecting the price with your own actions.
Second, these stocks are all so small that I wasn't able to find any options trading for them whatsoever.
Back in college, I toyed with the idea of doing something like this as my senior project. My name for it, at the time, was a "water hologram", which I think more-accurately describes what's going on. You're manipulating one or more wave sources in order to generate a specific pattern of constructive and destructive interference.
I'd like to see a video of it. I'm curious to see if the images are: A) constantly oscillating up and down, B) perpetually raised and not oscillating, or C) just there for a moment and then gone (like a rogue wave or something).
I see some problems with this particular approach. I haven't RTFA, but what if the natinoal popular vote is just barely in favor of candidate A, and everyone's ready to give all of the EC votes to them... and then, whoops!, Rhode Island decides to do a recount that drags on a long time like the Florida mess in 2000? It seems as though it would be like 50 people trying to rent a movie... "Okay... once again... are we certain that this is what we all want to watch?".
Another approach that is being tried as we speak is one where the individual states allocate their electoral votes proportionally based on their own statewide popular votes. In fact, I think a couple of states already do this. This eliminates the need for states to wait for each other. The catch is that they have to propose legislation in each state to get that state to change the way it awards EC votes. The only problem I have with this particular method is that they're trying to roll it out in some of the larger states, first; for example, California. California has been pretty solidly in the Democrat camp for about two decades, now, and it also carries the largest number of EC votes. If California switched to proportional allocation, and the Republicans got 1/3 or so of those EC votes, that's a big swing. A Democratic opponent would have to win a 3 or 4 smaller states just to make up that loss even though California's majority still favored him/her. We're talking just about the loss of EC votes from going from "all or nothing" to proportional. So, I think the way you have to roll this out is by introducing in the states with the fewest EC votes first, where you'd only see a swing of 1 or 2 EC votes, and then gradually introduce it to larger and larger states. Also, I figure that you'd want to introduce it to a Republican state and a Democratic state at the same time... with some clause where they both have to pass for either to go into effect, lest just one pass and that state's party gets shafted.
Although this isn't likely to change the actual outcome of many elections, it is far more likely to change the amount of spending in your state: both campaign spending and in "pork" (the allocation of gov't spending aimed at garnering favour with that area's voters). If you're living in a "battleground" state right now, you get bombed with campaign ads when election time comes around... yet you also see more than your fair share of highway bills, homeland-security funding, free gov't cheese, and whatnot. This would probably subside if your state had proportional EC votes. On the other side of the coin, states that are "locked up" solidly in one party's favor typically receive little campaign spending and pork. Back in 2004, my friend told me about visiting Nevada (a battleground state)... where it was non-stop campaign ads during the commercial breaks on TV. As soon as he drove over the border into California, he could hardly tell there was an election coming up.
Lastly, I remember reading an article a few years ago where a mathematician was able to show that you actually have more voting power with the electoral college than you would without it. Your vote only really matters if your candidate won, yet wouldn't have had you not voted (or voted for the opponent). With a popular election, this would require that the candidate win the national popular election by a single vote. This is, to understate things, "unlikely". To get this same scenario with an electoral college, two things need to happen. First, the candidate has to win by a number of electoral votes that is less than double the number that your state has... and your district's or state's popular vote has to be decided by a single vote.
The odds of the second scenario happening are, also, highly unlikely. However, this mathematician was able to show that the second scenario was marginally more-likely. So, if you like the idea of your vote being the "deciding vote", then the EC would have a benefit... and this benefit would still be there even in a proportional-allocation system.
Determining the full effects of Net neutrality can be difficult, however, in part because the concept is hard to define precisely
I think it's hard to define precisely because people are using different definitions.
The articles you see in papers and news sites all seem to boil it down to companies having to pay more for the larger amount of bandwidth they use. But that can't be it, because that's what's been happening all along. End users pay more for a faster DSL connection. ISP's pay more for a fatter pipe. This is the way it has always been... so to say that this is in danger of happening doesn't make any sense.
On the other end of the spectrum is the idea of charging based upon the nature of the content. VoIP, for example, being billed at a higher rate than, say, Usenet or web surfing. This is akin to the phone company charging you (or somebody) more if you use your telephone to dial the police or hospital than they do if you dial your mom. In fact, what might be more accurate is if the phone company charged less when you were talking to your mom about her meatloaf recipie... and then charged more when the nature of the conversation turned to "... make sure that you remember to give dad his heart medication!". That would be billing based upon the nature of the content.
I've seen some argue that this would merely be capitalism at work. It's charging what the market will bear. Getting the "heart medication" part of the message through is more crucial to you than getting the "meatloaf recipie" part through... and the phone companies should be able to charge what it's worth to you.... not what it costs them, right? Well, all I can say to that is that there are situations similar to that which the American people have pretty much agreed are unfair. Look at profiteering, for example. When a hurricane hits some region and the stores start charging $20 per gallon of water, we've pretty much agreed that that's crossing the line... partly because the increase in price had nothing to do with an increase in costs. (yet also partly because the predicament that the buyers are in wasn't forseeable).
Plane tickets would be a counter-example, however. Airlines have all kinds of tricks to get more money out of the people who can/will pay more. Charging more if you don't stay over a weekend is their way of getting more money out of the business travelers (who are traveling to a weekday conference, having the company pay for the airfare, and don't want to spend the weekend away from their family). This is an incarnation of price-discrimination that we've come to accept.
Which of those you feel NN falls into is up to you... but I think we need to start by giving our lawmakers some more-accurate analogies.... because the "fatter pipe" thing is just way off.
So they're getting paid 1/5 of the median household income for the area, before their employer takes half of it for living expenses.
I have a few retorts to this:
1 - I spend easily half of my income on living expenses (housing and food)
2 - In the US, with $50k as the median, 1/2 of 1/5 of that would be $5k for housing and food per year. That's pretty cheap. Now, many of the replies to this post have taken exception to your assertion that median income for that region of China is $3k. They're claiming that it's about 1/2 to 1/3 of what you claim... which would mean these ladies are getting housed and fed for somewhere around $2500-$1700 per year. So, yes they're getting paid a low wage compared to what we're used to in the west, but their living expenses are similarly low. So, let's dispense with this despair over their employer taking 1/2 of their wages for housing and food.
Now... as far as the low wages they're being paid, I agree, this is an outrage. We must switch immediately to having iPods assembled by robots so that these ladies are back out onto the street prostituting themselves like they should be.
For those of you having trouble parsing the sarcasm into the real message, what I'm getting at is that there are a lot of uneducated, unskilled people in this world. Their value to this planet extends only so far as they're able to put washers onto bolts and laces onto shoes for just slightly cheaper than it would cost for a machine to do it. If robots were just a little cheaper or just a little smarter, these ladies would be without housing and without food and we'd consider the situation to be more representative of our advanced world because we'd get to think that our iPods are assembled by sophisticated robots and we wouldn't make the connection between our iPods and the starving women... even though the scenario is probably more deplorable than having the women building the iPods.
The tragedy isn't that Apple has a way for these unskilled women to make themselves useful. The tragedy is that those women, and countless others in this world, have such a limited usefulness to start with.
The guy needs to realize that nothing he says is going make these folks magically wake up and say "OMFG, I was so wrong. Here's your phone back, sorry for the trouble. Peace be with you."
Exactly. When you're in a dispute, you have to consider the sophistication of your opponent. Getting into a pissing match with people from "da 'hood" isn't going to get you anywhere because pissing is their stock in trade. Check their myspace pages. The backgrounds are all pimped cars, gold chains, guns, and all that "gangsta" puffery. You might get them to "take shelter" (ie, changing their myspace page to ask to be left alone) for a little while, but you're *never* *ever* going to get them to go "I just can't take your relentless cyber-taunting a minute longer. Please, come take this cursed device away from me and never let it darken my doorstep again!!!". You must realize that, the biggest humiliation of all to them would be if they CAVED IN, so trying to humiliate them into doing just that is pointless.
What this guy really needed to do is give these punks no indication that he knew their identity until a nice officer knocked on their door.
So, why one over the other, structural concerns aside?
I think you're missing the real flaw, here. It has almost nothing to do with the support beams. It's about the way the support rods were implemented. The original design was for *single* rods to descend all the way from the ceiling, *through* the upper catwalk, and continue on to the lower one. This means that the upper catwalk beams have to support *one* catwalk's weight, since the lower catwalk's weight was transferred right up through the rod to the ceiling.
The flawed change is that they decided to use pairs of *two* rods of roughly half the length of the long ones. One rod went from the ceiling to the upper catwalk, and another went from the upper catwalk to the lower one. This meant that the lower catwalk was hanging from the *upper catwalk*, not the ceiling. If you take a moment to picture this, you'll see that this means that there is now the weight of *two* catwalks on the junction where the upper rod connects to the upper catwalk. That's where the failure occurred.
From the Wikipedia article:
The Havens Steel Company, responsible for manufacturing the rods, objected to the original plan of Jack D. Gillum and Associates, since it required the whole of the rod below the fourth floor to be threaded in order to screw on the nuts to hold the fourth floor walkway in place. These threads would probably have been damaged beyond use when the structure for the fourth floor was hoisted into position. Havens therefore proposed an alternate plan in which two separate sets of tie rods would be used; one connecting the fourth floor walkway to the ceiling, and the other connecting the second floor walkway to the fourth floor walkway. This also created a bending moment, or torque, on the connection.
As an aside, the History Channel has two great series that have dealt with all of the ones mentioned in the original post. One is "Engineering Disasters" (they're up to about 19 episodes so far) and "Inviting Disaster". They've covered Three Mile Island, Chernobyl, the Sodium Reactor Experiement in L.A., The San Fransiquito dam failure, the Baldwin Hills dam, the "mollasses flood"., the deHaviland Comet, the 737 rudder actuators, the DC-10, radio tower collapses, stadium roof collapses, the Chicago maintenance tunnel flood, and... my personal favorite, the Lake Peigneur disaster, where oil drillers in a huge lake acccidentally drilled into a huge salt mine. The entire lake drained into the mine over a day or so.... complete with a big whirlpool, just like in your bathtub!
Slashdot Mirror
What I think they really need is a "Web of Credibility" in the same form as PGP and Thawte have a "Web of Trust". People on the net could rate your comments/diatribes/rants on how factual they seem to be. How each person's rating of you figures into your overall credibility score would depend upon their credibility score. So, some 1337-dUdE who just flames everyone can go ahead and claim that you don't know what you're talking about, but it'll hardly affect your score. Meanwhile, if Stephen Hawking rates you as highly-credible, that would count much more.
Also, knowing that everything you put out there on the net, potentially, can affect your score (or "go onto your permanent record", as our high-school teachers used to threaten), it could make people a little more careful
We already have (or are on the cusp of having) all of the pieces necessary for this. Slashdot is only one of many sites that have things like karma, so we're already used to the paradigm of critiquing other people's comments. We also need a way of having universal identities so that people can't just create new ones after they burn one up by making asinine comments everywhere. Things like OpenID provide an entry into this... but we probably, ultimately, need something like Thawte's web-of-trust personal certificates.
Finally, there has to be a perceived need. I've been waiting over a decade for a web-of-credibility. Now, maybe this latest Wikipedia thing will provide the impetus to get it rolling. They're probably big enough to make it actually catch on.
The article states that, in about half of the failures, there were no SMART warnings at all. Okay, but what was the breakdown in the kinds of failures of these unpredicted ones? If they were all spindle motor and head traversal failures, then you can't blame SMART for that. If it turns out that SMART gave warnings for 95% of all failures that were media-degradation related (like bad sectors, etc... where the drive still talks to your machine properly, and just can't get the data you want), then I'd say SMART is pretty darn useful.
But, alas, I didn't see any breakdown for failure type....
Part of why the pros are pros is not because they have the fancy equipment... it's also because the know how to zoom without making the viewer puke, how to get angles with good lighting, and how to move the camera without jerking the camera too much. All that goes out the window when you've got some drunk dude with a RAZR capturing a street brawl in a dark alley.
What I totally didn't expect (at least not as quickly) was the "response" from the spammers of making the images captcha's. Within about a month, more than 75% of my image spam is captcha'd and sprinkled with confetti.
Now, what actually brings a smile to my face, however, is this: Several of the captcha images they've sent me are almost indecipherable by *me*. In order to make it past the OCR engine, the spammers seem to be avoiding "normal" fonts like Arial, Times, Courier, etc. and opting solely for those novelty fonts that look like the letters have snow on them, or that the letters are melting/bleeding. Combine that with random choices for the font and background colors and you often get an image where you *really* have to look hard to see what stock they're trying to get you to buy.
In other words, in an effort to be illegible to the spam filters, they're approaching making it illegible to their target audience. Nice!
I still say that we'll never get the upper-hand until we start crypo-signing all outgoing mail (preferably at the user level, but even doing it at the SMTP server level would be a big step) and start being more critical of any non-signed incoming mail.
Regarding the image spam that's on the rise, some spam filters are actually using OCR to turn the images into text and then scan them. There's a plugin for SpamAssassin called FuzzyOCR which does this. I'm testing it out and it actually succeeds on about half of the image spams I get (the other times, it crashes due to bugs in the various image converters that it relies upon).
It does jack the server loads up, as you'd expect. Fortunately, one of the features that it uses is that it keeps a hash value (and the spam score it got) for all of the images that it OCR's, so it only has to do each image once.
It is pretty surprising to see it work. With FuzzyOCR turned off, my test messages get scores of 2 or 3. With it turned on, the scores jump up to 20-30.
I voted today, but only for about 15% of the positions/initiatives on the ballot. The ones I didn't vote on fell into two categories:
- Issues where I didn't know enough about the candidates/initiative. In this case, I see it as pointless to just add coin-flip or party-line votes. I'll make it easier for the hand-recounters (if a recount is needed) by giving them one less thing to count.
- Issues where I knew the issue, but where the outcome really didn't affect me. In this case, casting a vote would dilute the votes of people who have a much larger stake in the issue. In a perfect world, maybe we could vote for someone/something along with a weight of 1-10 for how strongly we favored our choice. But, today, the weighting can only be binary (0 or 1), so I weighted most of my votes as 0 this time by not voting.
Besides... it's not like it matters anyway. I think of the polling place as more of a hands-on "democracy museum", where you get to experience what it was like when the people got to select their leaders.There's been so much focus on how awful the Diebold machines are, do you think there's the danger that many people will feel that "If we just ditch the Diebold machines, everything's okay", when Diebold is actually just the worse of a large group of bad e-voting providers?
Although it doesn't pertain so much to electronic voting machines as it does to allowing individual voters to play more of a role in verifying the integrity of the process, could you share your thoughts about PunchScan?
Even though the reports (opening voting machines with mini-bar keys, untraceable viruses that steal votes, etc.) keep getting more and more jaw-dropping, they still haven't caused any significant outrage on the part of the electorate. I tell my friends that the only thing that will do it will be for several nerds hack some voting machines to hand the election to some obscure candidate... by a margin that exceeds the number of registered voters in that district. Until then, I fear that people will just keep assuming that, as long as they're allowed to press a button next to their candidate, that their vote is assured to make it into that big tally they see on CNN that night. It's almost as though they think that all of these reports are just academic or theoretical until it is demonstrated that a *real* election can *really* be stolen.
Can you think of any other way to snap the electorate out of their apathy?
The lesson: Beware of any "combined" or "aggregate" score of competing products when the person doing the aggregating: A) has an interest in one product doing better than the others, and B) knows, beforehand, what the strengths of that product are.
I recently had a discussion with a friend about this whole Diebold mess, and how the public just has no clue about the vulnerabilities. I'm beginning to feel that an election version of "9/11" is the only way to do it. Have a bunch of operatives volunteer to work an election in some district and then, on election day, have the machines report twice as many votes as there are registered voters... and have them all be for the Green party or something. Until then, the sheeple will think that it's all ivory-tower theoretical stuff. They need to be shown that a real election can really be stolen.
Of course, this will mean jail time for the perps, but I have two things to say about that. 1) Some would consider that a small price to pay for preserving democracy. 2) You might be able to get a light sentence if you mailed, before the election, some letters (saying "Do not open until after the election" on them) to several news agencies declaring that you are going to rig the election in order to expose the dangers to our democracy.
I wonder if the programmers wrote in some cute "back doors" into the sentencing software... where, you could kill someone and, provided it was through bludgeoning, with a brown rock, on a Tuesday, with your left hand.... you only end up serving a week in prison.
So, it's not a score about how much you know... it would be a score about if you are *aware* of how much you know *and* if you are conscientious enough to stay quiet when you realize that there are more-knowledgeable folk in the discussion. For the *consumer* of this data (ie, somebody viewing your credibility score) it all boils down to "How likely is it that the information they're giving me is accurate?". If you spout off about stuff you have no clue about, you might not be aware that you haven't any clue... or... you might be perfectly aware that you have no clue and just not care. That doesn't matter to the consumer of your information. All that matters is the probability that the information you offer is accurate.
Sure, it would be nice if we had credibility scores for the millions of topics out there... but there are millions of them. What I think we'd have to settle for is a measure of how well each person "self-governs" the information they offer. And the information they offer doesn't necessarily need to be accurate, so long as they don't *portray* or qualify it as such. For example, how often do you seen Slashdotters start with "IANAL, but...."? So... maybe it would be a measure of how accurately *I* represent the accuracy of what I'm saying.
So, in that sense, it doesn't matter if I am an expert in that subject... as long as I realize and admit that I'm not.
About 10 years ago, when I learned of PGP's "Web of Trust" system (where I could choose to trust everybody that you trust), I turned to a colleague and said "What we need next is a 'Web of Credibility'..." where individuals in the community can bestow credibility points to others... and the points I can bestow to you would depend upon how credible the community thinks *I* am, and so on. In other words, if Noam Chomsky or Lester Thurow vouch that you're highly-credible, then that would boost your credibility more than a few glowing scores from your cable guy and the kid at the local sip-n-go. Ultimately, it would be a measure of how likely (or unlikely) you were to spout off on something that you had no clue about.
Having not yet RTFA, I'd just like to say that I agree, wholeheartedly, with the general notion... and I look forward to the day when our credibilities are incoporated into our digital signatures (that I hope we're also all using someday). - Joe
As I got each stock-pump spam, I'd record the ticker symbol and the date I first saw a spam for it. Then, I'd look at the stock's price after 1 week, 2 week, 4, 8, 12, and 16 weeks out. I was surprised at how consistent the results were. They'd be down to about 80% after a week or two, and then down to about 40%-60% of their value by weeks 12-16.
So, the prescribed course of action for a pattern like this is to "short" the stock... or buy "put" options on it. But, there are two problems:
- First, the volume on these stocks is fairly low... so low that, in order to make more than a couple hundred bucks, the volume *you* add is going to be on par with the (otherwise) "normal" volume. At that point, you're going to start affecting the price with your own actions.
- Second, these stocks are all so small that I wasn't able to find any options trading for them whatsoever.
Looks like I've got to keep my day job.Back in college, I toyed with the idea of doing something like this as my senior project. My name for it, at the time, was a "water hologram", which I think more-accurately describes what's going on. You're manipulating one or more wave sources in order to generate a specific pattern of constructive and destructive interference.
I'd like to see a video of it. I'm curious to see if the images are: A) constantly oscillating up and down, B) perpetually raised and not oscillating, or C) just there for a moment and then gone (like a rogue wave or something).
I see some problems with this particular approach. I haven't RTFA, but what if the natinoal popular vote is just barely in favor of candidate A, and everyone's ready to give all of the EC votes to them... and then, whoops!, Rhode Island decides to do a recount that drags on a long time like the Florida mess in 2000? It seems as though it would be like 50 people trying to rent a movie... "Okay... once again... are we certain that this is what we all want to watch?".
Another approach that is being tried as we speak is one where the individual states allocate their electoral votes proportionally based on their own statewide popular votes. In fact, I think a couple of states already do this. This eliminates the need for states to wait for each other. The catch is that they have to propose legislation in each state to get that state to change the way it awards EC votes. The only problem I have with this particular method is that they're trying to roll it out in some of the larger states, first; for example, California. California has been pretty solidly in the Democrat camp for about two decades, now, and it also carries the largest number of EC votes. If California switched to proportional allocation, and the Republicans got 1/3 or so of those EC votes, that's a big swing. A Democratic opponent would have to win a 3 or 4 smaller states just to make up that loss even though California's majority still favored him/her. We're talking just about the loss of EC votes from going from "all or nothing" to proportional. So, I think the way you have to roll this out is by introducing in the states with the fewest EC votes first, where you'd only see a swing of 1 or 2 EC votes, and then gradually introduce it to larger and larger states. Also, I figure that you'd want to introduce it to a Republican state and a Democratic state at the same time... with some clause where they both have to pass for either to go into effect, lest just one pass and that state's party gets shafted.
Although this isn't likely to change the actual outcome of many elections, it is far more likely to change the amount of spending in your state: both campaign spending and in "pork" (the allocation of gov't spending aimed at garnering favour with that area's voters). If you're living in a "battleground" state right now, you get bombed with campaign ads when election time comes around... yet you also see more than your fair share of highway bills, homeland-security funding, free gov't cheese, and whatnot. This would probably subside if your state had proportional EC votes. On the other side of the coin, states that are "locked up" solidly in one party's favor typically receive little campaign spending and pork. Back in 2004, my friend told me about visiting Nevada (a battleground state)... where it was non-stop campaign ads during the commercial breaks on TV. As soon as he drove over the border into California, he could hardly tell there was an election coming up.
Lastly, I remember reading an article a few years ago where a mathematician was able to show that you actually have more voting power with the electoral college than you would without it. Your vote only really matters if your candidate won, yet wouldn't have had you not voted (or voted for the opponent). With a popular election, this would require that the candidate win the national popular election by a single vote. This is, to understate things, "unlikely". To get this same scenario with an electoral college, two things need to happen. First, the candidate has to win by a number of electoral votes that is less than double the number that your state has... and your district's or state's popular vote has to be decided by a single vote.
The odds of the second scenario happening are, also, highly unlikely. However, this mathematician was able to show that the second scenario was marginally more-likely. So, if you like the idea of your vote being the "deciding vote", then the EC would have a benefit... and this benefit would still be there even in a proportional-allocation system.
The articles you see in papers and news sites all seem to boil it down to companies having to pay more for the larger amount of bandwidth they use. But that can't be it, because that's what's been happening all along. End users pay more for a faster DSL connection. ISP's pay more for a fatter pipe. This is the way it has always been... so to say that this is in danger of happening doesn't make any sense.
On the other end of the spectrum is the idea of charging based upon the nature of the content. VoIP, for example, being billed at a higher rate than, say, Usenet or web surfing. This is akin to the phone company charging you (or somebody) more if you use your telephone to dial the police or hospital than they do if you dial your mom. In fact, what might be more accurate is if the phone company charged less when you were talking to your mom about her meatloaf recipie... and then charged more when the nature of the conversation turned to "... make sure that you remember to give dad his heart medication!". That would be billing based upon the nature of the content.
I've seen some argue that this would merely be capitalism at work. It's charging what the market will bear. Getting the "heart medication" part of the message through is more crucial to you than getting the "meatloaf recipie" part through... and the phone companies should be able to charge what it's worth to you.... not what it costs them, right? Well, all I can say to that is that there are situations similar to that which the American people have pretty much agreed are unfair. Look at profiteering, for example. When a hurricane hits some region and the stores start charging $20 per gallon of water, we've pretty much agreed that that's crossing the line... partly because the increase in price had nothing to do with an increase in costs. (yet also partly because the predicament that the buyers are in wasn't forseeable).
Plane tickets would be a counter-example, however. Airlines have all kinds of tricks to get more money out of the people who can/will pay more. Charging more if you don't stay over a weekend is their way of getting more money out of the business travelers (who are traveling to a weekday conference, having the company pay for the airfare, and don't want to spend the weekend away from their family). This is an incarnation of price-discrimination that we've come to accept.
Which of those you feel NN falls into is up to you... but I think we need to start by giving our lawmakers some more-accurate analogies.... because the "fatter pipe" thing is just way off.
1 - I spend easily half of my income on living expenses (housing and food)
2 - In the US, with $50k as the median, 1/2 of 1/5 of that would be $5k for housing and food per year. That's pretty cheap. Now, many of the replies to this post have taken exception to your assertion that median income for that region of China is $3k. They're claiming that it's about 1/2 to 1/3 of what you claim... which would mean these ladies are getting housed and fed for somewhere around $2500-$1700 per year. So, yes they're getting paid a low wage compared to what we're used to in the west, but their living expenses are similarly low. So, let's dispense with this despair over their employer taking 1/2 of their wages for housing and food.
Now... as far as the low wages they're being paid, I agree, this is an outrage. We must switch immediately to having iPods assembled by robots so that these ladies are back out onto the street prostituting themselves like they should be.
For those of you having trouble parsing the sarcasm into the real message, what I'm getting at is that there are a lot of uneducated, unskilled people in this world. Their value to this planet extends only so far as they're able to put washers onto bolts and laces onto shoes for just slightly cheaper than it would cost for a machine to do it. If robots were just a little cheaper or just a little smarter, these ladies would be without housing and without food and we'd consider the situation to be more representative of our advanced world because we'd get to think that our iPods are assembled by sophisticated robots and we wouldn't make the connection between our iPods and the starving women... even though the scenario is probably more deplorable than having the women building the iPods.
The tragedy isn't that Apple has a way for these unskilled women to make themselves useful. The tragedy is that those women, and countless others in this world, have such a limited usefulness to start with.
What this guy really needed to do is give these punks no indication that he knew their identity until a nice officer knocked on their door.
The flawed change is that they decided to use pairs of *two* rods of roughly half the length of the long ones. One rod went from the ceiling to the upper catwalk, and another went from the upper catwalk to the lower one. This meant that the lower catwalk was hanging from the *upper catwalk*, not the ceiling. If you take a moment to picture this, you'll see that this means that there is now the weight of *two* catwalks on the junction where the upper rod connects to the upper catwalk. That's where the failure occurred.
From the Wikipedia article: As an aside, the History Channel has two great series that have dealt with all of the ones mentioned in the original post. One is "Engineering Disasters" (they're up to about 19 episodes so far) and "Inviting Disaster". They've covered Three Mile Island, Chernobyl, the Sodium Reactor Experiement in L.A., The San Fransiquito dam failure, the Baldwin Hills dam, the "mollasses flood"., the deHaviland Comet, the 737 rudder actuators, the DC-10, radio tower collapses, stadium roof collapses, the Chicago maintenance tunnel flood, and... my personal favorite, the Lake Peigneur disaster, where oil drillers in a huge lake acccidentally drilled into a huge salt mine. The entire lake drained into the mine over a day or so.... complete with a big whirlpool, just like in your bathtub!