Ever received "bad information" from a trusted friend? Or "bad information" from a friend of a trusted friend?
Do you remember the worms that used to propagate by emailing themselves out to everyone in the victim's Outlook address book? They'd spread like wildfire because everyone would trust who the message was coming from. More innocent times I know, but I'm damn sure it could happen again; people aren't any smarter than before. But my real point is that webs of trust really do get penetrated, and the result is usually disastrous because hardly anyone in the real world is security-conscious. At least with the CA/PKI model, the real security knowledge is only needed by a small group of people who damn well should know that they need to know what's going on.
The problem is the same with Moody's, actually: the central issue is that the people doing the auditing are being paid by the people they're auditing. Simply having browser users pay CAs (or investors pay rating agencies) would put the economic incentives in the right place, but that idea doesn't sit well with a lot of people.
So instead, we're left with imperfect and leaky regulation. CAs really should be subject to more regular audits, and their trust bits should be removed by browser vendors when they are abused.
Your analysis looks spot on to me, and the key issue is that browser "vendors" aren't being strict enough on enforcement. They're ideally placed to do so as gatekeepers for normal users, and they're also strongly incentivized to act as such. But if nobody enforces the rules then a system that depends on enforcement is just not going to work. After all, the only rule that really needs enforcing is this: verify that the certificate is issued to someone who has a right to it. (There's another critical rule for subsidiary CAs: you must enforce the first rule.) Everything else can be handled through normal contracts and making sure that people check certificate validity correctly, which is a software thing that we (the technologists) ought to be able to get right.
Isn't that why they charge huge amounts for the certs?
When was getting a signature from a crappy CA expensive? Is $30/yr for a basic server certificate terribly expensive all of a sudden (and I bet you can go cheaper than that if you hunt around) or are you deliberately going to a CA that is both bad and costly too? That'd be a new level of dumb...
In other news, emacs 24 is to be renamed "egacs" because the previous snarky backonym of "Eight Megabytes and Continually Swapping" is now the average footprint of "Hello World".
That's a invidious lie! It stands for "Escape Meta Alt Control Shift", and that's still a valid criticism (and I admit I really like emacs); there's only one shift-like key on this keyboard that it doesn't use and that's only because it's a laptop and so needs to put some functionality off a special "Function" shift...
Why does solar thermal energy seem so little loved around here? It looks like a rather cheap and proven technology to me: gather a few mirrors, a bathtub of salt and a steam machine in any sunny area and there you go.
Sounds like you're ready to go. Since it's obviously so proven, just build one and start producing power! (It's good for some parts of the world - especially deserts and semi-deserts - but other parts favor other solutions. What a surprise!)
*But*, the waste they do produce is *highly* radioactive (which is why it decays so fast), and *still* needs to be stored for *generations*.
If the waste is solid, insoluble and "hot", it's essentially self-protecting.
Right now, the real problem is that too many people are uncertain as to what the aim of protection is; it seems that many are after some magic whereby nothing bad can happen to anyone no matter how determined/stupid. Unsurprisingly, no such thing exists and that throws Greens in a total tizzy. (Of course, they mostly also think that Mother Nature is all good and can never hurt them provided they believe enough and hug the trees. Or am I overstating it a little?)
It's not that the containers cause fires (though I would guess the welding torches fed from them do from time to time), it's that properties catch fire for a whole number of reasons and those properties sometimes contain gas cylinders (of which acetelene is the nastiest common one but even things like butane and propane can be pretty nasty). Furthermore until the fire brigade can contact the owner they often don't know if cylinders are present and if so what they contain.
While you're right on one level, acetylene is a special nasty case. The issue is that it's unstable at high pressure, so that safety valves on acetylene cylinders have to give way at relatively low temperatures and pressures. OK, this does mean that you're less likely to have them acting like an explosive shell, but it does mean that you have a real risk of the cylinder going off like a burning explosive rocket. A 200m exclusion zone seems very sensible to me!
If you can't explain to me, down to every detail, what Google was allowed to include then they have no footing in court as Apple can simply claim they assumed X feature was allowed, but apple didn't authorize it.
Sounds to me like there's a case possible. (Which isn't the same as saying that it would be a sure-fire win.)
Your claim that some random person needs to explain in depth to you what was promised or a court case will be fatally lacking is frankly BS. Firstly, you have got no standing on the matter anyway. Secondly, there's no good reason to believe that the GP has either. Thirdly, courts rule on conflicting claims all the time, and "he promised me one thing then did another leaving me out of pocket, and I want redress" is pretty much a classic basis for a civil lawsuit. (At a guess, the real determining factors would be whether the person who made the promise was competent to do so, whether they represented themselves as competent to do so, and whether such a promise was actually made. The position of "senior vice president of Worldwide Product Marketing" sounds to me like it's arguable that a promise made by them would be believed by others and they'd be able to make it for real too.)
thanks! I'll check it out. There is also an IT stumbling block - the folks who implement Blackboard will be the ones to shuffle to Moodle, and given how entrenched Blackboard and all of its sucktasticness (sp?) in my school's IT managerial class, I don't know if I would even be allowed to use Moodle parallel to Blackboard, much less in place of...
Sigh.
We have both Blackboard and Moodle deployed at my employer. Well, let's be more accurate. We've outsourced the BB deployment to BB themselves, and I'm only aware of one (admittedly large) school using Moodle. BB is (according to my spies) slow, irritating and vastly expensive (as well as requiring extra file servers deployed by us to host "large" files like short videos) and the users of Moodle seem fairly satisfied with it.
For my teaching, I'm sticking to using plain old HTML and PDFs hosted in a normal webserver that allows anyone to download them. I don't other people seeing them after all, and if they want to teach courses based on them... well, good luck!
Sean Kovacs, main developer of GV Mobile, says that he had personal approval for his app from Phil Shiller, Apple's senior vice president of Worldwide Product Marketing, last April.
If this bit is true and documented, then sue for lost development time. Apple gave assurances they wouldn't do something, Google committed resources, then Apple did it. Whatever Apple's reasoning here for changing their minds, they can't yank the football away any more than a contest promoter could decide not to give awards to a winner.
Sounds to my (admittedly untrained) ear like a time to apply the legal doctrine of estoppel, especially promissory estoppel. If I was Google, I'd be looking to recover as much as possible from Apple here, or (better yet) force the app down their throat, as that would vastly annoy both Apple and AT&T.
After weeding out people who got faked out by placebos and "active challenges", they got 100% positive, 0% negative.
They may have had an MD, but they sure didn't have anyone who knows anything about statistics. 100% positive in something potentially involving human physiology? Give me a break.
If I can go out and get a certificate signed by "FishWithAHammer" for a couple dozen bucks from some CA which happens to have its root certificate in your browser by default (and I can), even CA-signed certificates aren't worth much. Actually, the fact that you think a CA-signed cert is much better than a self-signed one means to me that they are causing more harm than good in the form of false security.
If you're going to set out to "fix" the certificate system, please start by coming up with an alternative proposal. Until you've got that idea for how to go forward, you're just an ass on a soapbox.
FWIW, I think the root CAs and the browser makers need to be stricter on the rest of the system. In particular, browsers need to start verifying certificate validity properly, and a subsidiary CA needs to have its master certificate invalidated for failing to follow the policy it agreed when it got it signed. The hows of pain (and lawsuits) from the CA and its customers will be unfortunate, but it's the only way. And a lot of organizations need to put their own houses in order; strict enforcement of the rules from the top is the only way that's going to ever happen.
Verisign is untrustworthy, so why should I care if a certificate is signed or not?
Signed certificates are a complete racket: If you don't pay us then when your users show up they will get a giant warning shown in their face, telling them not to trust you. You wouldn't want that would you? Nope, don't care who you are, what you do, or why. $100 bucks please.
So what do you propose instead that doesn't require the user to do something obnoxious like phoning the website owner up (possibly in the middle of the night for them) to get the key fingerprint? About the only thing that doesn't involve central authorities (*ahem* CAs *ahem*) is a Web of Trust, and that's very vulnerable to human failures (e.g., trusting a blackhat who can then poison the whole trustweb, or a previously good person turning to the dark side, or an account getting hacked). Claiming that you'll only trust people that you've manually verified might work for you in your mom's basement, but definitely won't scale out.
Challenge/response authentication using a credit card number and PIN as the encryption key. Let the bank issue the challenge, have the e-commerce site pass that right on to the browser. Let the browser do the encryption, and pass it all back to the bank via the site.
Too difficult to use.
The problem of security is in getting the right balance between protection and usability. (This is true for physical security too.)
commercial CAs provide no real identity verification and anyone who relies on them to do so is a fool.
Some do, some don't. Some offer varying levels of assurance, and asserting that the cheap levels are the same as the expensive ones is very disingenuous.
FWIW, I think that the minimum level that any CA ought to provide should be that I can connect a certificate to a legal identity (either an individual or a corporate body). Sure you won't have anonymity, but that's whole blasted point of this type of PKI.
Now, you may disagree with that line of reasoning, but supposedly that has been part of Apple's rationale for continuing to ship single-button mice.
That'd be OK, except that Apple hide functionality on context menus too; you just need Control-Click to bring them up. (A keyboard/mouse chord is never going to be as intuitive as a right-mouse-click on its own.)
So far, the MacbookPro that replaced it has held up much better, with the exception of two issues. (One was fixed in minutes after taking it in for service, and the other took around a week.)
[Following up to myself... Meh.]
The issue that was fixed in minutes was a bust power supply which went wrong after less than a month of use (it would power the machine, but wouldn't charge it).
The issue that took a week was a broken HDD. It took a week because it's not a user-serviceable part and I was out of the country on business when they actually finished. Thankfully I'd taken a backup the week before. (Don't think you're immune to a disk failure either; take a backup. Do it now.)
For reference, the Dell never had any failures like this at all until it got to about 6 years old, but also weighed a lot more and was far larger. And it ran XP. Up until that 6 year mark, when it stopped booting at all. Happily, Linux (specifically, Ubuntu) let me recover the data.
There's a difference between how films are generated and how the images on the screen are generated (google). Games are unplayable at 24fps. 50FPS is acceptable, but not good.
The problem is that you're using low quality cables to connect your video output hardware to your display. Luckily, Denon sell just the things you need to improve your performance (provided you remember to plug them in the correct way round; get it wrong and the electrons will be forced against their geomagnetic polarity and will go slower, increasing the amount of lag experienced).
Slashdot Mirror
Ever received "bad information" from a trusted friend? Or "bad information" from a friend of a trusted friend?
Do you remember the worms that used to propagate by emailing themselves out to everyone in the victim's Outlook address book? They'd spread like wildfire because everyone would trust who the message was coming from. More innocent times I know, but I'm damn sure it could happen again; people aren't any smarter than before. But my real point is that webs of trust really do get penetrated, and the result is usually disastrous because hardly anyone in the real world is security-conscious. At least with the CA/PKI model, the real security knowledge is only needed by a small group of people who damn well should know that they need to know what's going on.
The problem is the same with Moody's, actually: the central issue is that the people doing the auditing are being paid by the people they're auditing. Simply having browser users pay CAs (or investors pay rating agencies) would put the economic incentives in the right place, but that idea doesn't sit well with a lot of people.
So instead, we're left with imperfect and leaky regulation. CAs really should be subject to more regular audits, and their trust bits should be removed by browser vendors when they are abused.
Your analysis looks spot on to me, and the key issue is that browser "vendors" aren't being strict enough on enforcement. They're ideally placed to do so as gatekeepers for normal users, and they're also strongly incentivized to act as such. But if nobody enforces the rules then a system that depends on enforcement is just not going to work. After all, the only rule that really needs enforcing is this: verify that the certificate is issued to someone who has a right to it. (There's another critical rule for subsidiary CAs: you must enforce the first rule.) Everything else can be handled through normal contracts and making sure that people check certificate validity correctly, which is a software thing that we (the technologists) ought to be able to get right.
Isn't that why they charge huge amounts for the certs?
When was getting a signature from a crappy CA expensive? Is $30/yr for a basic server certificate terribly expensive all of a sudden (and I bet you can go cheaper than that if you hunt around) or are you deliberately going to a CA that is both bad and costly too? That'd be a new level of dumb...
M-x butterfly
Knowing emacs, to actually issue that command, you would have to press all those buttons at once.
Never following M-x; that let's you just type in the command name. (M-: is even better; lets you type raw elisp...)
In other news, emacs 24 is to be renamed "egacs" because the previous snarky backonym of "Eight Megabytes and Continually Swapping" is now the average footprint of "Hello World".
That's a invidious lie! It stands for "Escape Meta Alt Control Shift", and that's still a valid criticism (and I admit I really like emacs); there's only one shift-like key on this keyboard that it doesn't use and that's only because it's a laptop and so needs to put some functionality off a special "Function" shift...
Why does solar thermal energy seem so little loved around here? It looks like a rather cheap and proven technology to me: gather a few mirrors, a bathtub of salt and a steam machine in any sunny area and there you go.
Sounds like you're ready to go. Since it's obviously so proven, just build one and start producing power! (It's good for some parts of the world - especially deserts and semi-deserts - but other parts favor other solutions. What a surprise!)
*But*, the waste they do produce is *highly* radioactive (which is why it decays so fast), and *still* needs to be stored for *generations*.
If the waste is solid, insoluble and "hot", it's essentially self-protecting.
Right now, the real problem is that too many people are uncertain as to what the aim of protection is; it seems that many are after some magic whereby nothing bad can happen to anyone no matter how determined/stupid. Unsurprisingly, no such thing exists and that throws Greens in a total tizzy. (Of course, they mostly also think that Mother Nature is all good and can never hurt them provided they believe enough and hug the trees. Or am I overstating it a little?)
May is usually great, but it's been a couple of years that June and even part of July is crappy.
I wonder if there are correlations here with summers in the UK; they've been shit for the past few years too (the last properly good one was 2006).
It's not that the containers cause fires (though I would guess the welding torches fed from them do from time to time), it's that properties catch fire for a whole number of reasons and those properties sometimes contain gas cylinders (of which acetelene is the nastiest common one but even things like butane and propane can be pretty nasty). Furthermore until the fire brigade can contact the owner they often don't know if cylinders are present and if so what they contain.
While you're right on one level, acetylene is a special nasty case. The issue is that it's unstable at high pressure, so that safety valves on acetylene cylinders have to give way at relatively low temperatures and pressures. OK, this does mean that you're less likely to have them acting like an explosive shell, but it does mean that you have a real risk of the cylinder going off like a burning explosive rocket. A 200m exclusion zone seems very sensible to me!
If you can't explain to me, down to every detail, what Google was allowed to include then they have no footing in court as Apple can simply claim they assumed X feature was allowed, but apple didn't authorize it.
Sounds to me like there's a case possible. (Which isn't the same as saying that it would be a sure-fire win.)
Your claim that some random person needs to explain in depth to you what was promised or a court case will be fatally lacking is frankly BS. Firstly, you have got no standing on the matter anyway. Secondly, there's no good reason to believe that the GP has either. Thirdly, courts rule on conflicting claims all the time, and "he promised me one thing then did another leaving me out of pocket, and I want redress" is pretty much a classic basis for a civil lawsuit. (At a guess, the real determining factors would be whether the person who made the promise was competent to do so, whether they represented themselves as competent to do so, and whether such a promise was actually made. The position of "senior vice president of Worldwide Product Marketing" sounds to me like it's arguable that a promise made by them would be believed by others and they'd be able to make it for real too.)
thanks! I'll check it out. There is also an IT stumbling block - the folks who implement Blackboard will be the ones to shuffle to Moodle, and given how entrenched Blackboard and all of its sucktasticness (sp?) in my school's IT managerial class, I don't know if I would even be allowed to use Moodle parallel to Blackboard, much less in place of...
Sigh.
We have both Blackboard and Moodle deployed at my employer. Well, let's be more accurate. We've outsourced the BB deployment to BB themselves, and I'm only aware of one (admittedly large) school using Moodle. BB is (according to my spies) slow, irritating and vastly expensive (as well as requiring extra file servers deployed by us to host "large" files like short videos) and the users of Moodle seem fairly satisfied with it.
For my teaching, I'm sticking to using plain old HTML and PDFs hosted in a normal webserver that allows anyone to download them. I don't other people seeing them after all, and if they want to teach courses based on them... well, good luck!
Sean Kovacs, main developer of GV Mobile, says that he had personal approval for his app from Phil Shiller, Apple's senior vice president of Worldwide Product Marketing, last April.
If this bit is true and documented, then sue for lost development time. Apple gave assurances they wouldn't do something, Google committed resources, then Apple did it. Whatever Apple's reasoning here for changing their minds, they can't yank the football away any more than a contest promoter could decide not to give awards to a winner.
Sounds to my (admittedly untrained) ear like a time to apply the legal doctrine of estoppel, especially promissory estoppel. If I was Google, I'd be looking to recover as much as possible from Apple here, or (better yet) force the app down their throat, as that would vastly annoy both Apple and AT&T.
Would you settle for a member of the perciformes order with an attitude?
Provided they're one of the Moronidae, sure.
Ah! The existential angst of whether the ball will return, the skill of sending it back with just an expert flick of the wrist...
How is anywhere on Long Island remote, compared to most of the US?
For the true Manhattanite, Queens is indeed remote.
After weeding out people who got faked out by placebos and "active challenges", they got 100% positive, 0% negative.
They may have had an MD, but they sure didn't have anyone who knows anything about statistics. 100% positive in something potentially involving human physiology? Give me a break.
Until then you're...well this lady who had her house covered in tin foil.
Don't knock it! At least she's safe from the orbital mind control lasers operated by the Thetan Illuminati from Mars!
Why is a 60-year-old book so important to our modern culture under someone's copyright control anyway?
Because it's younger than Mickey Mouse. Thank you, Disney!
If I can go out and get a certificate signed by "FishWithAHammer" for a couple dozen bucks from some CA which happens to have its root certificate in your browser by default (and I can), even CA-signed certificates aren't worth much. Actually, the fact that you think a CA-signed cert is much better than a self-signed one means to me that they are causing more harm than good in the form of false security.
If you're going to set out to "fix" the certificate system, please start by coming up with an alternative proposal. Until you've got that idea for how to go forward, you're just an ass on a soapbox.
FWIW, I think the root CAs and the browser makers need to be stricter on the rest of the system. In particular, browsers need to start verifying certificate validity properly, and a subsidiary CA needs to have its master certificate invalidated for failing to follow the policy it agreed when it got it signed. The hows of pain (and lawsuits) from the CA and its customers will be unfortunate, but it's the only way. And a lot of organizations need to put their own houses in order; strict enforcement of the rules from the top is the only way that's going to ever happen.
Verisign is untrustworthy, so why should I care if a certificate is signed or not?
Signed certificates are a complete racket: If you don't pay us then when your users show up they will get a giant warning shown in their face, telling them not to trust you. You wouldn't want that would you? Nope, don't care who you are, what you do, or why. $100 bucks please.
So what do you propose instead that doesn't require the user to do something obnoxious like phoning the website owner up (possibly in the middle of the night for them) to get the key fingerprint? About the only thing that doesn't involve central authorities (*ahem* CAs *ahem*) is a Web of Trust, and that's very vulnerable to human failures (e.g., trusting a blackhat who can then poison the whole trustweb, or a previously good person turning to the dark side, or an account getting hacked). Claiming that you'll only trust people that you've manually verified might work for you in your mom's basement, but definitely won't scale out.
Challenge/response authentication using a credit card number and PIN as the encryption key. Let the bank issue the challenge, have the e-commerce site pass that right on to the browser. Let the browser do the encryption, and pass it all back to the bank via the site.
Too difficult to use.
The problem of security is in getting the right balance between protection and usability. (This is true for physical security too.)
commercial CAs provide no real identity verification and anyone who relies on them to do so is a fool.
Some do, some don't. Some offer varying levels of assurance, and asserting that the cheap levels are the same as the expensive ones is very disingenuous.
FWIW, I think that the minimum level that any CA ought to provide should be that I can connect a certificate to a legal identity (either an individual or a corporate body). Sure you won't have anonymity, but that's whole blasted point of this type of PKI.
Now, you may disagree with that line of reasoning, but supposedly that has been part of Apple's rationale for continuing to ship single-button mice.
That'd be OK, except that Apple hide functionality on context menus too; you just need Control-Click to bring them up. (A keyboard/mouse chord is never going to be as intuitive as a right-mouse-click on its own.)
So far, the MacbookPro that replaced it has held up much better, with the exception of two issues. (One was fixed in minutes after taking it in for service, and the other took around a week.)
[Following up to myself... Meh.]
The issue that was fixed in minutes was a bust power supply which went wrong after less than a month of use (it would power the machine, but wouldn't charge it).
The issue that took a week was a broken HDD. It took a week because it's not a user-serviceable part and I was out of the country on business when they actually finished. Thankfully I'd taken a backup the week before. (Don't think you're immune to a disk failure either; take a backup. Do it now.)
For reference, the Dell never had any failures like this at all until it got to about 6 years old, but also weighed a lot more and was far larger. And it ran XP. Up until that 6 year mark, when it stopped booting at all. Happily, Linux (specifically, Ubuntu) let me recover the data.
There's a difference between how films are generated and how the images on the screen are generated (google). Games are unplayable at 24fps. 50FPS is acceptable, but not good.
The problem is that you're using low quality cables to connect your video output hardware to your display. Luckily, Denon sell just the things you need to improve your performance (provided you remember to plug them in the correct way round; get it wrong and the electrons will be forced against their geomagnetic polarity and will go slower, increasing the amount of lag experienced).