Capacitors, at least the few I have seen, generally want to release their stored energy all at once. How is this addressed when supercapacitors are used? For example, let's say you have a supercapacitor that can power a light bulb for eight hours. How do you make it actually provide a lower current over those eight hours instead of providing all of that energy in a single instant and frying the bulb?
A capacitor has a certain voltage (whatever you charged it to) and an internal resistance. The load (light bulb) you attach also has a certain resistance. The discharge rate is determined by those two resistances (added together) and the voltage. The "all at once" just means that the internal resistance is almost zero, so if you connect a load that also has zero resistance the capacitor will discharge very quickly. If your load doesn't have near-zero resistance, there won't be any real difference from using a battery.
My first thought would be that coming up with bearings for a flywheel that can handle the mass of the wheel yet be as close to frictionless as possible would be difficult and expensive to develop and then later to maintain.
Use the same tech they make maglev trains with, and put the whole thing in a big shell you can pump all the air out of.
For instance, the government doesn't know how many people actually have broadband or what they pay for it
Wait, I thought telcos giving the government open access to their records was a bad thing.
Come'on folks. You know anything the government touches will be abused. Stop giving these appointees more power for next to no real gain.
You don't see a difference between "Here's a log of all calls to/from 555-1234, registered to Joe Nacho" and "here's a list of how many subscribers we have at each speed in each zip code"? I tend to see personal information as rather different from statistics.
Central planning will always lead to ditches to nowhere.
[citation needed]
Without an ability to perform rationale economic calculations, an economy cannot function. Any effort by the State to manipulate or direct economic planning will lead to increasing economic irrationality and inefficiency.
The perfect efficiency that markets try to approach is short-sighted (this also makes them unstable, with a tendency to collapse to monopoly/oligopoly). Some diversion of resources towards longer-term goals is useful, why do you think any country has a public education system?
The only way to maximize the efficient use of resources is to remove government coercion from the marketplace, and let voluntary cooperation and aggregate individual choices locate the closet to optimally possible solution to any problem.
You also have to eliminate all other forms of coercion and tying and collusion, and provide everyone with perfect information and zero transaction costs. And you still end up with that short-sightedness.
It's referring to the Verizon portal which is third-party to the web service. Note the singular use of "the third party Web site" rather than sites.
How about someone settle this once and for all and call fairpoint asking for clarification.
Customers with questions can call FairPoint at (800) 240-5019.
But Verizon isn't a "third party", they're with Fairpoint as a first party (making the changes) against the subscribers as the other first party (who the changes were made to). Now, it sounds like Verizon might have a site with "third-party" in the name, so if that's the case...
It is eminently plausible that the author of the article was confused, AND Fairpoint was talking about the third-party Verizon portal for accessing Yahoo mail and other webmail services.
...Verizon has a site called the "third-party portal"? Yeah, I can see how that could get confusing in a situation where there are real third parties, none of whom are Verizon.
Verizon provided a service to IT'S customers where they can read webmail of another provider on their web page. Fairpoint is saying that after x date that if you still want that kind of service you have to go through THEIR web page. You can still go to Yahoo, Gmail, AOL, and Hotmail, and read your mail from those pages directly.
Article says:
Web-based e-mail users can continue to access their e-mail at the Verizon Web site until Feb. 6. After that date, Fastiggi said users will need to log on to www.MyFairPoint.net. Customers then click on Web mail and type in their existing user name@myfairpoint.net and existing password.
AOL, Yahoo! and MSN subscribers will continue to have access to content but will no longer be able to access their e-mail through the third party Web site. Instead, Yahoo! and other third party e-mail will be accessed directly at the MyFairPoint.net portal.
So, (1) customersl ISP email addresses are changing, and (2) people with certain non-Fairpoint webmails will have to read their email through MyFairPoint.net instead of "the third party Web site" which I take to mean the AOL/MSN/etc site ("third party" being, not Fairpoint or the subscriber).
I DID RTFA, and it certainly seems to be saying what you say it isn't. Do you have a better-worded article?
If we define notebooks as small laptops with processors in the Atom class
...then we're confusing "notebooks" and "netbooks". The former are what used to be called "laptops" before the litigious folks realized they could fry your balls, and the latter are overgrown PDAs that still don't have enough pixels to do anything useful with.
SSL is all about trust, and the CA you trust for your site's cert does matter. There's only so much a CA can practically do to establish your (corporate) identity, but that's what you trust them to do every time they are asked to issue a certificate in your name. It's not rocket science, this is a basic level of trust that everyone needs. We expect, and trust ANYONE who handles our personal information to properly, consistently verify our identity.
The reason it doesn't matter, is that there's nothing to prevent some other CA from also issuing certs for your site. Even when you have absolutely no contact with or trust in that other CA.
What?? Let me get this straight... you want to pay for a whitelist(s) of the Internet?
CAs are already a whitelist mechanism, the only difference is who pays. I think it should be whoever benefits from the list being accurate, so the list maintainers have an incentive to do their job properly.
The whole point of a CA is for them to verify who you (site owner) are.
And yet somehow, nobody cares. If people did care, the UI would reflect that (like it does for EV certs).
This is as if you said that a shopping center should not pay for the security guards, but the customers that visit this shopping center should.
The web site provides services to its users (usually not for free - even if only for the revenue from advertising), and security is just one of these services.
The site cannot provide (this kind of) security, it can only be implemented on the user's system. Because of this, it should be the responsibility of the user (or realistically, the browser maker) rather than the site owner.
If there is a problem with this system, it is not a design problem, it's a problem with implementation. If users are ignoring the warnings, then they should be more persistent and explanatory (like the Firefox warnings are now, though I'm not so sure about the "explanatory" part). If the CAs sign keys without verifying their owners, then they should be punished. If there is no authority that regulates the CAs (checking whether their verification processes are thorough enough, etc.), then maybe there should.
The "need" for punishment (which would have collateral damage) and regulation is how you should know that the system is broken. The built-in incentives do not align properly to make the system work on its own, so external forces must be imposed. That makes it a bad design.
My point is really that shifting the burden of proof from one entity to the other doesn't really address the real problem. The problem isn't that one party has more interest in security than the other (sometimes true, sometimes not), it's that the weakest CA ruins it for everyone else.
So make it so that you don't need multiple authorities, or can use multiple authorities with AND instead of OR for combining individual results. The way to do this is to make there be only one business relationship, end-user <-> verifier. The verifier can check that a site consistenly presents the same cert across time and network paths, and does not need a relationship with the site owner to do this. If a verifier proves to be incompetent, there's no collateral damage from dropping them. I can use multiple verifiers without site owners needing to care or even knowing.
What do you mean they have no incentive for competent CA's?
The security of my site's users doesn't depend on the competence of whichever CA I choose, it depends on the competence of the least competent CA trusted by their browser. So there's no incentive for me to pay extra for a more competent CA, because their competence (or lack thereof) doesn't really affect anything.
A better system would involve only the CA who the site owner contracted with has to be trusted for that sites security.
I don't think that's possible, how do you know which CA that is?
Pesronally I'd like VISA and Mastercard to give me Root CA certs to use for purchasing on line - then if they foul up and someone gets a dodgy cert then they pick up the bill.
Hey, I like that idea. But it would need a way for the site to have multiple certificates where the user chooses which one they want (does SSL/TLS permit this?), and support for this in the browser UI... I suppose the root cert would probably come on a mini-CD or USB stick when they mail your card.
(they can't know which CA the site owner actually paid)
Isn't that what the 'issued by' field is for in a certificate?
No, that says which CA issues that particular cert. Which may not be the same CA that the site owner paid, for example if I pay Verisign for a cert for my site and when you visit you get a cert from "one of Comodo's resellers", you really have no way to know that I didn't buy a cert from them and you shouldn't trust it.
Gotta think over the SSL certs one more. I never really liked the mechanism behind it, i like it even less now.
The current mechanism is that the site owner pays a particular CA to identify them, and end-users/browsers trust any CA to identify any site (they can't know which CA the site owner actually paid). Site owners (the ones paying the bills) have no incentive to demand that the CA be competent.
A better system would have the end-user pay someone they trust to identify the site; they are directly paying for the identification service and can take their business elsewhere if they get crap service. This would also mean that the site owners don't have to pay someone who, really, can't actually provide any assurances to the end-users (because all CA-signed certs are treated the same). Better to just have everyone go self-signed, and then let someone (paid by the end-users) keep records of who used what cert when as seen by what network routes.
Mapping to real-world identities is a separate issue (only provided by "extended validation" or whatever certs due to browser UI issues), and is (1) rather expensive because you need people involved to look at paperwork and such and (2) mostly isn't needed, because you'll generally find IRL groups' sites by communication from those groups (eg, my electric bill has the electric company's URL printed on it, I don't need to look them up in google and then verify that I got pointed to the right place).
Except that previous drives have well-known severe problems with random IO, so I'm kinda suspicious that they specified sequential speeds rather than take the opportunity to say "see, we don't have that bug that everyone else does".
there haven't been any articles recently addressing fundamental problems for long term practical use.
I wasn't aware that there were any, just a few implementation issues with the early drives (bad wear leveling and f-ed up controllers that can't multitask).
Slashdot Mirror
Capacitors, at least the few I have seen, generally want to release their stored energy all at once. How is this addressed when supercapacitors are used? For example, let's say you have a supercapacitor that can power a light bulb for eight hours. How do you make it actually provide a lower current over those eight hours instead of providing all of that energy in a single instant and frying the bulb?
By understanding I=V/R.
A capacitor has a certain voltage (whatever you charged it to) and an internal resistance. The load (light bulb) you attach also has a certain resistance. The discharge rate is determined by those two resistances (added together) and the voltage. The "all at once" just means that the internal resistance is almost zero, so if you connect a load that also has zero resistance the capacitor will discharge very quickly. If your load doesn't have near-zero resistance, there won't be any real difference from using a battery.
My first thought would be that coming up with bearings for a flywheel that can handle the mass of the wheel yet be as close to frictionless as possible would be difficult and expensive to develop and then later to maintain.
Use the same tech they make maglev trains with, and put the whole thing in a big shell you can pump all the air out of.
For instance, the government doesn't know how many people actually have broadband or what they pay for it
Wait, I thought telcos giving the government open access to their records was a bad thing.
Come'on folks. You know anything the government touches will be abused. Stop giving these appointees more power for next to no real gain.
You don't see a difference between "Here's a log of all calls to/from 555-1234, registered to Joe Nacho" and "here's a list of how many subscribers we have at each speed in each zip code"? I tend to see personal information as rather different from statistics.
Central planning will always lead to ditches to nowhere.
[citation needed]
Without an ability to perform rationale economic calculations, an economy cannot function. Any effort by the State to manipulate or direct economic planning will lead to increasing economic irrationality and inefficiency.
The perfect efficiency that markets try to approach is short-sighted (this also makes them unstable, with a tendency to collapse to monopoly/oligopoly). Some diversion of resources towards longer-term goals is useful, why do you think any country has a public education system?
The only way to maximize the efficient use of resources is to remove government coercion from the marketplace, and let voluntary cooperation and aggregate individual choices locate the closet to optimally possible solution to any problem.
You also have to eliminate all other forms of coercion and tying and collusion, and provide everyone with perfect information and zero transaction costs. And you still end up with that short-sightedness.
It's referring to the Verizon portal which is third-party to the web service. Note the singular use of "the third party Web site" rather than sites.
How about someone settle this once and for all and call fairpoint asking for clarification.
Customers with questions can call FairPoint at (800) 240-5019.
But Verizon isn't a "third party", they're with Fairpoint as a first party (making the changes) against the subscribers as the other first party (who the changes were made to). Now, it sounds like Verizon might have a site with "third-party" in the name, so if that's the case...
It is eminently plausible that the author of the article was confused, AND Fairpoint was talking about the third-party Verizon portal for accessing Yahoo mail and other webmail services.
...Verizon has a site called the "third-party portal"? Yeah, I can see how that could get confusing in a situation where there are real third parties, none of whom are Verizon.
Verizon provided a service to IT'S customers where they can read webmail of another provider on their web page. Fairpoint is saying that after x date that if you still want that kind of service you have to go through THEIR web page. You can still go to Yahoo, Gmail, AOL, and Hotmail, and read your mail from those pages directly.
Article says:
So, (1) customersl ISP email addresses are changing, and (2) people with certain non-Fairpoint webmails will have to read their email through MyFairPoint.net instead of "the third party Web site" which I take to mean the AOL/MSN/etc site ("third party" being, not Fairpoint or the subscriber).
I DID RTFA, and it certainly seems to be saying what you say it isn't. Do you have a better-worded article?
...why didn't I see Gmail on their list?
Sure, it would make some businesses angry, but they don't vote.
I thought campaign contributions were considered more valuable than individual votes.
If we define notebooks as small laptops with processors in the Atom class
...then we're confusing "notebooks" and "netbooks". The former are what used to be called "laptops" before the litigious folks realized they could fry your balls, and the latter are overgrown PDAs that still don't have enough pixels to do anything useful with.
SSL is all about trust, and the CA you trust for your site's cert does matter. There's only so much a CA can practically do to establish your (corporate) identity, but that's what you trust them to do every time they are asked to issue a certificate in your name. It's not rocket science, this is a basic level of trust that everyone needs. We expect, and trust ANYONE who handles our personal information to properly, consistently verify our identity.
The reason it doesn't matter, is that there's nothing to prevent some other CA from also issuing certs for your site. Even when you have absolutely no contact with or trust in that other CA.
What?? Let me get this straight... you want to pay for a whitelist(s) of the Internet?
CAs are already a whitelist mechanism, the only difference is who pays. I think it should be whoever benefits from the list being accurate, so the list maintainers have an incentive to do their job properly.
The whole point of a CA is for them to verify who you (site owner) are.
And yet somehow, nobody cares. If people did care, the UI would reflect that (like it does for EV certs).
This is as if you said that a shopping center should not pay for the security guards, but the customers that visit this shopping center should.
The web site provides services to its users (usually not for free - even if only for the revenue from advertising), and security is just one of these services.
The site cannot provide (this kind of) security, it can only be implemented on the user's system. Because of this, it should be the responsibility of the user (or realistically, the browser maker) rather than the site owner.
If there is a problem with this system, it is not a design problem, it's a problem with implementation. If users are ignoring the warnings, then they should be more persistent and explanatory (like the Firefox warnings are now, though I'm not so sure about the "explanatory" part). If the CAs sign keys without verifying their owners, then they should be punished. If there is no authority that regulates the CAs (checking whether their verification processes are thorough enough, etc.), then maybe there should.
The "need" for punishment (which would have collateral damage) and regulation is how you should know that the system is broken. The built-in incentives do not align properly to make the system work on its own, so external forces must be imposed. That makes it a bad design.
My point is really that shifting the burden of proof from one entity to the other doesn't really address the real problem. The problem isn't that one party has more interest in security than the other (sometimes true, sometimes not), it's that the weakest CA ruins it for everyone else.
So make it so that you don't need multiple authorities, or can use multiple authorities with AND instead of OR for combining individual results. The way to do this is to make there be only one business relationship, end-user <-> verifier. The verifier can check that a site consistenly presents the same cert across time and network paths, and does not need a relationship with the site owner to do this. If a verifier proves to be incompetent, there's no collateral damage from dropping them. I can use multiple verifiers without site owners needing to care or even knowing.
What do you mean they have no incentive for competent CA's?
The security of my site's users doesn't depend on the competence of whichever CA I choose, it depends on the competence of the least competent CA trusted by their browser. So there's no incentive for me to pay extra for a more competent CA, because their competence (or lack thereof) doesn't really affect anything.
A better system would involve only the CA who the site owner contracted with has to be trusted for that sites security.
I don't think that's possible, how do you know which CA that is?
Pesronally I'd like VISA and Mastercard to give me Root CA certs to use for purchasing on line - then if they foul up and someone gets a dodgy cert then they pick up the bill.
Hey, I like that idea. But it would need a way for the site to have multiple certificates where the user chooses which one they want (does SSL/TLS permit this?), and support for this in the browser UI... I suppose the root cert would probably come on a mini-CD or USB stick when they mail your card.
(they can't know which CA the site owner actually paid)
Isn't that what the 'issued by' field is for in a certificate?
No, that says which CA issues that particular cert. Which may not be the same CA that the site owner paid, for example if I pay Verisign for a cert for my site and when you visit you get a cert from "one of Comodo's resellers", you really have no way to know that I didn't buy a cert from them and you shouldn't trust it.
Gotta think over the SSL certs one more. I never really liked the mechanism behind it, i like it even less now.
The current mechanism is that the site owner pays a particular CA to identify them, and end-users/browsers trust any CA to identify any site (they can't know which CA the site owner actually paid). Site owners (the ones paying the bills) have no incentive to demand that the CA be competent.
A better system would have the end-user pay someone they trust to identify the site; they are directly paying for the identification service and can take their business elsewhere if they get crap service. This would also mean that the site owners don't have to pay someone who, really, can't actually provide any assurances to the end-users (because all CA-signed certs are treated the same). Better to just have everyone go self-signed, and then let someone (paid by the end-users) keep records of who used what cert when as seen by what network routes.
Mapping to real-world identities is a separate issue (only provided by "extended validation" or whatever certs due to browser UI issues), and is (1) rather expensive because you need people involved to look at paperwork and such and (2) mostly isn't needed, because you'll generally find IRL groups' sites by communication from those groups (eg, my electric bill has the electric company's URL printed on it, I don't need to look them up in google and then verify that I got pointed to the right place).
Since every body of water is exactly 0 feet deep at the coastline, by definition, that last bit is pretty obvious.
What about when the coastline is actually a sheer cliff?
Isn't this classical "act of terrorism" ?
But reduced connectivity isn't really that terrifying, at least to us humans. Maybe if you were a distributed AI...
Right here!
He cared deeply about the human dilemma and the rape of Earth.
He advocates killing 90% of the worlds population with the ebola virus. http://www.sas.org/tcs/weeklyIssues_2006/2006-04-07/feature1p/index.html
IIRC, it was more "considers it inevitable" rather than "advocates".
Since SSDs don't have any seek time (or if they do, it's probably not measurable) the sequential and non-sequential I/O should be indifferent.
Yes, should. But most current SSDs have buggy controllers that can't really handle random IO.
Except that previous drives have well-known severe problems with random IO, so I'm kinda suspicious that they specified sequential speeds rather than take the opportunity to say "see, we don't have that bug that everyone else does".
there haven't been any articles recently addressing fundamental problems for long term practical use.
I wasn't aware that there were any, just a few implementation issues with the early drives (bad wear leveling and f-ed up controllers that can't multitask).
I only see numbers for sequential access (240MB/s read, 200MB/s write). I don't suppose anyone knows how it does for random read/write speed?