Re:This is how economics is supposed to work!
on
The SUV Is Dethroned
·
· Score: 2, Insightful
Yes - as long as the cost you pay for gasoline is the true cost, including externalities like its effect on the environment. Which will be a bit higher than just the cost of getting it out of the ground.
However, if you wrote and published an essay 'Why Child Abuse Is A Good Idea', you would also be encouraging crimes against children. Yet few argue you should be arrested for doing so. It's hard to see why computer graphics 'encourage abuse' any more than talking about it or writing books about it.
SSL to an untrusted host is just as bad as no ssl because the man-in-the-middle (which is kind of the definition of an ISP) could easily produce a certificate that says, "hey, I'm what ever page you wanted to look at". And the insert ads.
It depends on your definition of easily. As I said earlier, I don't think the technical staff at ISPs are whizzkids enough to understand and implement man-in-the-middle attacks against SSL, there certainly isn't off-the-shelf hardware to do it, it's computationally expensive, and you can potentially get in a lot of legal trouble for it (much more than just injecting advertisements).
So I don't agree that https without a signed certificate is no better than unencrypted http. It may not be good enough for Paypal or even Gmail, but it's a lot better than plain http for ordinary websites that just want to serve a page and have the user see it without tampering by the ISP.
Bandying about GHz numbers and arguing about whether or not it's a real PowerPC processor doesn't get you anywhere. Let's see some results from the only benchmark that matters: the Linux kernel compile benchmark. (It can be accepted by all sides that the PS3 has some fairly monstrous floating point hardware, but sadly that doesn't speed up gcc.)
Yeah it sucks that you have to either pay money or endure scary messages from the web browser. There should be a way to label your site as self-signed where it wouldn't get the special secure icon or magic green glowing bar in the web browser, but on the other hand the user wouldn't be pestered about an invalid certificate (unless the cert offered really has changed since last time the user visited the site).
Doing man-in-the middle attacks on SSL connections is beyond the technical ability of ISPs, even if users don't bother to check certificates. And the potential for them to get in trouble for it is a lot higher (e.g. if they ended up intercepting financial information, and then the ISP's servers got cracked...). So https is still the right answer here.
It's 2008, why aren't most websites just using https by default? A low-volume site can handle the load with today's superfast CPUs, and high-volume sites can afford to buy one of those crypto engine thingies.
Medical services in the US? Check. Water utilities in the UK? Check. Power companies in the US? Check. Major ISPs in Australia, Canada, the US?
In most of those cases, there is very little competition, so if you get poor service you're stuck with it. Private enterprise generally does better when there is strong competition between providers and you can choose the one you want. Clearly this can be the case with schools - there is no need for all schools in an area to be run by the same MegaSchoolCorp.
NB there is no reason why the organization running a school must be for-profit. Universities are independent from government, they are not run for profit, and they seem to work okay. If you don't think a for-profit company can do a good job, you can send your child to one of the not-for-profit schools. That kind of choice is the whole idea.
The problem is that it's easy enough to trap antiprotons and positrons in electromagnetic fields. It's even fairly straightforwad to put them together so that they form antihydrogen. The problem is that antihydrogen is neutral and simply falls out of the trap.
This surely contains a clue about whether antihydrogen falls up or down...
I downloaded Safari for Windows yesterday but it's just... urgh. The font rendering is all smudgy and changing the preferences doesn't seem to help. The window decorations and controls are weird and non-standard. Dialogue boxes do bizarre things like stretching themselves up and down without warning. There used to be a time when Apple was all in favour of user interface guidelines and making apps look and feel consistent, but they seem unable to even try to do this when developing for a platform they don't control.
The one bright spot is that the 'back' button is hella fast. Why can't Firefox cache the bitmap image of the previous page, or something, and snap to it instantly when you go back?
Yes I realize that InnoDB doesn't have these crazy limitations, but MyISAM is much faster,
That in a nutshell is the problem with any discussion of MySQL - or indeed with MySQL itself. Any limitation like 'doesn't support tables >4Gibyte' can be answered with 'well use a different table type'. Meanwhile, any benchmark or scalability test can be won by MySQL by using MyISAM tables. It's not really meaningful to talk about 'MySQL' at all; you should say 'MySQL/MyISAM', 'MySQL/InnoDB' and so on. They are different database systems with different characteristics.
This assumes that a Windows machine has a meaningful distinction between per-user and system-wide settings, which is sadly not the case, whatever the original good intentions of the Windows designers.
You just do it. In your typical ZODB application there is no difference between doing something on objects in memory and doing it on objects stored in the database.
There must be a small difference in that changes to several database objects can be rolled back or committed together as a single transaction. (If the database doesn't support that, then it's not really worthy of the name.) But you are talking about the higher-level interface. If under the covers your code is still operating procedurally one record at a time, then it's never going to be as efficient as the set-oriented commands available in a relational system. That doesn't matter a lot of the time. Most applications are not dealing with large data volumes, and if the data takes 0.05s to process on modern hardware, who cares about making it ten times faster? But then again, if performance is not an issue, using a clunky object-relational mapping on top of an RDBMS will also get the job done, and you can still access set-based queries and update commands if you need them for efficiency.
A relational database system usually allows you to write stored procedures to do the data manipulation entirely in the server without roundtrips. That's not the issue. What matters is your second point 'forcing you to express the idea with relational operators only'. Sometimes relational operators are a good fit for the data model and, if you twist your neurons in the right way to think of the problem relationally, allow you a much higher-level way to express an algorithm than procedural code, while keeping good efficiency. But there are many cases where the relational model doesn't seem to fit.
The question is, what powerful primitives does an OODBMS offer you? What efficiency advantages do they have over naive 'fetch record; do something; fetch another record' procedural code for dealing with large amounts of data? How can they take advantage of indexing or optimize cache usage to reduce disk seeks?
What are those "problematic ideas" you are taling about and how does an RDBMS help us with analyzing large tree and/or graph datasets, knowledge bases and other forms of data that does not fit the relational model?
How does an object-oriented database system help with them?
35. Santa Monica-based incubator eCompanies pays $7.5 million for the domain name Business.com in November 1999; explaining the purchase, eCompanies co-founder Sky Dayton tells Internet World, "It is going to be the bargain of the century. It is going to look like we bought the island of Manhattan for $7.5 million and some beads."
Without RST packets, how are you supposed to know if the remote host is legitimately closing the connection?
Um, IPsec?
Point is, if your ISP spoofs RST packets, you cannot know when the remote host is legitimately closing the connection. If you get such a packet it could be genuine or it could be a fake. So it doesn't tell you much. You need some means for the remote host to sign every packet it sends out so they can't be spoofed, or else stop trusting them.
Unicomp also make keyboards with a 'nipple' pointing device in the middle like on Thinkpads. The Endurapro is buckling spring with pointing device and is available as USB. The only downside is that they can't ship the USB version outside the USA.
I'd like to get one but currently I have a good stock of Model Ms for my typing needs.
What I really want to use is the old PC or PC-XT keyboard - buckling spring but even heavier and better built than the Model M. However the electronics are different. I think I saw an adapter on sale for $100 somewhere but that's a bit steep.
Bah. The yardstick is not 'has nothing to do with Microsoft Windows' but rather 'gives you and others freedom to use, share and change the software'. If you just wanted to eliminate Microsoft you could buy a Mac and not buy Office for it. Mono is completely free and open source software. Yes, it is a clone of a proprietary system, just as GNU started out as a clone of proprietary Unix.
Slashdot Mirror
Yes - as long as the cost you pay for gasoline is the true cost, including externalities like its effect on the environment. Which will be a bit higher than just the cost of getting it out of the ground.
I think it jumped the shark with 2.0.
Why else would you buy one?
However, if you wrote and published an essay 'Why Child Abuse Is A Good Idea', you would also be encouraging crimes against children. Yet few argue you should be arrested for doing so. It's hard to see why computer graphics 'encourage abuse' any more than talking about it or writing books about it.
So I don't agree that https without a signed certificate is no better than unencrypted http. It may not be good enough for Paypal or even Gmail, but it's a lot better than plain http for ordinary websites that just want to serve a page and have the user see it without tampering by the ISP.
Bandying about GHz numbers and arguing about whether or not it's a real PowerPC processor doesn't get you anywhere. Let's see some results from the only benchmark that matters: the Linux kernel compile benchmark. (It can be accepted by all sides that the PS3 has some fairly monstrous floating point hardware, but sadly that doesn't speed up gcc.)
Yeah it sucks that you have to either pay money or endure scary messages from the web browser. There should be a way to label your site as self-signed where it wouldn't get the special secure icon or magic green glowing bar in the web browser, but on the other hand the user wouldn't be pestered about an invalid certificate (unless the cert offered really has changed since last time the user visited the site).
Doing man-in-the middle attacks on SSL connections is beyond the technical ability of ISPs, even if users don't bother to check certificates. And the potential for them to get in trouble for it is a lot higher (e.g. if they ended up intercepting financial information, and then the ISP's servers got cracked...). So https is still the right answer here.
It's 2008, why aren't most websites just using https by default? A low-volume site can handle the load with today's superfast CPUs, and high-volume sites can afford to buy one of those crypto engine thingies.
NB there is no reason why the organization running a school must be for-profit. Universities are independent from government, they are not run for profit, and they seem to work okay. If you don't think a for-profit company can do a good job, you can send your child to one of the not-for-profit schools. That kind of choice is the whole idea.
I downloaded Safari for Windows yesterday but it's just... urgh. The font rendering is all smudgy and changing the preferences doesn't seem to help. The window decorations and controls are weird and non-standard. Dialogue boxes do bizarre things like stretching themselves up and down without warning. There used to be a time when Apple was all in favour of user interface guidelines and making apps look and feel consistent, but they seem unable to even try to do this when developing for a platform they don't control.
The one bright spot is that the 'back' button is hella fast. Why can't Firefox cache the bitmap image of the previous page, or something, and snap to it instantly when you go back?
We have decided to crush your worthless civilization. Prepare for WAR!
This assumes that a Windows machine has a meaningful distinction between per-user and system-wide settings, which is sadly not the case, whatever the original good intentions of the Windows designers.
A relational database system usually allows you to write stored procedures to do the data manipulation entirely in the server without roundtrips. That's not the issue. What matters is your second point 'forcing you to express the idea with relational operators only'. Sometimes relational operators are a good fit for the data model and, if you twist your neurons in the right way to think of the problem relationally, allow you a much higher-level way to express an algorithm than procedural code, while keeping good efficiency. But there are many cases where the relational model doesn't seem to fit.
The question is, what powerful primitives does an OODBMS offer you? What efficiency advantages do they have over naive 'fetch record; do something; fetch another record' procedural code for dealing with large amounts of data? How can they take advantage of indexing or optimize cache usage to reduce disk seeks?
My favourite:
Point is, if your ISP spoofs RST packets, you cannot know when the remote host is legitimately closing the connection. If you get such a packet it could be genuine or it could be a fake. So it doesn't tell you much. You need some means for the remote host to sign every packet it sends out so they can't be spoofed, or else stop trusting them.
Unicomp also make keyboards with a 'nipple' pointing device in the middle like on Thinkpads. The Endurapro is buckling spring with pointing device and is available as USB. The only downside is that they can't ship the USB version outside the USA.
I'd like to get one but currently I have a good stock of Model Ms for my typing needs.
What I really want to use is the old PC or PC-XT keyboard - buckling spring but even heavier and better built than the Model M. However the electronics are different. I think I saw an adapter on sale for $100 somewhere but that's a bit steep.
Have you seriously investigated the hardware costs of building your own cluster? With equivalent specs to this one?
Hint: it's not about how many CPUs you have or how fast they are, it's how fast the interlinks are between processors.
Bah. The yardstick is not 'has nothing to do with Microsoft Windows' but rather 'gives you and others freedom to use, share and change the software'. If you just wanted to eliminate Microsoft you could buy a Mac and not buy Office for it. Mono is completely free and open source software. Yes, it is a clone of a proprietary system, just as GNU started out as a clone of proprietary Unix.