Note that what Dell is selling is "Dell-branded" printers. Dell doesn't manufacture printers -- Dell doesn't manufacture much of anything (at best, they assemble things). HP, on the other hand, actually makes printers. (Who knows, maybe some of the printers that Dell sells will be made by HP some day.)
Dell has sold printers for a long time. As far as I can tell, they target buyers who like to buy everything through one web site. The peripherals they sell are nothing special, and the prices aren't that good, but it's easy and convenient to buy everything with one click.
People who want the best are usually willing to shop around for it. Hopefully HP won't be run out of business if Dell is successful in undermining their market, and the next time I want a good, dependable printer I won't have to buy a re-branded Lexmark or some other similar junk.
The problem with "plain vanilla ASCII" is that expresses less information than is present in the original and is a PITA to recover this information. This is especially true for texts that use non-ASCII characters (or illustrations of any kind).
I agree that flavor of the month representations are bad, but markup languages have been around for a long time and it wouldn't have been hard to use something (like small subset of SGML) to add a bit more formatting info. Then when people want to look at the text in the flavor-of-the-month format, it's just a matter of writing a translator for that format. (illustrations are another matter, but I suspect that things have converged enough in this area so that something could be done from this point forward.)
Don't get me wrong -- I have a huge respect for PG and what they're doing is a benefit to humanity. I just wish that they would aim a little higher than the lowest common denominator for representation, and support other character sets in a simpler manner.
4. The fourth folk belief is that passwords based on mnemonic phrases are harder to remember than naively selected passwords.
Is this a typo, or is there a new meaning of "mnemonic"? The whole point of mnemonic passwords is that they're easy to remember. That's what mnemonic means.
I was having all kinds of problems browsing the web last night: about half the sites I tried to visit (including slashdot and freebsd.org) simply failed to connect. The others were perfectly fine. I didn't see any pattern to it, but I wasn't looking very hard.
Since I've had problems like this with my ISP, I figured it was something local. I guess not.
OK, moderate me redundant because now I see a million other people saw the same thing...
This might seem completely nuts, but in the good old USA there are many companies that use personality profiling tests, often based on debunked psychology theories, to decide who to hire and what tasks to give them. The faith that people put in these tests is astonishing.
Clearly the person who wrote this is not a genuine nerd -- only a nerd wannabe would make such a mistake, not knowing that to nerds the misuse of jargon is like waving a red flag in front of a bull.
When they come out with one that has god-like coding skills, that's when I'll get interested. Maybe.
Whether or not the Java programming language is horrible is not even tangentially related to this topic. The Java desktop has nothing to do with the Java programming language! (why Sun decided to call this the JDS is a mystery to me -- maybe they're just nuts)
The rest of your comment is troll-ish. If you want people to use your language, you either do a media blitz (Ada, Java, C#, Borland this and that, etc) or else you wait a long time (5-10 years) for word-of-mouth to spread (C, Perl, Python...). I can't think of any languages that just appeared one day and suddenly became popular.
The time that you're most awake and alert? Right before you become a parent.
About 72 hours after the arrival of my first child, I was more tired than I'd ever been in my life. Just when things were starting to improve, a few years later, along came the second.
At this point, I'm not even making the interest payments on my sleep debt, and the Sandman keeps threatening to send his goons around to beat the rest of it out of me.
I was hoping that robots would be free of original sin, but no. Well, I guess if I can't find a robot that's smart enough to vacuum my living room without getting wedged beneath the couch, I shouldn't expect to find a robot smart enough to resist forbidden fruit.
Well, let's hope all they want is a little revenge, and not their come-uppence.
Wrong... By focussing the rays, it's easy to get something hotter than the source. (although in this case, "easy" means "the mathematics are easy", non "practical") For this particular exercise:
Figure out how much energy the sun is emitting. Imagine all this energy is coming from a point source at the center of the sun.
Divide the total energy by the surface area of the sun. This gives you "heat" at the surface of the sun. Call this S.
Divide the total energy by the surface area of a sphere with its center at the center of the sun and its ratius the radius of earth's orbit. This gives you the amount of energy per unit area that is falling on your orbital fresnel lens. Call this E.
If you have a fresnel lens that has a focus ratio of greater than S/E, then the energy at the focus will be greater (per unit area) than that at the surface of the sun.
It wouldn't work as a laser per se, because the light would not be coherent. Only the light beams that are more or less parallel to the end of the fiber would find their way through the cable (although this could be increased with a few more lenses along the way).
As far as the fresnel-lens-in-space idea is concerned, it scares the crap out of me. If the thing gets knocked out alignment (maybe dinged by some of the tons of space junk already in orbit you suddenly have a death ray tracing random patterns across the planet, with no off switch (short of destroying the lens).
If they have identified these people and have some evidence, why are they waiting to act until "later next year?" The longer they're on the loose, the more chance they will have to move their operations overseas, earn money to hire better lawyers, etc. And, of course, the more spam they will inflict on us and the more it will cost *us*.
I say arrest them as soon as the prosecuting attorney is happy with the case.a judge will sign a warrant. Why wait?
My new hobby, it seems, is pointing out that this isn't news. Google makes a nice search engine, but do we really need them to tell us that spyware is bad? Sheesh.
The only interesting part of the article is:
We are alarmed by the size of this problem, which we estimate to be causing hundreds of millions of dollars to be changing hands annually. Because of this magnitude and user impact, strong action by the industry is imperative.
I think that what they're trying to say is that if money is changing hands, they want a cut.
This is a very simple case, and so it is not surprising that the contract and the code look similar. The proof of the pudding is whether this works for non-trivial specifications and algorithms.
For example, will this work with your favorite sorting algorithm? Presumably all sorting algorithms for sets drawn from a given domain will have the same pre and post conditions, but very different algorithms.
A programmer at Google forgot to put the decimal point in the right place, or forgot to divide by 1K, or something similar. As a result, absolutely nothing happened except that the quota was displayed incorrectly. BFD.
Not everything Google does is automatically interesting, or news.
Check your math: you changed bits to bytes there. 33MHz times 32bits = 1056 Mb/s = 132 MB/s.
However, it is true that this is just a theoretical maximum; there's other overhead in the PCI protocol and some implementations are worse than others. A typical PCI bus will drive a single device at about 50 MB/s (give or take). This is faster than a 100 Mb/s card can go, but not as fast as a gigabit card, so if you want to squeeze out all the performance, you'll need a wider/faster version of PCI. So your conclusion is correct, even if your math isn't.
Not quite.
Don't confuse the ethernet frame size with the TCP MSS; there's no fixed relationship between these (although of course there are combinations that make no sense). The ethernet driver is free to pack more than one TCP segment into the same ethernet frame if they're headed for the same destination, and smart drivers implement this because it boosts performance quite a bit.
Not to mention that there are other protocols like UPD that don't even attempt to negotiate the packet size...
But the worst thing is that if you try to send jumbo frames through a cheap switch, you're out of luck. The cheap switches I've tried take a look at the frame header, decide that the packet length is impossibly large, note the error and drop the packet into the bit bucket.
It's been my experience that it is not possible to use jumbo frames on a network unless EVERYTHING on that network is capable of supporting them.
At my school, it's more than 19%. In our intro CS course, we had ~200 pre-bubble, reaching 420 in 1996, then levelling off in the 300's. This year, 80.
Slashdot Mirror
I'm confused about the same thing. Can someone explain how an earthquake can possibly add stress to the tectonic plates?
Dell has sold printers for a long time. As far as I can tell, they target buyers who like to buy everything through one web site. The peripherals they sell are nothing special, and the prices aren't that good, but it's easy and convenient to buy everything with one click.
People who want the best are usually willing to shop around for it. Hopefully HP won't be run out of business if Dell is successful in undermining their market, and the next time I want a good, dependable printer I won't have to buy a re-branded Lexmark or some other similar junk.
I agree that flavor of the month representations are bad, but markup languages have been around for a long time and it wouldn't have been hard to use something (like small subset of SGML) to add a bit more formatting info. Then when people want to look at the text in the flavor-of-the-month format, it's just a matter of writing a translator for that format. (illustrations are another matter, but I suspect that things have converged enough in this area so that something could be done from this point forward.)
Don't get me wrong -- I have a huge respect for PG and what they're doing is a benefit to humanity. I just wish that they would aim a little higher than the lowest common denominator for representation, and support other character sets in a simpler manner.
Is this a typo, or is there a new meaning of "mnemonic"? The whole point of mnemonic passwords is that they're easy to remember. That's what mnemonic means.
Since I've had problems like this with my ISP, I figured it was something local. I guess not.
OK, moderate me redundant because now I see a million other people saw the same thing...
This might seem completely nuts, but in the good old USA there are many companies that use personality profiling tests, often based on debunked psychology theories, to decide who to hire and what tasks to give them. The faith that people put in these tests is astonishing.
Possessed by whom? Do you mean "he codes like he is Linda Blair in the Exorcist?" Do you really want that in the next cubicle?
Let's take a poll: would you rather be described as ungodly, or god-like?
When they come out with one that has god-like coding skills, that's when I'll get interested. Maybe.
The rest of your comment is troll-ish. If you want people to use your language, you either do a media blitz (Ada, Java, C#, Borland this and that, etc) or else you wait a long time (5-10 years) for word-of-mouth to spread (C, Perl, Python...). I can't think of any languages that just appeared one day and suddenly became popular.
About 72 hours after the arrival of my first child, I was more tired than I'd ever been in my life. Just when things were starting to improve, a few years later, along came the second.
At this point, I'm not even making the interest payments on my sleep debt, and the Sandman keeps threatening to send his goons around to beat the rest of it out of me.
Well, let's hope all they want is a little revenge, and not their come-uppence.
As far as the fresnel-lens-in-space idea is concerned, it scares the crap out of me. If the thing gets knocked out alignment (maybe dinged by some of the tons of space junk already in orbit you suddenly have a death ray tracing random patterns across the planet, with no off switch (short of destroying the lens).
I say arrest them as soon as the prosecuting attorney is happy with the case.a judge will sign a warrant. Why wait?
The only interesting part of the article is:
I think that what they're trying to say is that if money is changing hands, they want a cut.
For example, will this work with your favorite sorting algorithm? Presumably all sorting algorithms for sets drawn from a given domain will have the same pre and post conditions, but very different algorithms.
A programmer at Google forgot to put the decimal point in the right place, or forgot to divide by 1K, or something similar. As a result, absolutely nothing happened except that the quota was displayed incorrectly. BFD. Not everything Google does is automatically interesting, or news.
Cobalt servers were based on MIPS, and then migrated to AMD-K6 processors.
Not that they wouldn't have worked just fine with ARM, but as far as I can tell the idea never even came up.
However, it is true that this is just a theoretical maximum; there's other overhead in the PCI protocol and some implementations are worse than others. A typical PCI bus will drive a single device at about 50 MB/s (give or take). This is faster than a 100 Mb/s card can go, but not as fast as a gigabit card, so if you want to squeeze out all the performance, you'll need a wider/faster version of PCI. So your conclusion is correct, even if your math isn't.
Not quite. Don't confuse the ethernet frame size with the TCP MSS; there's no fixed relationship between these (although of course there are combinations that make no sense). The ethernet driver is free to pack more than one TCP segment into the same ethernet frame if they're headed for the same destination, and smart drivers implement this because it boosts performance quite a bit. Not to mention that there are other protocols like UPD that don't even attempt to negotiate the packet size... But the worst thing is that if you try to send jumbo frames through a cheap switch, you're out of luck. The cheap switches I've tried take a look at the frame header, decide that the packet length is impossibly large, note the error and drop the packet into the bit bucket. It's been my experience that it is not possible to use jumbo frames on a network unless EVERYTHING on that network is capable of supporting them.
At my school, it's more than 19%. In our intro CS course, we had ~200 pre-bubble, reaching 420 in 1996, then levelling off in the 300's. This year, 80.