1. There is no elevated trains in downtown/midtown Manhattan
I'm led to understand that it's supposed to be the 9th St el, which was actually torn down a long time ago. So maybe it's not a spacial error, but a temporal one.
2. Shots are frequently switching between a background of midtown, brooklyn, queens, and the village.
Sometimes I take those as supposed to be representing Spidey moving a lot. But yes, they're not as careful as Stan Lee was about representing New York precisely.
No, conveniently located New York City buildings. You know, those tall things that are all over Manhattan?
You'll notice that he generally will traverse the street. In other words, to go forwards, he first shoots a web to the top of a building that's in front of him and to the left. Before he smacks into a building, he shoots a web to the top of a building that's in front of him and to the right. Then left, then right. Some artists are more careless about this than others, but the movies seem to be good about that. (The video game that came out along with the first movie was careful to not let you swing as you neared the height of the tallest buildings, but in the interests of fun wasn't too picky about the precise geometry.)
The trick to this is, you need some variations to keep it up. (The rest of this web-slinging description is my own speculation.) If you start at rest, then you can only go down in the above-described manner; by the time you start an upswing, you've smacked into a building. While it's not as precise of a problem once you have a little forward momentum, it's still something to think about. You can get around this a little bit by starting a web pretty close to you in the forward direction (but still across the street and high). This lets you trade momentum to get back some height. But it's still not an easy game; you need one or two more tricks.
One handy trick is to swing on convenient out-jutting overhangings: gargoyles, horizontal flagpoles, etc. This lets you start an upswing without bleeding off as much momentum as you would if you used the across-the-street web trick.
Now, everybody who got through HS physics learned that an unpowered body can't keep this up indefinately, no matter how many geometry tricks you play. You're constantly exchanging kinetic and potential energy, but also bleeding some of that energy off as friction. You occassionally need a boost of energy. Fortunately, Spidey isn't an unpowered body. A quick yank upwards, timed right, and you can introduce a little energy into the system. If you need to stop and look around, you can also climb a nearby skyscraper to give yourself a nice big potential energy bank. Spidey traditionally can do this in a hurry by shooting a webline high and yanking hard on it (proportional strength of a spider, remember) to propel himself upwards. Or, if he's not in a hurry, good ol' wall-crawling works too.
In the original comic, does the webbing actually come from his body, or is it an invention of Peter Parkers?
Yup. Peter was a science whiz, and developed his own webbing material. It's strong, initially adhesive but quick-setting, and breaks down in a couple of hours. This is how it's been in every Spider-Man medium I've seen (lots of them), except the movies.
In the movies, they use organic web shooters. This is mostly to avoid explaining how a high school kid comes up with an adhesive that DuPont Chemicals would kill for. In the comics, it's addressed only vaguely: Peter suspects that he gained some sort of innate understanding of a spider's web when he was bitten. Even this was only discussed years after the comic began.
Spidey normally kept some extra web fluid cartridges on his belt, and sometimes would come up with specialty fluids for defeating particular foes (conductive fluid, geletainizing fluid, etc). But, being the hard-luck superhero, Spidey would inevitably run out of web fluid at the worst possible times. The "out of fluid" moments are almost a cliche of Spidey stories.
In the comics, most people-- heros, civilians, and villians-- assume that the webbing is an innate ability. I believe that he used that to fool villians once into thinking he had his powers when he didn't, but I could be wrong about that.
When 2 ethernet NIC's transmit at the same time in normal operation we don't call it jamming.
I know what you're saying, but you may want to use a different example. From the comp.dcom.lans.ethernet FAQ:
[5.3] What is jam?
When a collision is recognized by a transmitting station, a bit sequence called jam is transmitted. This jam is 32 bits long, which is long enough to traverse the entire collision domain so that all transmitting stations can detect the collision.
At one point, I assembled a computer, but plugged in the switch contacts wrong. The switch wasn't labeled, you see, and I mistook how it was put together for a sort of labeling. Like a fool, I wired it so that when I threw the switch, it shorted the 120 line to ground.
And on that day, I saw the light of ATX.
The light happened to come in the form of sparks seemingly everywhere and the power cord catching fire. Then the building power went out. Fortunately, my boss was a journeyman electrician and fixed the problem, but nobody was terribly amused by the incident.
The bundling started, I think, with IE 2 which I think got bundled with NT4.
Ah, I didn't know about that. I didn't work with NT much; we were mostly targeting home and small business.
It might also have been on some early copies of 95. It was upgraded to IE 3 pretty quickly, though (in OSR1?).
I only recall seeing one version of IE bundled with 95, but that was a long time ago and my memory isn't what it used to be, I think. I believe IE started in at OSR2. I do remember that the price on Windows went up something like $7 the day they started bundling the free web browser.
I worked in retail back then. We dealt with a lot of Microsoft products, and I saw their play many times.
First, you find a market with a clear leader. Then, you produce a knock-off, and use marketing to move eyeballs towards your product, convincing the masses that it's superior. (This is the only part that the actual product quality plays. If it sucks so bad that nobody will be fooled into thinking it's superior, then the quality needs to be better.) Finally, if it looks like the market leader will survive, then buy them out; otherwise, drop your price to something minimal and wait it out.
This was played out nowhere as clearly as Quicken. Microsoft made MS Money (which sucked terribly). MS did everything they could to make people see Money. Then they tried to buy Intuit, the makers of Quicken (but Unc' Sam put a stop to that).
Microsoft was clearly dumbfounded. Their three-step plan didn't work. What could they do? MS Money thrashed in agony for a year or two until Microsoft realized they might actually have to put some engineers into improving their product.
Not long after that, I left retail, and knocked the last dust of Microsoft products off my boots. So I don't know what's happened since then; only that every bank I've used supports MS Money downloads.
Most of us watched something similar in the browser wars, but more pronouced. Didja notice that IE was constantly improving lots, right up until IE 4? That's when they started to bundle it with the OS to get eyeballs instead of having to rely on other people who might be able to form opinions of their own. (Actually, the bundling started with IE 3 IIRC, but towards the end of its lifecycle.)
Anyway, when you think of things in those terms, then you want eyeballs. You want people thinking about MS Money as long as possible. That's your only goal. Meeting customer demand is irrelevant, so long as you don't fail by enough to lose eyeballs. And eyeballs are what marketroids know about (well, that and gin).
This entire business strategy is exactly the way for a successful monopoly in one market to expand into other markets. (Leveraging the monopolized markets, like happened with IE, is good too when you can pull it off.) It's terrible for the society, because it mutilates Adam Smith's invisible hand and leaves one finger. But it's good for the share prices.
As far as I can tell, the TRO doesn't require the customer to continue paying for service at the old rate.
No, but it doesn't require the ISP to provide service. They just can't recycle the IPs, or sabatoge the customer's migration (such as by announcing the IPs). Unless this ISP is starved for IP space, it's not a big deal to set aside a few IPs for a little while. The ISP isn't having to pay for bandwidth, or anything.
When I say the status quo, I'm only referring to the parts relevant to the dispute, specifically the IPs. The carrier service isn't under dispute, so that's gone.
So what's the big deal? Sure, the customer in question has a severe case of recto-cranial inversion. But why is everybody saying that this TRO heralds the doom of the route tables?
The judge doesn't know the technical issues, so he's issued the TRO to keep things static until he can examine everything and issue a ruling.
Note that the judge isn't insisting that the customer be able to take his numbers, just that the ISP can't prevent it. In other words, they can't BGP-advertise those numbers, or sell them to another customer, etc. The judge is just asking (okay, ordering) the ISP to set those IPs aside for the time being. If the customer can find somebody who'll advertise 'em, then that's fine too.
In a little while, the judge will have studied the situation, and gotten amicus curiae briefs, and probably expert testimony, and will issue a fair ruling (which, I expect, will tell the customer to go away and quit whining about his IPs). But for him to be fair in his ruling, he has to make sure that those IPs aren't recycled first, and that's why he issued the TRO.
The article makes it sound like the judge ruled that the IPs are portable; even the subject says it: "Can a Customer take their IP's with them? (Court says yes!)". The article talks about this as a ruling that may set a precident. It's just a TRO; the judge is putting the brakes on things until he can figure out what's what. There's no ruling, there's no precident, and I expect everything will go back to normal soon.
MIPS. It's a good baseline RISC system. There's free simulators (SPIM) and good books (Computer Organization and Design) for it. It's a straightforward and orthogonal system, so you learn about how assembly language works, rather than all the funky quirks of the architecture you're being taught on. The only thing that I might consider "quirky" is the prefetch execution slot, and that's common on RISCs anyway. You can ignore it by just sticking NOPs after branches until you're ready to learn about pipelining.
When you are ready to tackle pipelining, the orthogonality of the instruction set makes it easy to discuss. When you're ready to do some gate work, you can make a basic CPU design for it in under a week.
IMHO, you don't teach assembly so people use exactly what you teach to write programs after they graduate. Only the very worst CS programs think of things that way. You teach assembly so that the students understand computers. If you're teaching carpentry, you start by building birdhouses and tables that are easier to buy from a factory than build. Not because the student is going to build birdhouses, but because it's a good way to start teaching the basics.
If you're still wanting something you can write to a machine you own, then skip x86; it sucks for learning. The Palms run on either a Dragonball or ARM. The Dragonball is 68000-descended. While that's better than x86, it's still CISC, which can be confusing to learn (paradoxically, since CISC has stuff to make assembly theoretically easier to write). The ARM is cool, but it has its own quirks that might not translate well to other machines. PowerPC (Mac, RS/6000, GameCube, series 1 TiVo) is probably your best bet if it's widespread enough for you.
Are any of the released packages neato super keen simulations of stuff?
Read the article. It has a list and descriptions of the packages they released so far.
Where exactly IS Nasa Ames?
Exactly? I can't remember. About? In California. In the Silicon Valley. In Sunnyvale. If you're on 237 eastbound, look left just before you hit 101 (pretty much near the overpass with the red warning lightpoles next to the golf course); look for the big blimp hangers. It's nestled in with Moffet Field and Lockheed-Martin, and across the street from Juniper.
What areas of technology do they focus on?
Lots and lots of stuff, and it changes. Anything that has to do with aeronautics or space, even indirectly. This includes weather, materials research, supercomputers, health, nanotechnology, AI, to cherry-pick just a few. Look at the AMES web page for more information.
Are they part of creating the generic space probe operating system software that the Mars Rovers were saying was such a good thing?
I'm not quite sure which bit you're referring to. The rover was primarily the JPL's baby, but AMES did a lot of the mission support software. One of the coolest things I think I saw in that software development was the C Global Surveyor system.
Can ordinary Schmo's like me contribute to any projects or are they so esoteric and strange it would be useless?
Some projects are quite advanced. Some are pretty much glorified xearth's. Again, read the article.
You can have incremental backup for single files as well. Look at all our version control systems.
These rely on being able to diff the last recorded version with the version on-disk. This means having one tape drive reading your old backup, while another one writes to the new tape.
Most incremental backup systems work off of timestamps, which means all you know is that the file changed. You don't know what changed.
To put it another way, find me a real-life (commercial or free) backup suite that does this. While it may be possible in theory, I've never seen one in practice, and doubt it would go well.
Of course, I'm presently referring to non-FS-aware systems. It may be possible with some types of journalling FSs with support from the FS itself, but I'd have to research that.
What?!? What's wrong with an incremental backup? Surely all those millions of messages aren't *changing* every day?!?
That depends on how their email system works. If it stores each user in a single file, then that file is changing every day. If they're using a file-based backup system...
Because everybody sees the parens and runs like hell. "Aaaah! Its syntax is not like C's!"
Because Lisp is seen as the language of the smartest coders, and therefore inaccessible to the masses. While it is (IMHO) the language of the smartest coders, that doesn't mean the language itself is inaccessible. However, the things that the smart people can do with it may be. (Imagine explaining shell scripts to somebody who's been using command.com all their life.)
Because there's a lot of Lisp myths still floating around.
Because most programmers and PHBs don't realize that different languages make different problems easy, so they assume that everybody can use a language that is widely-known. This language is almost always a clear ALGOL descendant. As peoples' careers spread, they learn other ALGOL-descended languages, and decide that all computer languages are pretty similar. This perpetuates the problem.
Because there's still a lot of repressed resentment among PHBs who associate it with AI, which was the.com bust of the 80s. A lot of the PHB culture has forgotten the AI heyday and bust, but they haven't forgotten the prejudice against Lisp.
Because most PHBs learned a week's worth of Lisp-- often something ancient like LISP 1.5-- in college, and assume that they learned all it had to offer.
sorry but roulette with just picking the color gives your 50% odds of winning.
Bzzt, sorry, wrong, but thanks for playing.
Almost all roulette wheels in use today have 0 and 00 slots. (I say "almost", because last time I checked, there were four wheels in use-- only four-- with only a 0. But a similar discussion applies there.) These are green, not red or black. There's 18 each of the red and black slots. So betting on one of these gives you 18/38 odds, slightly less than 50%.
at least not without counting cards; which is illegal BTW
Wrong again. Counting cards in your head is legal, but will get you legally kicked out in Nevada. That's because in Nevada, you can be ejected from private property for any reason whatsoever.
In Atlantic City, the courts ruled (in somebody v. Ken Uston) that card counting is legal.
I'm getting 28k down, 23k up. Did you remember to cap your upload speed? That's a common problem with bittorrent on asymmetric links; you've got to cap your u/l speed or your downloads will suck.
You can also use bandwidth shaping to make sure that the outbound ACKs are getting enough of a chunk of the pipe. That's what I do here. It keeps my web and ssh at good speeds, and gets the ACKs out so my d/l doesn't suck.
Are there any geeks on expensive synchronous connections that can keep a tracker up long enough for us to use the Torrents?
I'm having no trouble with the tracker. I've got the download running as we speak. Not to sound condescending, but are you sure everything's okay on your end?
My dad has a box that's no good for any Windows past 98, and now that it's EOL'd, I'm thinking about recommending he move to Linux. (I'm a FreeBSD guy myself, but it looks like a lot of the Linux distros are more friendly to the Windows emigrant.)
All he uses the box for is email, web, and reading his digital camera.
Can anybody comment on Mandrake 10's suitability for such a user? Ease of use (particuarly when the geek is four states away)? Any anecdotes about Mandrake's support?
I'm going to install it here on a VMware box to evaluate myself, but would like some opinions from longer-term users.
Mind you, I made a point of explaining to the customer something to the effect of "I believe exactly what your saying and have no doubt [x] is defective, however, if we don't do some basic troubleshooting, the replacement will be rejected. I will keep this as brief as possible..." EUs were usually happy to follow the steps at that point...
That would have been nice. I figgered the fella didn't understand what I was saying. If I thought it was a mandatory procedure for RMAs, I'd have gone along with it earlier.
Slashdot Mirror
Stuck it under my tire and peeled out.
1. There is no elevated trains in downtown/midtown Manhattan
I'm led to understand that it's supposed to be the 9th St el, which was actually torn down a long time ago. So maybe it's not a spacial error, but a temporal one.
2. Shots are frequently switching between a background of midtown, brooklyn, queens, and the village.
Sometimes I take those as supposed to be representing Spidey moving a lot. But yes, they're not as careful as Stan Lee was about representing New York precisely.
3. There is no D'Agostinos on St. Marks
Are you now just being silly?
Conveniently located blimps?
No, conveniently located New York City buildings. You know, those tall things that are all over Manhattan?
You'll notice that he generally will traverse the street. In other words, to go forwards, he first shoots a web to the top of a building that's in front of him and to the left. Before he smacks into a building, he shoots a web to the top of a building that's in front of him and to the right. Then left, then right. Some artists are more careless about this than others, but the movies seem to be good about that. (The video game that came out along with the first movie was careful to not let you swing as you neared the height of the tallest buildings, but in the interests of fun wasn't too picky about the precise geometry.)
The trick to this is, you need some variations to keep it up. (The rest of this web-slinging description is my own speculation.) If you start at rest, then you can only go down in the above-described manner; by the time you start an upswing, you've smacked into a building. While it's not as precise of a problem once you have a little forward momentum, it's still something to think about. You can get around this a little bit by starting a web pretty close to you in the forward direction (but still across the street and high). This lets you trade momentum to get back some height. But it's still not an easy game; you need one or two more tricks.
One handy trick is to swing on convenient out-jutting overhangings: gargoyles, horizontal flagpoles, etc. This lets you start an upswing without bleeding off as much momentum as you would if you used the across-the-street web trick.
Now, everybody who got through HS physics learned that an unpowered body can't keep this up indefinately, no matter how many geometry tricks you play. You're constantly exchanging kinetic and potential energy, but also bleeding some of that energy off as friction. You occassionally need a boost of energy. Fortunately, Spidey isn't an unpowered body. A quick yank upwards, timed right, and you can introduce a little energy into the system. If you need to stop and look around, you can also climb a nearby skyscraper to give yourself a nice big potential energy bank. Spidey traditionally can do this in a hurry by shooting a webline high and yanking hard on it (proportional strength of a spider, remember) to propel himself upwards. Or, if he's not in a hurry, good ol' wall-crawling works too.
In the original comic, does the webbing actually come from his body, or is it an invention of Peter Parkers?
Yup. Peter was a science whiz, and developed his own webbing material. It's strong, initially adhesive but quick-setting, and breaks down in a couple of hours. This is how it's been in every Spider-Man medium I've seen (lots of them), except the movies.
In the movies, they use organic web shooters. This is mostly to avoid explaining how a high school kid comes up with an adhesive that DuPont Chemicals would kill for. In the comics, it's addressed only vaguely: Peter suspects that he gained some sort of innate understanding of a spider's web when he was bitten. Even this was only discussed years after the comic began.
Spidey normally kept some extra web fluid cartridges on his belt, and sometimes would come up with specialty fluids for defeating particular foes (conductive fluid, geletainizing fluid, etc). But, being the hard-luck superhero, Spidey would inevitably run out of web fluid at the worst possible times. The "out of fluid" moments are almost a cliche of Spidey stories.
In the comics, most people-- heros, civilians, and villians-- assume that the webbing is an innate ability. I believe that he used that to fool villians once into thinking he had his powers when he didn't, but I could be wrong about that.
When 2 ethernet NIC's transmit at the same time in normal operation we don't call it jamming.
I know what you're saying, but you may want to use a different example. From the comp.dcom.lans.ethernet FAQ:
After the second, I'd probably get the store to install it for me.
Unless, of course, you're buying from Fry's. In which case they may have been crushed before you got them.
At one point, I assembled a computer, but plugged in the switch contacts wrong. The switch wasn't labeled, you see, and I mistook how it was put together for a sort of labeling. Like a fool, I wired it so that when I threw the switch, it shorted the 120 line to ground.
And on that day, I saw the light of ATX.
The light happened to come in the form of sparks seemingly everywhere and the power cord catching fire. Then the building power went out. Fortunately, my boss was a journeyman electrician and fixed the problem, but nobody was terribly amused by the incident.
The bundling started, I think, with IE 2 which I think got bundled with NT4.
Ah, I didn't know about that. I didn't work with NT much; we were mostly targeting home and small business.
It might also have been on some early copies of 95. It was upgraded to IE 3 pretty quickly, though (in OSR1?).
I only recall seeing one version of IE bundled with 95, but that was a long time ago and my memory isn't what it used to be, I think. I believe IE started in at OSR2. I do remember that the price on Windows went up something like $7 the day they started bundling the free web browser.
I worked in retail back then. We dealt with a lot of Microsoft products, and I saw their play many times.
First, you find a market with a clear leader. Then, you produce a knock-off, and use marketing to move eyeballs towards your product, convincing the masses that it's superior. (This is the only part that the actual product quality plays. If it sucks so bad that nobody will be fooled into thinking it's superior, then the quality needs to be better.) Finally, if it looks like the market leader will survive, then buy them out; otherwise, drop your price to something minimal and wait it out.
This was played out nowhere as clearly as Quicken. Microsoft made MS Money (which sucked terribly). MS did everything they could to make people see Money. Then they tried to buy Intuit, the makers of Quicken (but Unc' Sam put a stop to that).
Microsoft was clearly dumbfounded. Their three-step plan didn't work. What could they do? MS Money thrashed in agony for a year or two until Microsoft realized they might actually have to put some engineers into improving their product.
Not long after that, I left retail, and knocked the last dust of Microsoft products off my boots. So I don't know what's happened since then; only that every bank I've used supports MS Money downloads.
Most of us watched something similar in the browser wars, but more pronouced. Didja notice that IE was constantly improving lots, right up until IE 4? That's when they started to bundle it with the OS to get eyeballs instead of having to rely on other people who might be able to form opinions of their own. (Actually, the bundling started with IE 3 IIRC, but towards the end of its lifecycle.)
Anyway, when you think of things in those terms, then you want eyeballs. You want people thinking about MS Money as long as possible. That's your only goal. Meeting customer demand is irrelevant, so long as you don't fail by enough to lose eyeballs. And eyeballs are what marketroids know about (well, that and gin).
This entire business strategy is exactly the way for a successful monopoly in one market to expand into other markets. (Leveraging the monopolized markets, like happened with IE, is good too when you can pull it off.) It's terrible for the society, because it mutilates Adam Smith's invisible hand and leaves one finger. But it's good for the share prices.
As far as I can tell, the TRO doesn't require the customer to continue paying for service at the old rate.
No, but it doesn't require the ISP to provide service. They just can't recycle the IPs, or sabatoge the customer's migration (such as by announcing the IPs). Unless this ISP is starved for IP space, it's not a big deal to set aside a few IPs for a little while. The ISP isn't having to pay for bandwidth, or anything.
When I say the status quo, I'm only referring to the parts relevant to the dispute, specifically the IPs. The carrier service isn't under dispute, so that's gone.
So what's the big deal? Sure, the customer in question has a severe case of recto-cranial inversion. But why is everybody saying that this TRO heralds the doom of the route tables?
The judge doesn't know the technical issues, so he's issued the TRO to keep things static until he can examine everything and issue a ruling.
Note that the judge isn't insisting that the customer be able to take his numbers, just that the ISP can't prevent it. In other words, they can't BGP-advertise those numbers, or sell them to another customer, etc. The judge is just asking (okay, ordering) the ISP to set those IPs aside for the time being. If the customer can find somebody who'll advertise 'em, then that's fine too.
In a little while, the judge will have studied the situation, and gotten amicus curiae briefs, and probably expert testimony, and will issue a fair ruling (which, I expect, will tell the customer to go away and quit whining about his IPs). But for him to be fair in his ruling, he has to make sure that those IPs aren't recycled first, and that's why he issued the TRO.
The article makes it sound like the judge ruled that the IPs are portable; even the subject says it: "Can a Customer take their IP's with them? (Court says yes!)". The article talks about this as a ruling that may set a precident. It's just a TRO; the judge is putting the brakes on things until he can figure out what's what. There's no ruling, there's no precident, and I expect everything will go back to normal soon.
Polarize the hull plating!
MIPS. It's a good baseline RISC system. There's free simulators (SPIM) and good books (Computer Organization and Design) for it. It's a straightforward and orthogonal system, so you learn about how assembly language works, rather than all the funky quirks of the architecture you're being taught on. The only thing that I might consider "quirky" is the prefetch execution slot, and that's common on RISCs anyway. You can ignore it by just sticking NOPs after branches until you're ready to learn about pipelining.
When you are ready to tackle pipelining, the orthogonality of the instruction set makes it easy to discuss. When you're ready to do some gate work, you can make a basic CPU design for it in under a week.
IMHO, you don't teach assembly so people use exactly what you teach to write programs after they graduate. Only the very worst CS programs think of things that way. You teach assembly so that the students understand computers. If you're teaching carpentry, you start by building birdhouses and tables that are easier to buy from a factory than build. Not because the student is going to build birdhouses, but because it's a good way to start teaching the basics.
If you're still wanting something you can write to a machine you own, then skip x86; it sucks for learning. The Palms run on either a Dragonball or ARM. The Dragonball is 68000-descended. While that's better than x86, it's still CISC, which can be confusing to learn (paradoxically, since CISC has stuff to make assembly theoretically easier to write). The ARM is cool, but it has its own quirks that might not translate well to other machines. PowerPC (Mac, RS/6000, GameCube, series 1 TiVo) is probably your best bet if it's widespread enough for you.
Are any of the released packages neato super keen simulations of stuff?
Read the article. It has a list and descriptions of the packages they released so far.
Where exactly IS Nasa Ames?
Exactly? I can't remember. About? In California. In the Silicon Valley. In Sunnyvale. If you're on 237 eastbound, look left just before you hit 101 (pretty much near the overpass with the red warning lightpoles next to the golf course); look for the big blimp hangers. It's nestled in with Moffet Field and Lockheed-Martin, and across the street from Juniper.
What areas of technology do they focus on?
Lots and lots of stuff, and it changes. Anything that has to do with aeronautics or space, even indirectly. This includes weather, materials research, supercomputers, health, nanotechnology, AI, to cherry-pick just a few. Look at the AMES web page for more information.
Are they part of creating the generic space probe operating system software that the Mars Rovers were saying was such a good thing?
I'm not quite sure which bit you're referring to. The rover was primarily the JPL's baby, but AMES did a lot of the mission support software. One of the coolest things I think I saw in that software development was the C Global Surveyor system.
Can ordinary Schmo's like me contribute to any projects or are they so esoteric and strange it would be useless?
Some projects are quite advanced. Some are pretty much glorified xearth's. Again, read the article.
You can have incremental backup for single files as well. Look at all our version control systems.
These rely on being able to diff the last recorded version with the version on-disk. This means having one tape drive reading your old backup, while another one writes to the new tape.
Most incremental backup systems work off of timestamps, which means all you know is that the file changed. You don't know what changed.
To put it another way, find me a real-life (commercial or free) backup suite that does this. While it may be possible in theory, I've never seen one in practice, and doubt it would go well.
Of course, I'm presently referring to non-FS-aware systems. It may be possible with some types of journalling FSs with support from the FS itself, but I'd have to research that.
What?!? What's wrong with an incremental backup? Surely all those millions of messages aren't *changing* every day?!?
That depends on how their email system works. If it stores each user in a single file, then that file is changing every day. If they're using a file-based backup system...
Because everybody sees the parens and runs like hell. "Aaaah! Its syntax is not like C's!"
Because Lisp is seen as the language of the smartest coders, and therefore inaccessible to the masses. While it is (IMHO) the language of the smartest coders, that doesn't mean the language itself is inaccessible. However, the things that the smart people can do with it may be. (Imagine explaining shell scripts to somebody who's been using command.com all their life.)
Because there's a lot of Lisp myths still floating around.
Because most programmers and PHBs don't realize that different languages make different problems easy, so they assume that everybody can use a language that is widely-known. This language is almost always a clear ALGOL descendant. As peoples' careers spread, they learn other ALGOL-descended languages, and decide that all computer languages are pretty similar. This perpetuates the problem.
Because there's still a lot of repressed resentment among PHBs who associate it with AI, which was the .com bust of the 80s. A lot of the PHB culture has forgotten the AI heyday and bust, but they haven't forgotten the prejudice against Lisp.
Because most PHBs learned a week's worth of Lisp-- often something ancient like LISP 1.5-- in college, and assume that they learned all it had to offer.
I'm not coming off as bitter, am I?
sorry but roulette with just picking the color gives your 50% odds of winning.
Bzzt, sorry, wrong, but thanks for playing.
Almost all roulette wheels in use today have 0 and 00 slots. (I say "almost", because last time I checked, there were four wheels in use-- only four-- with only a 0. But a similar discussion applies there.) These are green, not red or black. There's 18 each of the red and black slots. So betting on one of these gives you 18/38 odds, slightly less than 50%.
at least not without counting cards; which is illegal BTW
Wrong again. Counting cards in your head is legal, but will get you legally kicked out in Nevada. That's because in Nevada, you can be ejected from private property for any reason whatsoever.
In Atlantic City, the courts ruled (in somebody v. Ken Uston) that card counting is legal.
I'm getting 28k down, 23k up. Did you remember to cap your upload speed? That's a common problem with bittorrent on asymmetric links; you've got to cap your u/l speed or your downloads will suck.
You can also use bandwidth shaping to make sure that the outbound ACKs are getting enough of a chunk of the pipe. That's what I do here. It keeps my web and ssh at good speeds, and gets the ACKs out so my d/l doesn't suck.
Are there any geeks on expensive synchronous connections that can keep a tracker up long enough for us to use the Torrents?
I'm having no trouble with the tracker. I've got the download running as we speak. Not to sound condescending, but are you sure everything's okay on your end?
My dad has a box that's no good for any Windows past 98, and now that it's EOL'd, I'm thinking about recommending he move to Linux. (I'm a FreeBSD guy myself, but it looks like a lot of the Linux distros are more friendly to the Windows emigrant.)
All he uses the box for is email, web, and reading his digital camera.
Can anybody comment on Mandrake 10's suitability for such a user? Ease of use (particuarly when the geek is four states away)? Any anecdotes about Mandrake's support?
I'm going to install it here on a VMware box to evaluate myself, but would like some opinions from longer-term users.
When you install a copy of Linux, what's your support agreement?
Have you ever actually tried to get Microsoft to support their product as part of the purchase price?
Not the nerd/geek as they are in real life but rather as tv would like to show them.
Quite a statement, coming from you.
How about one of Benoit Mandelbrot?
The manufacturing process can't do fine enough detail.
They've been playing with a lot of time travel stuff.
Mind you, I made a point of explaining to the customer something to the effect of "I believe exactly what your saying and have no doubt [x] is defective, however, if we don't do some basic troubleshooting, the replacement will be rejected. I will keep this as brief as possible..." EUs were usually happy to follow the steps at that point...
That would have been nice. I figgered the fella didn't understand what I was saying. If I thought it was a mandatory procedure for RMAs, I'd have gone along with it earlier.