It doesn't matter what everyone else's experience is. This person is trying WINE and it's not working for him. He can run both Linux and Windows successfully. He can't run WINE. Ergo, there's a problem. Hint: the problem is not with the user.
Are you associated with WINE development? If you are, they should throw you out on your ear. You are hurting uptake of their project.
Well, Dungeon Keeper 1 was a great idea, but was badly flawed. DK2 fixed a great number of the problems, and was really a much better game.
Because of that, I suspect B&W2 will probably end up being what 1 should have been.... and should be quite fun. There was the core of a great game in B&W1, but they shipped it too soon.
It's the NEXT game, the Big New Thing, that will suck.:)
You are correct, but it's a little more complex than that. There were at least four separate evolutionary steps.
1. The original ramdisks... volatile, didn't survive reboots. Many people did partial or full copies from floppy on boot. Each boot took ages(5 or 6 minutes for a full copy!), but the machine ran a lot faster afterward. 2. The first recoverable ramdisk. My memory agrees with yours that it was the RRD: device. Because of how the machine allocated RAM, you could be reasonably sure the ramdisk area wouldn't be touched on a warm boot. You still had to boot from floppies, but once you'd partially booted, your scripts could mount the ramdisk and look for the existence of files. If it was populated, you could transfer boot to there. This was much faster after the first poweron in a given day, but was still slow...the process to bring up the ramdisk took probably 45 seconds to a minute. Much better than five minutes, and faster than a normal boot from floppies (about three minutes) but still slow. 3. In Workbench 1.2, Commodore added, um, maybe it was called Autoboot or Autoconfig? It was the ability to recognize and initialize storage devices very early in the boot process, much like the PC BIOS detects hard disks. It was more general than the PCs of the time... it could recognize any device that used the spec, rather than just IDE hard disks. This allowed booting from arbitrary devices. I THINK you could then boot completely from ramdisk, but I'm not 100% sure that this was yet supported in the system ROM. 4. Finally, in the 1.3 release, Commodore included a recoverable, bootable RAM drive right in the OS, the RAMB0: device. You could DEFINITELY boot from this ramdisk, whether or not you could in stage 3. This made warm reboots extremely, extremely fast.... 10 seconds tops. (Floppies were really dismal on the Amiga.)
So the GP poster may have been entirely correct in what he said. He had an A1000 and used it for a number of years before finding out about this, so the cvolution may have completed before he discovered it at all.
Thanks for jogging my memory. I was a little blurry on this in my main post.
I had been bugging my dad for months, and he finally broke down on Christmas Eve, 1985. We showed up to Computerland kinda late and they were trying to close, but they opened right up when my dad hollered "You wanna sell a computer?"
We brought home an Amiga 1000 with 512 megs of RAM, two floppies, the 1084 monitor, and a box of floppies. (The floppies cost $50 for 10!! And just 880k.. double-sided was still very rare and expensive.) Altogether I think he spent a shade over $3000. We had Deluxe Paint, Archon, and Dr. J and Larry Bird Go One on One. Wow, did we ever play a lot of Archon.
The system was incredibly complex, and learning it well was very hard, because there wasn't anyone around that knew it but me. I never did get very good at programming it. The OS was bafflingly complex for the time... I could do assembly on an Apple 2 and was completely stumped by the Amiga. Just not smart enough to get it, at least not on my own. But I got very good at USING it.
Eventually I bought an A500 of my own, and then awhile later I was offered a job at the local computer store... wow was I in heaven.:) I was about 19 when I started there... great job! I moved a heck of a lot of machines, out of a tiny corner store far away from any malls.
I remember that there was a DRAM crunch, and for awhile, the price of a 2 megabyte memory expansion on the side of the A500 cost more than $1200. (anyone griping over $50 for 512 megs should remember this!) The Amiga's main competitor, the Atari 1040ST, actually used as a major selling point that it had 1 megabyte for 'under $1000!'. Eventually, memory got more reasonable and memory expansions became commonplace.
You could do semi roll-your-own memory expansions, too... you could buy an outboard box for a couple hundred bucks, and then populate it yourself with chips. RAM chips at the time were about 3/4 of an inch long, and had either 16 or 32 pins, I forget which. You mounted them in sockets, being very careful not to bend any of the legs (pins), and to orient them all in the right direction. (not a keyed socket). It was slow and painstaking work, you had to get it just right or you'd potentially blow up (very expensive) 16k RAM chips.:)
As you mention, if you had enough RAM, from Kickstart 1.3 onward you could boot from a RAM disk. So I carefully crafted my Workbench disk to, when booted, initialize the ramdisk device and copy itself there. (there were several varieties of recoverable ram disk... I think I used the RRD: device, but I think later version of Workbench included a RAMB0: device that did the same thing.) It took several minutes for the machine to boot from floppy the first time, but from then on, if I needed to reboot (which at that time was still pretty common), I could do it in about 10 seconds. And the machine FLEW with the whole operating system in RAM... instant response. If you didn't have enough RAM to hold the entire Workbench floppy, you could load individual pieces of the OS instead. You could avoid a lot of the pain of floppies even if you had only 1MB of RAM.
The Amiga didn't truly shine until you had a hard disk and 2.5 megs of RAM on it... once you got it that far, it really sat up and started to sing. And the fact that you could expand even the very first Amigas to 8.5 megs, and hang absolutely ludicrous (for the time) amounts of storage off them, gave them incredible longevity.. this was when the 640k barrier was a very big problem in the PC world. As far as I know, one of my friends is STILL using her Amiga, and can do pretty much any of the basic web stuff... browsing, email, file transfers, what have you. (downloadable video doesn't work though, the Amiga just isn't powerful enough to do that... which is a shame, considering how video-oriented the machine was.)
Being able to extend a system that old, that far, *without changing its fundamental nature*, is truly amazing. The PC has extended itself similarly, but it did it by adding
I think 'fragile' would be a better description. It was (very!) easy to break, but if you had good software you could go for ages. It wasn't inherently untsable in and of itself, it just had no protection whatsoever against something ELSE making a mistake.
Of course, no self-respecting OS should ever be killable by a user-level program, but ya gotta realize... this was TWENTY YEARS AGO, and written in two or three years by a handful of (brilliant) programmers. Linux has a cast of hundreds and has been in development for more than ten years, and XP has a cast of thousands. It's not exactly fair to compare them on a level playing field.
The fact that you can even BEGIN to compare them is mind-boggling. The Amiga is clearly of the same pedigree as modern operating systems, limited and fragile/unstable as it is. None of the other personal computer OSes from the time can make that claim.... not even close.
I can't speak to the stability of the later machines. I think perhaps you had a harder time than some, because I did have friends with 3000s and they didn't curse their systems. I know part of it was choosing your programs and hardware very, very carefully.
I had a pretty beefy 2000, and it was quite stable for me. Yeah, I locked it up sometimes, but on very rare occasion. Windows 98 was probably about equivalent in terms of crashiness.... it DID happen, but not enough to keep me from getting work done. It never got CLOSE, however, to the stability we expect of modern OSes. Linux, OSX, and XP all make that early OS look like a toy.
In some ways, it was... but it had many of the capabilities we take for granted today, when most folks were running one text-mode program at a time in DOS on 128k machines, and being excited about TSRs.
It was much more like what you're using now than like the PCs of the era, which speaks very, very highly of it.
I have often said that had Apple been selling the Amiga, we'd all be running offshoots of that platform, rather than the PC. It was ten years ahead of its time, and I mean that almost literally; it wasn't until 1994, running Linux, that I could get even CLOSE to what I could do on my Amiga in 1985.
In looking back, it is amazing the number of things they absolutely nailed wih the Amiga. It was the first machine to use fully-programmable custom chips for sound and graphics support. That hardware was immensely powerful; it could do memory copies (the blitter), palette shifts (the "copper", and I don't remember why it was called that anymore), sprites, collision detection, four-channel stereo sound, and probably many other things I'm forgetting, without even using the main CPU at all. (well, except to set things up, at any rate.)
The system could display separate programs with separate resolutions and color palettes on the same screen at the same time. You could literally grab the Workbench screen and drag it down, revealing some cool demo running behind it..."Boing!" in the top half, Workbench in the bottom. This was done by some clever copper tricks... on the fly, over the space of about two scanlines, the copper would shift the entire display mode and palette, and start displaying screen data from a different arbitrary program.
Later, a variant on this technique was used to create the best graphics the Amiga could manage...Sliced HAM, or S-HAM. The default 'high color' graphics mode, HAM, could have any 32 base colors out of the palette of 4,096. Any pixel could either have one of the base colors, or it could H)old the color of the previous pixel A)nd M)odify either the red, green, or blue component. S-HAM took this a step further, and swapped the base 32 colors on *every scan line*, so that you could have many more colors available. Some of the S-HAM pictures were absolutely stunning. It did, however, put a huge load on the graphics hardware... the machine really crawled when running that mode. So it was really only useful for slideshows... you couldn't animate that mode, to my knowledge.
Then, on top of that, they mostly nailed the OS. There were three major components to the AmigaOS; Exec, Intuition, and AmigaDOS.
Exec was the multitasking core, what we'd probably think of as the kernel in Linux land. It was immensely efficient. The task switching method that RJ Mical came up with was so fast that it ended up going into the Motorola programming manuals. I can't find the numbers offhand, but I believe the Amiga could task switch in less than twenty clock cycles. Whatever the actual number was, it was FAST.
Intuition provided the windowing libraries; it was what kept windows properly layered and coordinated, and routed user input. That would be roughly the equivalent of X, though much simpler. Workbench, the built-in graphic UI, was an optional load; you could stay in 'console mode' if you wished. The Amiga had no true text-only mode, however. Even if you had just a single CLI window open with nothing else, it was still drawn in graphics mode. (scrolling on the Amiga was never very fast because of this).
AmigaDOS, I believe, did all the disk and file I/O. It was rather Unixish, but it was very slow and had an absolutely horrible user interface. (Fortunately, it was easy to replace the DOS programs with better ones, and most people who really used their machines did so.) Filesystems were abstracted too, which was a good thing.... the early filesystem on the Amiga was very fragile and very slow. Later on, the Fast File System was introduced, which sped things up a heck of a lot. With FFS, hard drives were quite comfortable, but floppies were never very good. There were many special custom loaders that sped things up (much like on the C64), but the floppies were always slow, no matter what.
Of the three major components of the OS, AmigaDOS was the weakest, and was responsible for a lot of the early (justified) griping about the pl
One possibility would be to use a separate, dedicated local network, possibly just one machine sitting by its lonesome. That machine should still be firewalled, especially outbound, and your local network shouldn't trust it at all. Ideally, you'd set it to only be able to reach the networks of whoever you're collaborating with, but if that's likely to change frequently, you may have to open it up to the world.
If your hospital provides network services to the outside world, it's likely that your IT group will already have implemented the separate-network idea... it's generally called a 'DMZ'. If they already have one, it should be easy to run a wire to a certain, specific machine, ideally one in a conference room. And they'll want to lock down the port so it will only talk to that machine, so that visitors or staff don't try to use the DMZ net by mistake. (Some outfits treat the DMZ as already compromised, so if you connect to it by mistake, your machine is declared untrusted, wiped, and reloaded. Not a lot of fun for anyone involved. Most places aren't that anal, though perhaps they should be.)
Setting up a machine in an existing DMZ will involve some work on the part of the IT crew. If they're creating a DMZ from scratch, it could potentially be quite a bit of work. If they're jammed, they may not have any brainpower free to help you with it.
If they can't help you, but you have some budget and live in a big city, you could potentially hook up via a laptop and one of the high-speed cell data networks. In many areas of the country, this is fast enough to be perfectly functional, and you shouldn't need much help getting it running.
However, if you do this without IT's blessing, you should NEVER AGAIN connect that machine to their network... with that data card in it, that machine can potentially be a back door straight into their systems. If you have any VPN software on that machine with credentials to connect to your network, remove it. Before you plug in the cell card, turn it completely into a standalone system, and make sure that all the passwords are different than any others used on any of your other networks, including your home machines. You are assuming that it WILL be compromised, and you're doing your level best to make sure that there's no data on the machine you care about, nor any sensitive passwords that could give an attacker a wedge into your corporate network.
Remember, even if you pull the card out, the machine could already be compromised, so connecting it to the internal network could let it 'phone home' to a hacker and give him full access to your systems. So it should never again be connected to the work network in any way, shape, or form. You should always think of it as already compromised.... dangerous and radioactive. This may or may not be actually TRUE, but it's not a bad habit to be in, especially when you're not an expert.
Disable the Ethernet port so you remember that you don't trust that machine. If you need to move data, burn a CD and sneakernet it. Turn the machine off when you're not using it.
If you don't have the technical chops to lock it down well, wipe and reload it on a fairly regular basis. Use new passwords every time.
I don't recommend you do this for a whole bunch of reasons (not the least of which is the potential of getting fired), but if you can't get IT to help you, setting up a cell card like this should be pretty safe. Just make SURE that it knows NOTHING about your internal net before hooking it up.
Even if IT is otherwise busy, they might be able to give you a 'clean' machine that you can do whatever you want with.
I read an article by a knowledgeable-seeming science-fiction author a few years back, about antimatter drives.
The first interesting thing I got out of the article was that any rocket would carry about the same amount of reaction mass for a trip of any distance. Whether going to the Moon or going to Jupiter, it would carry a fixed amount of some cheaply available gas. What would alter would be how much antimatter would be mixed into the propellant.. for a Moon trip you might use a hundredth as much as you would go to go Jupiter. By mixing more antimatter in, they'd increase the boost without needing more (or less) inert propellant.
He didn't address your issue directly, but one thing he pointed out is that both antimatter and fusion reactions produce gamma rays in abundance. If I understood him correctly, gamma rays are impossible to block completely. They can be attenuated, but there is no substance known that will completely block them. Some will always get through. You could have a mass of solid lead the size of the Earth and some rays would still penetrate.
This means that astronauts would be taking a heck of a dose from their engines, no matter what shielding was used. Add in the radiation from the Sun and astronauts probably wouldn't last long out in space, at least in these parts.
Unless we discover some way to reduce or eliminate radiation damage, it's not likely we'll be sending manned ships anywhere using either of these technologies. They could be perfectly fine for unmanned probes, but barring major medical advances, human evolution, or some kind of radiation shielding technology, humans aren't going far on antimatter drives.
One of the things I like to think about are the limitations of rationality. Rational thinking involves ignoring most of what's true about something, in order to deal with the aspect in which we're interested. When we're thinking about a chair as something to sit on, we're not really thinking about it as kindling, or as potential building supplies for a chair pyramid, nor the possibility of using one of the chair legs as a club. We tend to think about just one property of an object at any given time, which (temporarily) blinds us to its other properties.
I really wonder if the whole time travel metaphor thing isn't an artifact of bad modeling, rather than anything true. Time is often claimed to be 'a dimension'. A one-dimensional object is a line, and a two-dimensional object is a plane. Three dimensions is real life, what we see, and the fourth dimension, ergo, must be time.
This is all very pretty, but there's one big problem... there are no one- or two-dimensional objects. Nothing in real life has less than three dimensions. Never, never, not ever. One and two dimensional objects are imaginary, they do not and can never exist. Dimensions are a highly useful mental model, but since there are NO objects with less than three dimensions, there's a good chance they're not actually true in any meaningful sense.
Like most rational thought, dealing with space in terms of wholly imaginary dimensions allows us to do some really amazing stuff. But it also blinds us to most other things that are true about the universe while we're thinking that way.
I very strongly suspect that time is not, in fact, a dimension, and that all the mathematical hand-wringing that goes on about time travel is trying to make an impossible thing work out of a false, imaginary premise. There are no one- and two-dimensional objects. Objects always have height, depth, and width. Always.
If the first and the second dimensions are imaginary, it seems entirely reasonable to assume that the fourth must also be imaginary. We like to count up, and having invented three dimensions, we may be looking for a fourth. It looks to me like we're confusing our map for the territory.
I strongly suspect that ultimately we'll find out that time is a force or a process, and that the model of it being a line we're traveling along is false.
Your business obviously thought this software was valuable enough to spend two million bucks making it. It's a competitive advantage; you have it, and the other people in your industry don't.
If you open source your software, then your competitors get it too. But instead of spending two million bucks, they get it for free, so they have money to spend on OTHER stuff that YOUR company doesn't have. You are weakening yourselves considerably, hoping to get a payoff of equal or greater value.
So you'd need outside code contributed that would be worth at least another couple million. The chances of that happening are laughably small. Unless your software addresses a very broad horizontal market (at least tens and probably hundreds of thousands of possible installations), there's no way you'd ever get two million bucks' worth of patches.
The LAMP programs, Linux, Apache, mysql, and Perl, have probably gotten that level of free time donated. But there are very few others, and you most likely don't have 'the new Apache' on your hands. Let's charitably put that chance at 'tiny'.
Not all code should be or needs to be free. You would definitely be doing your customers a favor if you included the source code with the product, with strong restrictions on what they can do with it. But just releasing it into the wild directly and immediately harms you a great deal, and has only a small chance of paying off.
From the very rudimentary data provided, it looks like open sourcing your product is almost exactly the worst thing you could do.
For a long time, Intel et al have been shipping processors that emulate the X86 instruction set. They don't actually use it themselves; they read an X86 bytecode instruction stream and translate it, on the fly, to their native micro-ops, whatever they might be. They use all kinds of tricks to allow more X86 instructions to be executed simultaneously, stuff so complex it's mind numbing.
But from YOUR perspective as a programmer, you're still using X86 and it's just as brain-dead as it ever was. (X86-64 is a little better.) You don't have any access to the native, 'real' instruction set, so the fact that modern chips are just hardware emulators of X86 doesn't matter. From your perspective, it's the same old crap, just a lot faster.
It's really a shame that this ugly architecture has gotten so far and so completely dominated everything. But, in fairness, very few programmers are working down that low anymore, and compilers hide the worst of the nastiness.
Still can't help but think that if all that effort had gone into accelerating a cleaner architecture to begin with, we'd have come farther and faster.
It's a very weird idea... it feels a bit like waking up in the Twilight Zone or something.
I forget the name offhand, but there was an early computer company, possibly Amstrad, that preannounced that they were coming out with this really amazing new machine. Customers loved the idea, and stopped buying all the existing machines. Sales dropped to 0 and they died before they could get the new hotness out the door.
Apple has always known about this, and has been very aggressive about not preannouncing ANYTHING, so as to not hurt sales of existing products. This violates that precept so badly that I can't help but think it's prompted by panic (or anger, as some have hypothesized). I'm not sure whether or not it's a bad idea. Their existing sales are going to drop, perhaps precipitously, while they try to figure out how to make OSX on Intel work. But if they didn't preannounce, then the developers wouldn't have time to get ready for the transition, which could potentially be worse.
As an aside, PPC emulation is never going to be very good, and all that specially tuned Altivec code has just been junked. That's a huge investment down the drain.
They have to know that this is a bet-the-business move. They'll have to execute nearly perfectly, and very quickly, to make this happen without pissing off too many people. And they're going to have to continue to execute flawlessly for a long while. Apple's good at that, but this is asking an awful lot of their engineers. They must really believe their backs are against the wall.
I wish them well, but there are soooo many ways they could screw this up.
I doubt they'll get to full human-brain awareness level anytime soon, but... what if they do? What happens if they create a sentient being inside their simulator? When they're done with the simulation and it's time to start on something new, is turning off the machine killing the 'creature' inside?
And even if it's not as smart as a human, what then? What ethical guidelines are appropriate? When is it okay to destroy a thinking being, even if you created it yourself? And how complex must it be? Killing a beagle or a dolphin isn't murder, after all, but it's still considered wrong in many cases to do so.
Are AIs cute and cuddly and protected by humane-treatment laws, or scary and kill-on-sight, like spiders and snakes are for many people?
How smart does an AI have to be to have rights against termination?
We've been sort of doodling around with these thoughts for a long time, but it's getting to the point where we may actually need the answers.....
My mom is out in the boonies. She's been on Starband for a number of years, and has spent a ridiculous amount of money on equipment.... her cable modem has melted down twice.
This last time, it stopped working completely, even to the point of talking to the computer at all. So I called Starband. It took two hours on hold before I could talk to someone. This was actually two calls, because I called, waited an hour, and then the system hung up on me. The second call took another hour and I finally got rhrough to a technician. He said "oh, sorry, your service is provided by Dish Network, you'll have to talk to them."
So I called Dish, who answered within a few minutes... but they couldn't help me. "Sorry, you'll have to talk to Starband, we don't know anything about the hardware." I had him ask his supervisor and the supervisor confirmed it.. I had to talk to Starband.
So I called Starband back. It took about 90 minutes... they had the good grace to disconnect me after only 30 minutes of waiting on my first call, so I got started on my second hour a little sooner. I explained my problem to the guy with the strong Indian accent. He said "You have to talk to Dish about that." And I said "No, I can't, it's hardware and you guys have to do hardware." He said back, "You have to talk to Dish about that." And then he hung up on me.
And note that this is the SECOND $1000 modem she's bought from them. I'm not sure she paid the full $1k for the second, but it was still a lot, over $500 for eure.
Dish Network and Starband are THE WORST network providers I have EVER seen. You are better off with carrier pigeon protocol. DO NOT sign up with these people. They define customer hostility. Their network is atrocious and their customer service... well, see the above paragraph. This company deserves to be nuked from orbit. You've gotta realize... even when the net was WORKING, it was only for a very, very loose definition of working. Starband would be a perfect synonym on Wikipedia for 'incompetence'.
So I went back in and told my mother she was changing providers. We've had various misadventures before she found one she liked. She's in a fairly remote area, but has reasonable wireless coverage, so we figured we'd try setting her up that way. (she's about 1.25 hours NE out of Atlanta, Georgia.)
In that area, Verizon's service was terrible, nearly as bad as Starband's. My guess was that the cell network was okay, but that it was grossly underprovisioned at the cell tower(s). At 4AM I got pretty good download speeds. At 4pm the network was essentially unusable. It appears Verizon is willing to accept very, very poor levels of service quality while still happily signing people up. If you go with Verizon, MAKE SURE you have a money-back guarantee. Their network was no better than Starband's, and that's as profound a statement of dissatisfaction as I can make.
Next she tried Suncom, which is the local AT&T affiliate. Their service plan is very reasonable. It's almost exactly like a 56k modem, and quite reliable. You won't get your pages quickly, but you WILL get them, which is more than I can say for either Starband or Verizon. Suncom's pricing is very good too.. the minimum each month is $5, and the maximum is $50. So if you're not using it, you don't get soaked, and the most you can pay for any given month is $50. Suncom has very good, customer-oriented price plans, and if you're in the South, they're a good choice.
Unfortunately, my mother kept roaming into Cingular space, even though she was just sitting in her armchair in her living room. Suncom tried to bill her several hundred dollars for roaming, backed off on that, but told her she couldn't use it, in the future, when the network said Cingular. It was doing that more and more, so she just cancelled/returned that service and tried Cingular's flavor.
In her area, Cingular appears to be quite good. I haven't had a chance to try it myself yet, but she says i
Dude, you're not thinking. That isn't my argument at all. I don't know where the hell your 'prove a 50% probability' crap came from.
No matter HOW improbable intelligent life is, it has happened at least once. Therefore, it is possible. In the vastness of the Universe, asserting that it hasn't happened at least twice is STUPID. No matter how small the chance for life on any given planet, the sheer vastness of the Universe essentially guarantees we are not the only intelligent life. The probability isn't 50%, it's 99.99999... %, going to a very large number of nines.
But the distances are so great -- the Universe is so large -- that if the chance is small, each race will probably go extinct without ever seeing another intelligent species. Only if complex life is exceedingly common will we be at all likely to encounter alien intelligence.
Well, there are two pretty compelling bits of evidence.
1) Intelligence, such as it is, has evolved at least once. 2) The Universe is unimaginably huge. Just our own galaxy is vast beyond the ability of humans to even imagine. One of the early Hubble Deep Field studies, looking at one of the darkest places in the sky, saw 40,000 GALAXIES in an area of the sky equivalent to a grain of sand held at arm's length.
Given those two facts, doubting alien intelligence strikes me as profoundly stupid. However, unless it is extremely common (which I doubt), the chances of any of that intelligence being within a distance we could detect is pretty darn small.
The probability of alien intelligence, in other words, is essentially indistinguishable from 1. Given the constraints of lightspeed, however, the chance that we could ever meet and TALK TO such aliens is probably very close to 0.
As I understand it, true randomness only comes from measuring effects at the quantum level, like radioactive decay. (mentioned in other threads here). As nearly as we can determine, individual quantum events are absolutely random and completely unpredictable. We can make fairly precise predictions about groups of events, but we can't predetermine when, say, a given atom will decay. We can tell you about how many will decay in any given second, and this prediction will become more and more precise as events accumulate over time, but we can't tell ANYTHING about an individual event until after it happens. True random number generators depend on this.
A step up from that are events derived from things believed to be random, like user input. However, they are always mediated by other factors, like the circuitry and response time in the keyboard or mouse, for instance. This is probably pseudo-random data at least some of the time. The numbers are still pretty random, but in theory, skilled cryptographers might be able to tease patterns out of the bitstreams. This could potentially allow them to successfully mount mathematical attacks on encrypted data.
Another method of generating numbers is with an algorithm... given a seed number, it will produce a stream of 'random' numbers. If the seed and the algorithm are known, many of these streams can be cracked. Because they aren't really random, but have many of the features of true random numbers, they're called pseudo-random. With a strong enough algorithm, at least in theory, encryption based on pseudo-random numbers should be just about as difficult to crack as encryption based on REAL random numbers.
But even if we can't easily detect it, any number stream generated by an algorithm DOES have a pattern. We may not be smart enough to find it yet, but there's a good chance we may someday be that smart (or have that much computing power). Encryption based on truly random numbers should be crackable only by brute force; no analysis should ever reveal a pattern, since by definition there will be no pattern to find. (If we ever DID find consistent patterns in true random numbers, this would shake our understanding of the universe right down to the foundations... just about to the point of having to start over from scratch.)
Also note that a number stream generated by a Random Number Generator(RNG) can't be *too* random... if the chaos is too perfect, it becomes order, and is attackable. (weird, eh?)
I'm neither a cryptographer nor a mathematician. This is how I understand things to be, but I'm just a layperson and could easily be wrong. Pay attention to replies.
You gotta realize, that even by Empire standards, the Death Star is unimaginably huge. It would be a titanic feat of engineering to build it. Consider: the Earth has a huge amount of iron in it... if the Death Star were made out of iron, it would require stripping away like a SIXTH of the Earth's ENTIRE MASS to build it. (those numbers are, admittedly, pulled out of thin air and are guesses, but I think they're in the right size range).
Now, they'd probably go get asteroids and use those, and they'd probably have an enormous assembly fleet, but it would still be an unbelievable amount of work. Doing it in just 20 years would strain even a Galactic Empire. And having the skeleton built is about 0.01% of the whole job. Consider houses... they'll have the frame up in a couple days, but it still takes six months to build. Compare that to the enormous internal volume of a Death Star versus its relatively small surface area.
The fact that they had a second largely built in just a few more years strikes me as far less believable than the long timeframe to do the first. They would almost have to have started it halfway through the first project.
Rome wasn't built in a day, and I very much doubt the Death Star was built in a decade.
Oh, absolutely. I'm not a big Microsoft fan by any means, and they do it too. But I'm strongly suspicious that the PS3 may not actually deliver what they claim... given today's technology, the specs they're quoting are going cost them like $750/unit. I don't think they're going to want to ship that thing at $300.
Basically, I'll believe it when they actually ship it -- in both cases. One of the mainstays of FUD is to announce your own much better product that won't come out for a long time, to try to distract momentum from the product you can actually buy TODAY.
Admittedly, in neither case can you actually buy the consoles in question, but Sony is going to be late to the party... so even if they don't ship what they promise to, or if it doesn't work as well as they claim, they might believe that the backlash would be less damaging to them than letting the 360 get a solid foothold in the market.
I personally find it hard to believe that they can ship that much hardware at a resaonable price point. They've been known to overstate future products before. And the XBox appears more likely to be able to hit $299, which I think is critical to any kind of mass-market acceptance.
And then, of course, you have Nintendo, whose claim to fame appears to be "Our console is much smaller." Yeah, that's a big selling point, for sure.:-)
Good god, how many times does this have to be said? Free as in beer is great. Free as in libre is better. If I use free-beer code and become dependent on it, then I am giving the person who wrote it some degree of control over my life. And I may not really realize the consequences of doing so until much later. See Linus Torvalds and Bitkeeper. See also Microsoft, who likes to impose new license changes unilaterally in service packs, or modify file-save formats to force you to buy another multi-hundred-dollar upgrade to your office suite.
People are upset because Sun can change the terms of Java whenever they choose. They have not been willing to really give up control of Java. They desperately want to screw people the way Microsoft gets to. They don't seem to realize that nobody else will EVER be able to print money like Microsoft does, at least not in the computer field. Between Microsoft already existing, and Open Source coming up from underneath, the best Sun can hope for is an honest living... they're not going to be able to both maintain control over their platform and also have it be a market-determining force. They can have one or the other, but not both. Only Microsoft gets both.
People are uptight because using Java-based OO means making themselves vulnerable to Sun, in the same way that Word users are vulnerable to Microsoft. They want the assurance that they can't be orphaned, forced to upgrade, or forced to accept some other unilateral license change on Sun's part.
I don't think I've ever heard ANYONE argue that OO is actually better than the Office suite. People use it because it is cheap and free-as-in-libre, and get most annoyed when the libre part starts going away. The few serious users they do have exist BECAUSE of Microsoft's abuse, and those customers are highly snsitive to behavior that looks even vaguely similar.
Slashdot Mirror
It doesn't matter what everyone else's experience is. This person is trying WINE and it's not working for him. He can run both Linux and Windows successfully. He can't run WINE. Ergo, there's a problem. Hint: the problem is not with the user.
Are you associated with WINE development? If you are, they should throw you out on your ear. You are hurting uptake of their project.
Well, Dungeon Keeper 1 was a great idea, but was badly flawed. DK2 fixed a great number of the problems, and was really a much better game.
:)
Because of that, I suspect B&W2 will probably end up being what 1 should have been.... and should be quite fun. There was the core of a great game in B&W1, but they shipped it too soon.
It's the NEXT game, the Big New Thing, that will suck.
You are correct, but it's a little more complex than that. There were at least four separate evolutionary steps.
1. The original ramdisks... volatile, didn't survive reboots. Many people did partial or full copies from floppy on boot. Each boot took ages(5 or 6 minutes for a full copy!), but the machine ran a lot faster afterward.
2. The first recoverable ramdisk. My memory agrees with yours that it was the RRD: device. Because of how the machine allocated RAM, you could be reasonably sure the ramdisk area wouldn't be touched on a warm boot. You still had to boot from floppies, but once you'd partially booted, your scripts could mount the ramdisk and look for the existence of files. If it was populated, you could transfer boot to there. This was much faster after the first poweron in a given day, but was still slow...the process to bring up the ramdisk took probably 45 seconds to a minute. Much better than five minutes, and faster than a normal boot from floppies (about three minutes) but still slow.
3. In Workbench 1.2, Commodore added, um, maybe it was called Autoboot or Autoconfig? It was the ability to recognize and initialize storage devices very early in the boot process, much like the PC BIOS detects hard disks. It was more general than the PCs of the time... it could recognize any device that used the spec, rather than just IDE hard disks. This allowed booting from arbitrary devices. I THINK you could then boot completely from ramdisk, but I'm not 100% sure that this was yet supported in the system ROM.
4. Finally, in the 1.3 release, Commodore included a recoverable, bootable RAM drive right in the OS, the RAMB0: device. You could DEFINITELY boot from this ramdisk, whether or not you could in stage 3. This made warm reboots extremely, extremely fast.... 10 seconds tops. (Floppies were really dismal on the Amiga.)
So the GP poster may have been entirely correct in what he said. He had an A1000 and used it for a number of years before finding out about this, so the cvolution may have completed before he discovered it at all.
Thanks for jogging my memory. I was a little blurry on this in my main post.
I had been bugging my dad for months, and he finally broke down on Christmas Eve, 1985. We showed up to Computerland kinda late and they were trying to close, but they opened right up when my dad hollered "You wanna sell a computer?"
:) I was about 19 when I started there... great job! I moved a heck of a lot of machines, out of a tiny corner store far away from any malls.
:)
We brought home an Amiga 1000 with 512 megs of RAM, two floppies, the 1084 monitor, and a box of floppies. (The floppies cost $50 for 10!! And just 880k.. double-sided was still very rare and expensive.) Altogether I think he spent a shade over $3000. We had Deluxe Paint, Archon, and Dr. J and Larry Bird Go One on One. Wow, did we ever play a lot of Archon.
The system was incredibly complex, and learning it well was very hard, because there wasn't anyone around that knew it but me. I never did get very good at programming it. The OS was bafflingly complex for the time... I could do assembly on an Apple 2 and was completely stumped by the Amiga. Just not smart enough to get it, at least not on my own. But I got very good at USING it.
Eventually I bought an A500 of my own, and then awhile later I was offered a job at the local computer store... wow was I in heaven.
I remember that there was a DRAM crunch, and for awhile, the price of a 2 megabyte memory expansion on the side of the A500 cost more than $1200. (anyone griping over $50 for 512 megs should remember this!) The Amiga's main competitor, the Atari 1040ST, actually used as a major selling point that it had 1 megabyte for 'under $1000!'. Eventually, memory got more reasonable and memory expansions became commonplace.
You could do semi roll-your-own memory expansions, too... you could buy an outboard box for a couple hundred bucks, and then populate it yourself with chips. RAM chips at the time were about 3/4 of an inch long, and had either 16 or 32 pins, I forget which. You mounted them in sockets, being very careful not to bend any of the legs (pins), and to orient them all in the right direction. (not a keyed socket). It was slow and painstaking work, you had to get it just right or you'd potentially blow up (very expensive) 16k RAM chips.
As you mention, if you had enough RAM, from Kickstart 1.3 onward you could boot from a RAM disk. So I carefully crafted my Workbench disk to, when booted, initialize the ramdisk device and copy itself there. (there were several varieties of recoverable ram disk... I think I used the RRD: device, but I think later version of Workbench included a RAMB0: device that did the same thing.) It took several minutes for the machine to boot from floppy the first time, but from then on, if I needed to reboot (which at that time was still pretty common), I could do it in about 10 seconds. And the machine FLEW with the whole operating system in RAM... instant response. If you didn't have enough RAM to hold the entire Workbench floppy, you could load individual pieces of the OS instead. You could avoid a lot of the pain of floppies even if you had only 1MB of RAM.
The Amiga didn't truly shine until you had a hard disk and 2.5 megs of RAM on it... once you got it that far, it really sat up and started to sing. And the fact that you could expand even the very first Amigas to 8.5 megs, and hang absolutely ludicrous (for the time) amounts of storage off them, gave them incredible longevity.. this was when the 640k barrier was a very big problem in the PC world. As far as I know, one of my friends is STILL using her Amiga, and can do pretty much any of the basic web stuff... browsing, email, file transfers, what have you. (downloadable video doesn't work though, the Amiga just isn't powerful enough to do that... which is a shame, considering how video-oriented the machine was.)
Being able to extend a system that old, that far, *without changing its fundamental nature*, is truly amazing. The PC has extended itself similarly, but it did it by adding
I think 'fragile' would be a better description. It was (very!) easy to break, but if you had good software you could go for ages. It wasn't inherently untsable in and of itself, it just had no protection whatsoever against something ELSE making a mistake.
Of course, no self-respecting OS should ever be killable by a user-level program, but ya gotta realize... this was TWENTY YEARS AGO, and written in two or three years by a handful of (brilliant) programmers. Linux has a cast of hundreds and has been in development for more than ten years, and XP has a cast of thousands. It's not exactly fair to compare them on a level playing field.
The fact that you can even BEGIN to compare them is mind-boggling. The Amiga is clearly of the same pedigree as modern operating systems, limited and fragile/unstable as it is. None of the other personal computer OSes from the time can make that claim.... not even close.
I can't speak to the stability of the later machines. I think perhaps you had a harder time than some, because I did have friends with 3000s and they didn't curse their systems. I know part of it was choosing your programs and hardware very, very carefully.
I had a pretty beefy 2000, and it was quite stable for me. Yeah, I locked it up sometimes, but on very rare occasion. Windows 98 was probably about equivalent in terms of crashiness.... it DID happen, but not enough to keep me from getting work done. It never got CLOSE, however, to the stability we expect of modern OSes. Linux, OSX, and XP all make that early OS look like a toy.
In some ways, it was... but it had many of the capabilities we take for granted today, when most folks were running one text-mode program at a time in DOS on 128k machines, and being excited about TSRs.
It was much more like what you're using now than like the PCs of the era, which speaks very, very highly of it.
I have often said that had Apple been selling the Amiga, we'd all be running offshoots of that platform, rather than the PC. It was ten years ahead of its time, and I mean that almost literally; it wasn't until 1994, running Linux, that I could get even CLOSE to what I could do on my Amiga in 1985.
In looking back, it is amazing the number of things they absolutely nailed wih the Amiga. It was the first machine to use fully-programmable custom chips for sound and graphics support. That hardware was immensely powerful; it could do memory copies (the blitter), palette shifts (the "copper", and I don't remember why it was called that anymore), sprites, collision detection, four-channel stereo sound, and probably many other things I'm forgetting, without even using the main CPU at all. (well, except to set things up, at any rate.)
The system could display separate programs with separate resolutions and color palettes on the same screen at the same time. You could literally grab the Workbench screen and drag it down, revealing some cool demo running behind it..."Boing!" in the top half, Workbench in the bottom. This was done by some clever copper tricks... on the fly, over the space of about two scanlines, the copper would shift the entire display mode and palette, and start displaying screen data from a different arbitrary program.
Later, a variant on this technique was used to create the best graphics the Amiga could manage...Sliced HAM, or S-HAM. The default 'high color' graphics mode, HAM, could have any 32 base colors out of the palette of 4,096. Any pixel could either have one of the base colors, or it could H)old the color of the previous pixel A)nd M)odify either the red, green, or blue component. S-HAM took this a step further, and swapped the base 32 colors on *every scan line*, so that you could have many more colors available. Some of the S-HAM pictures were absolutely stunning. It did, however, put a huge load on the graphics hardware... the machine really crawled when running that mode. So it was really only useful for slideshows... you couldn't animate that mode, to my knowledge.
Then, on top of that, they mostly nailed the OS. There were three major components to the AmigaOS; Exec, Intuition, and AmigaDOS.
Exec was the multitasking core, what we'd probably think of as the kernel in Linux land. It was immensely efficient. The task switching method that RJ Mical came up with was so fast that it ended up going into the Motorola programming manuals. I can't find the numbers offhand, but I believe the Amiga could task switch in less than twenty clock cycles. Whatever the actual number was, it was FAST.
Intuition provided the windowing libraries; it was what kept windows properly layered and coordinated, and routed user input. That would be roughly the equivalent of X, though much simpler. Workbench, the built-in graphic UI, was an optional load; you could stay in 'console mode' if you wished. The Amiga had no true text-only mode, however. Even if you had just a single CLI window open with nothing else, it was still drawn in graphics mode. (scrolling on the Amiga was never very fast because of this).
AmigaDOS, I believe, did all the disk and file I/O. It was rather Unixish, but it was very slow and had an absolutely horrible user interface. (Fortunately, it was easy to replace the DOS programs with better ones, and most people who really used their machines did so.) Filesystems were abstracted too, which was a good thing.... the early filesystem on the Amiga was very fragile and very slow. Later on, the Fast File System was introduced, which sped things up a heck of a lot. With FFS, hard drives were quite comfortable, but floppies were never very good. There were many special custom loaders that sped things up (much like on the C64), but the floppies were always slow, no matter what.
Of the three major components of the OS, AmigaDOS was the weakest, and was responsible for a lot of the early (justified) griping about the pl
One possibility would be to use a separate, dedicated local network, possibly just one machine sitting by its lonesome. That machine should still be firewalled, especially outbound, and your local network shouldn't trust it at all. Ideally, you'd set it to only be able to reach the networks of whoever you're collaborating with, but if that's likely to change frequently, you may have to open it up to the world.
If your hospital provides network services to the outside world, it's likely that your IT group will already have implemented the separate-network idea... it's generally called a 'DMZ'. If they already have one, it should be easy to run a wire to a certain, specific machine, ideally one in a conference room. And they'll want to lock down the port so it will only talk to that machine, so that visitors or staff don't try to use the DMZ net by mistake. (Some outfits treat the DMZ as already compromised, so if you connect to it by mistake, your machine is declared untrusted, wiped, and reloaded. Not a lot of fun for anyone involved. Most places aren't that anal, though perhaps they should be.)
Setting up a machine in an existing DMZ will involve some work on the part of the IT crew. If they're creating a DMZ from scratch, it could potentially be quite a bit of work. If they're jammed, they may not have any brainpower free to help you with it.
If they can't help you, but you have some budget and live in a big city, you could potentially hook up via a laptop and one of the high-speed cell data networks. In many areas of the country, this is fast enough to be perfectly functional, and you shouldn't need much help getting it running.
However, if you do this without IT's blessing, you should NEVER AGAIN connect that machine to their network... with that data card in it, that machine can potentially be a back door straight into their systems. If you have any VPN software on that machine with credentials to connect to your network, remove it. Before you plug in the cell card, turn it completely into a standalone system, and make sure that all the passwords are different than any others used on any of your other networks, including your home machines. You are assuming that it WILL be compromised, and you're doing your level best to make sure that there's no data on the machine you care about, nor any sensitive passwords that could give an attacker a wedge into your corporate network.
Remember, even if you pull the card out, the machine could already be compromised, so connecting it to the internal network could let it 'phone home' to a hacker and give him full access to your systems. So it should never again be connected to the work network in any way, shape, or form. You should always think of it as already compromised.... dangerous and radioactive. This may or may not be actually TRUE, but it's not a bad habit to be in, especially when you're not an expert.
Disable the Ethernet port so you remember that you don't trust that machine. If you need to move data, burn a CD and sneakernet it. Turn the machine off when you're not using it.
If you don't have the technical chops to lock it down well, wipe and reload it on a fairly regular basis. Use new passwords every time.
I don't recommend you do this for a whole bunch of reasons (not the least of which is the potential of getting fired), but if you can't get IT to help you, setting up a cell card like this should be pretty safe. Just make SURE that it knows NOTHING about your internal net before hooking it up.
Even if IT is otherwise busy, they might be able to give you a 'clean' machine that you can do whatever you want with.
I read an article by a knowledgeable-seeming science-fiction author a few years back, about antimatter drives.
The first interesting thing I got out of the article was that any rocket would carry about the same amount of reaction mass for a trip of any distance. Whether going to the Moon or going to Jupiter, it would carry a fixed amount of some cheaply available gas. What would alter would be how much antimatter would be mixed into the propellant.. for a Moon trip you might use a hundredth as much as you would go to go Jupiter. By mixing more antimatter in, they'd increase the boost without needing more (or less) inert propellant.
He didn't address your issue directly, but one thing he pointed out is that both antimatter and fusion reactions produce gamma rays in abundance. If I understood him correctly, gamma rays are impossible to block completely. They can be attenuated, but there is no substance known that will completely block them. Some will always get through. You could have a mass of solid lead the size of the Earth and some rays would still penetrate.
This means that astronauts would be taking a heck of a dose from their engines, no matter what shielding was used. Add in the radiation from the Sun and astronauts probably wouldn't last long out in space, at least in these parts.
Unless we discover some way to reduce or eliminate radiation damage, it's not likely we'll be sending manned ships anywhere using either of these technologies. They could be perfectly fine for unmanned probes, but barring major medical advances, human evolution, or some kind of radiation shielding technology, humans aren't going far on antimatter drives.
One of the things I like to think about are the limitations of rationality. Rational thinking involves ignoring most of what's true about something, in order to deal with the aspect in which we're interested. When we're thinking about a chair as something to sit on, we're not really thinking about it as kindling, or as potential building supplies for a chair pyramid, nor the possibility of using one of the chair legs as a club. We tend to think about just one property of an object at any given time, which (temporarily) blinds us to its other properties.
I really wonder if the whole time travel metaphor thing isn't an artifact of bad modeling, rather than anything true. Time is often claimed to be 'a dimension'. A one-dimensional object is a line, and a two-dimensional object is a plane. Three dimensions is real life, what we see, and the fourth dimension, ergo, must be time.
This is all very pretty, but there's one big problem... there are no one- or two-dimensional objects. Nothing in real life has less than three dimensions. Never, never, not ever. One and two dimensional objects are imaginary, they do not and can never exist. Dimensions are a highly useful mental model, but since there are NO objects with less than three dimensions, there's a good chance they're not actually true in any meaningful sense.
Like most rational thought, dealing with space in terms of wholly imaginary dimensions allows us to do some really amazing stuff. But it also blinds us to most other things that are true about the universe while we're thinking that way.
I very strongly suspect that time is not, in fact, a dimension, and that all the mathematical hand-wringing that goes on about time travel is trying to make an impossible thing work out of a false, imaginary premise. There are no one- and two-dimensional objects. Objects always have height, depth, and width. Always.
If the first and the second dimensions are imaginary, it seems entirely reasonable to assume that the fourth must also be imaginary. We like to count up, and having invented three dimensions, we may be looking for a fourth. It looks to me like we're confusing our map for the territory.
I strongly suspect that ultimately we'll find out that time is a force or a process, and that the model of it being a line we're traveling along is false.
Your business obviously thought this software was valuable enough to spend two million bucks making it. It's a competitive advantage; you have it, and the other people in your industry don't.
If you open source your software, then your competitors get it too. But instead of spending two million bucks, they get it for free, so they have money to spend on OTHER stuff that YOUR company doesn't have. You are weakening yourselves considerably, hoping to get a payoff of equal or greater value.
So you'd need outside code contributed that would be worth at least another couple million. The chances of that happening are laughably small. Unless your software addresses a very broad horizontal market (at least tens and probably hundreds of thousands of possible installations), there's no way you'd ever get two million bucks' worth of patches.
The LAMP programs, Linux, Apache, mysql, and Perl, have probably gotten that level of free time donated. But there are very few others, and you most likely don't have 'the new Apache' on your hands. Let's charitably put that chance at 'tiny'.
Not all code should be or needs to be free. You would definitely be doing your customers a favor if you included the source code with the product, with strong restrictions on what they can do with it. But just releasing it into the wild directly and immediately harms you a great deal, and has only a small chance of paying off.
From the very rudimentary data provided, it looks like open sourcing your product is almost exactly the worst thing you could do.
You're right and wrong at the same time.
For a long time, Intel et al have been shipping processors that emulate the X86 instruction set. They don't actually use it themselves; they read an X86 bytecode instruction stream and translate it, on the fly, to their native micro-ops, whatever they might be. They use all kinds of tricks to allow more X86 instructions to be executed simultaneously, stuff so complex it's mind numbing.
But from YOUR perspective as a programmer, you're still using X86 and it's just as brain-dead as it ever was. (X86-64 is a little better.) You don't have any access to the native, 'real' instruction set, so the fact that modern chips are just hardware emulators of X86 doesn't matter. From your perspective, it's the same old crap, just a lot faster.
It's really a shame that this ugly architecture has gotten so far and so completely dominated everything. But, in fairness, very few programmers are working down that low anymore, and compilers hide the worst of the nastiness.
Still can't help but think that if all that effort had gone into accelerating a cleaner architecture to begin with, we'd have come farther and faster.
VHS has won again.
It's a very weird idea... it feels a bit like waking up in the Twilight Zone or something.
I forget the name offhand, but there was an early computer company, possibly Amstrad, that preannounced that they were coming out with this really amazing new machine. Customers loved the idea, and stopped buying all the existing machines. Sales dropped to 0 and they died before they could get the new hotness out the door.
Apple has always known about this, and has been very aggressive about not preannouncing ANYTHING, so as to not hurt sales of existing products. This violates that precept so badly that I can't help but think it's prompted by panic (or anger, as some have hypothesized). I'm not sure whether or not it's a bad idea. Their existing sales are going to drop, perhaps precipitously, while they try to figure out how to make OSX on Intel work. But if they didn't preannounce, then the developers wouldn't have time to get ready for the transition, which could potentially be worse.
As an aside, PPC emulation is never going to be very good, and all that specially tuned Altivec code has just been junked. That's a huge investment down the drain.
They have to know that this is a bet-the-business move. They'll have to execute nearly perfectly, and very quickly, to make this happen without pissing off too many people. And they're going to have to continue to execute flawlessly for a long while. Apple's good at that, but this is asking an awful lot of their engineers. They must really believe their backs are against the wall.
I wish them well, but there are soooo many ways they could screw this up.
I doubt they'll get to full human-brain awareness level anytime soon, but ... what if they do? What happens if they create a sentient being inside their simulator? When they're done with the simulation and it's time to start on something new, is turning off the machine killing the 'creature' inside?
And even if it's not as smart as a human, what then? What ethical guidelines are appropriate? When is it okay to destroy a thinking being, even if you created it yourself? And how complex must it be? Killing a beagle or a dolphin isn't murder, after all, but it's still considered wrong in many cases to do so.
Are AIs cute and cuddly and protected by humane-treatment laws, or scary and kill-on-sight, like spiders and snakes are for many people?
How smart does an AI have to be to have rights against termination?
We've been sort of doodling around with these thoughts for a long time, but it's getting to the point where we may actually need the answers.....
I can certainly see not upmodding this one, as it's a single example in one specific geographic area, but -1, Redundant?
I really gotta wonder about the mods here sometimes.
My mom is out in the boonies. She's been on Starband for a number of years, and has spent a ridiculous amount of money on equipment.... her cable modem has melted down twice.
This last time, it stopped working completely, even to the point of talking to the computer at all. So I called Starband. It took two hours on hold before I could talk to someone. This was actually two calls, because I called, waited an hour, and then the system hung up on me. The second call took another hour and I finally got rhrough to a technician. He said "oh, sorry, your service is provided by Dish Network, you'll have to talk to them."
So I called Dish, who answered within a few minutes... but they couldn't help me. "Sorry, you'll have to talk to Starband, we don't know anything about the hardware." I had him ask his supervisor and the supervisor confirmed it.. I had to talk to Starband.
So I called Starband back. It took about 90 minutes... they had the good grace to disconnect me after only 30 minutes of waiting on my first call, so I got started on my second hour a little sooner. I explained my problem to the guy with the strong Indian accent. He said "You have to talk to Dish about that." And I said "No, I can't, it's hardware and you guys have to do hardware." He said back, "You have to talk to Dish about that." And then he hung up on me.
And note that this is the SECOND $1000 modem she's bought from them. I'm not sure she paid the full $1k for the second, but it was still a lot, over $500 for eure.
Dish Network and Starband are THE WORST network providers I have EVER seen. You are better off with carrier pigeon protocol. DO NOT sign up with these people. They define customer hostility. Their network is atrocious and their customer service... well, see the above paragraph. This company deserves to be nuked from orbit. You've gotta realize... even when the net was WORKING, it was only for a very, very loose definition of working. Starband would be a perfect synonym on Wikipedia for 'incompetence'.
So I went back in and told my mother she was changing providers. We've had various misadventures before she found one she liked. She's in a fairly remote area, but has reasonable wireless coverage, so we figured we'd try setting her up that way. (she's about 1.25 hours NE out of Atlanta, Georgia.)
In that area, Verizon's service was terrible, nearly as bad as Starband's. My guess was that the cell network was okay, but that it was grossly underprovisioned at the cell tower(s). At 4AM I got pretty good download speeds. At 4pm the network was essentially unusable. It appears Verizon is willing to accept very, very poor levels of service quality while still happily signing people up. If you go with Verizon, MAKE SURE you have a money-back guarantee. Their network was no better than Starband's, and that's as profound a statement of dissatisfaction as I can make.
Next she tried Suncom, which is the local AT&T affiliate. Their service plan is very reasonable. It's almost exactly like a 56k modem, and quite reliable. You won't get your pages quickly, but you WILL get them, which is more than I can say for either Starband or Verizon. Suncom's pricing is very good too.. the minimum each month is $5, and the maximum is $50. So if you're not using it, you don't get soaked, and the most you can pay for any given month is $50. Suncom has very good, customer-oriented price plans, and if you're in the South, they're a good choice.
Unfortunately, my mother kept roaming into Cingular space, even though she was just sitting in her armchair in her living room. Suncom tried to bill her several hundred dollars for roaming, backed off on that, but told her she couldn't use it, in the future, when the network said Cingular. It was doing that more and more, so she just cancelled/returned that service and tried Cingular's flavor.
In her area, Cingular appears to be quite good. I haven't had a chance to try it myself yet, but she says i
We would most likely have hunted any such species into extinction long ago, probably during prehistoric times.
:)
There can be only one.
Dude, you're not thinking. That isn't my argument at all. I don't know where the hell your 'prove a 50% probability' crap came from.
No matter HOW improbable intelligent life is, it has happened at least once. Therefore, it is possible. In the vastness of the Universe, asserting that it hasn't happened at least twice is STUPID. No matter how small the chance for life on any given planet, the sheer vastness of the Universe essentially guarantees we are not the only intelligent life. The probability isn't 50%, it's 99.99999... %, going to a very large number of nines.
But the distances are so great -- the Universe is so large -- that if the chance is small, each race will probably go extinct without ever seeing another intelligent species. Only if complex life is exceedingly common will we be at all likely to encounter alien intelligence.
Well, there are two pretty compelling bits of evidence.
1) Intelligence, such as it is, has evolved at least once.
2) The Universe is unimaginably huge. Just our own galaxy is vast beyond the ability of humans to even imagine. One of the early Hubble Deep Field studies, looking at one of the darkest places in the sky, saw 40,000 GALAXIES in an area of the sky equivalent to a grain of sand held at arm's length.
Given those two facts, doubting alien intelligence strikes me as profoundly stupid. However, unless it is extremely common (which I doubt), the chances of any of that intelligence being within a distance we could detect is pretty darn small.
The probability of alien intelligence, in other words, is essentially indistinguishable from 1. Given the constraints of lightspeed, however, the chance that we could ever meet and TALK TO such aliens is probably very close to 0.
As I understand it, true randomness only comes from measuring effects at the quantum level, like radioactive decay. (mentioned in other threads here). As nearly as we can determine, individual quantum events are absolutely random and completely unpredictable. We can make fairly precise predictions about groups of events, but we can't predetermine when, say, a given atom will decay. We can tell you about how many will decay in any given second, and this prediction will become more and more precise as events accumulate over time, but we can't tell ANYTHING about an individual event until after it happens. True random number generators depend on this.
A step up from that are events derived from things believed to be random, like user input. However, they are always mediated by other factors, like the circuitry and response time in the keyboard or mouse, for instance. This is probably pseudo-random data at least some of the time. The numbers are still pretty random, but in theory, skilled cryptographers might be able to tease patterns out of the bitstreams. This could potentially allow them to successfully mount mathematical attacks on encrypted data.
Another method of generating numbers is with an algorithm... given a seed number, it will produce a stream of 'random' numbers. If the seed and the algorithm are known, many of these streams can be cracked. Because they aren't really random, but have many of the features of true random numbers, they're called pseudo-random. With a strong enough algorithm, at least in theory, encryption based on pseudo-random numbers should be just about as difficult to crack as encryption based on REAL random numbers.
But even if we can't easily detect it, any number stream generated by an algorithm DOES have a pattern. We may not be smart enough to find it yet, but there's a good chance we may someday be that smart (or have that much computing power). Encryption based on truly random numbers should be crackable only by brute force; no analysis should ever reveal a pattern, since by definition there will be no pattern to find. (If we ever DID find consistent patterns in true random numbers, this would shake our understanding of the universe right down to the foundations... just about to the point of having to start over from scratch.)
Also note that a number stream generated by a Random Number Generator(RNG) can't be *too* random... if the chaos is too perfect, it becomes order, and is attackable. (weird, eh?)
I'm neither a cryptographer nor a mathematician. This is how I understand things to be, but I'm just a layperson and could easily be wrong. Pay attention to replies.
You do realize that I'm just talking about the internal self-consistency of a plotline, right?
You gotta realize, that even by Empire standards, the Death Star is unimaginably huge. It would be a titanic feat of engineering to build it. Consider: the Earth has a huge amount of iron in it... if the Death Star were made out of iron, it would require stripping away like a SIXTH of the Earth's ENTIRE MASS to build it. (those numbers are, admittedly, pulled out of thin air and are guesses, but I think they're in the right size range).
Now, they'd probably go get asteroids and use those, and they'd probably have an enormous assembly fleet, but it would still be an unbelievable amount of work. Doing it in just 20 years would strain even a Galactic Empire. And having the skeleton built is about 0.01% of the whole job. Consider houses... they'll have the frame up in a couple days, but it still takes six months to build. Compare that to the enormous internal volume of a Death Star versus its relatively small surface area.
The fact that they had a second largely built in just a few more years strikes me as far less believable than the long timeframe to do the first. They would almost have to have started it halfway through the first project.
Rome wasn't built in a day, and I very much doubt the Death Star was built in a decade.
Oh, absolutely. I'm not a big Microsoft fan by any means, and they do it too. But I'm strongly suspicious that the PS3 may not actually deliver what they claim... given today's technology, the specs they're quoting are going cost them like $750/unit. I don't think they're going to want to ship that thing at $300.
Basically, I'll believe it when they actually ship it -- in both cases. One of the mainstays of FUD is to announce your own much better product that won't come out for a long time, to try to distract momentum from the product you can actually buy TODAY.
:-)
Admittedly, in neither case can you actually buy the consoles in question, but Sony is going to be late to the party... so even if they don't ship what they promise to, or if it doesn't work as well as they claim, they might believe that the backlash would be less damaging to them than letting the 360 get a solid foothold in the market.
I personally find it hard to believe that they can ship that much hardware at a resaonable price point. They've been known to overstate future products before. And the XBox appears more likely to be able to hit $299, which I think is critical to any kind of mass-market acceptance.
And then, of course, you have Nintendo, whose claim to fame appears to be "Our console is much smaller." Yeah, that's a big selling point, for sure.
Good god, how many times does this have to be said? Free as in beer is great. Free as in libre is better. If I use free-beer code and become dependent on it, then I am giving the person who wrote it some degree of control over my life. And I may not really realize the consequences of doing so until much later. See Linus Torvalds and Bitkeeper. See also Microsoft, who likes to impose new license changes unilaterally in service packs, or modify file-save formats to force you to buy another multi-hundred-dollar upgrade to your office suite.
People are upset because Sun can change the terms of Java whenever they choose. They have not been willing to really give up control of Java. They desperately want to screw people the way Microsoft gets to. They don't seem to realize that nobody else will EVER be able to print money like Microsoft does, at least not in the computer field. Between Microsoft already existing, and Open Source coming up from underneath, the best Sun can hope for is an honest living... they're not going to be able to both maintain control over their platform and also have it be a market-determining force. They can have one or the other, but not both. Only Microsoft gets both.
People are uptight because using Java-based OO means making themselves vulnerable to Sun, in the same way that Word users are vulnerable to Microsoft. They want the assurance that they can't be orphaned, forced to upgrade, or forced to accept some other unilateral license change on Sun's part.
I don't think I've ever heard ANYONE argue that OO is actually better than the Office suite. People use it because it is cheap and free-as-in-libre, and get most annoyed when the libre part starts going away. The few serious users they do have exist BECAUSE of Microsoft's abuse, and those customers are highly snsitive to behavior that looks even vaguely similar.