Running out of room in space? Well, Douglas Adams put it fairly well: "Space is really big! You may think it's a long way down the road to the chemists but that's peanuts compared to space."
They're not exactly running out of room where "the corporations" usually put their satellites. Look, geosynchronous orbit is about 40 Mm from the center of the earth. That means that there's about 240 Mm of linear space in geosynchronous orbit. I say "linear" because they all want to be in a circular orbit in the plane of the equator. Since each satellite has to be about 3 degrees from its neighbor because of the beamwidth of the signals being sent to it, that means that there are only about 120 active satellites in geosynchronous orbit at any one time. Of course, there are dead satellites and spares, but each slot is 2000 kilometers wide. That could easily soak up ten satellites or so and they'd still be so far apart that you wouldn't be able to seen one from another. In other words, they're not exactly running out of real estate.
I once read a wonderful cure for insomnia that was a NASA report on the odds of colliding with space junk. The upshot was this: The odds of a significant collision was highest in those orbits closest to the earth. There are two reasons for this. First, even though lower orbits decay so there's a kind of a natural cleaning process, there's a lot more junk close to the earth and the fact that there's less volume near the earth, what with the volume of a thin shell of a particular thickness being roughly proportional to the square of the radius, so the density of stuff is a lot higher. Second, lower orbits tend to be inclined with the equator. That means that the closing speeds tend to be much higher and so the potential damage is much larger.
Now, if space travel to geosynchronous was routine, it seems likely that there would be an effort to salvage dead satellites, which would, in my opinion, be beneficial to many people, but NASA's big thing is that only "steely-eyed missle men" get to fly into space. Ragmen need not apply, so it'll never happen while NASA holds the keys to low earth orbit.
All this does is point out the folly of attempting to label some versions of software as "stable". How do you, or anyone, KNOW that it's going to be stable? Well, the only reliable way to do that is to run said software for a while under production conditions. Even that is uncertain. So, either you adopt a kernel early on and expect problems, or you wait for others to try the kernel first and hope that they discover all the problems for you. I do the former, others do the latter.
From my perspective, the difference between the "experimental" kernels and the "stable" kernels lies in a difference in direction and emphasis. With the "stable" kernels, changes are usually more about fixing defects than about changing or adding features. With "experimental" kernels, the emphasis is more on changing or adding features with less concern for the people who might try to use the kernel to do real work. However, that doesn't mean that a "stable" kernel will work, for fixes may beget other problems and it doesn't mean that an "unstable" kernel might not be solid as a rock, for the intent is to make the new features work well.
Early "stable" releases are probably best approached with caution, and not just because the kernel itself may be unstable. In my case, the problems I ran into with 2.6.0-testN kernels had to do with the fact that there's some pretty big changes in the way things are supposed to be done so a lot of the supporting software had to be fixed. (See, for example, some of the defect logs for Debian's initrd-tools package, where my email address appears a few times as I reported problems so that Mr. Xu could fix them.) Once I got the kernels installed and bootable, I didn't have issues with stability. Not at the house.
On the other hand, the SCSI driver from 2.6.1 refused to function at all on my work machine, but it did work on similar hardware at home. I never nailed that problem down because 2.6.2 appears to work on both machines.
Re:Global IPv6 Service launch event in Brussels
on
The State of IPv6
·
· Score: 2, Insightful
Bluefirebird wrote:
There is still some way to go on IPv6 but the main problem is the lack of IPv6 link requests from users that makes the ISPs ignore IPv6 as an important issue.
This is why I can't take the IPV6 folks seriously. Demand for addresses comes from the leaves, not the root. So what if every backbone provider has native IPv6 routing throughout they're backbones? They're not the ones who use addresses by the ton!
I've got an IPv6 tunnel and addresses from a TLA, but I can't get native IPv6 access because neither the cable modem that I use nor the equipment it talks to upstream knows anything about IPv6. In fact, there is little, if any, end user WAN equipment that speaks IPv6 natively. Availability of that kind of equipment is necessary before a "global service launch" has any kind of meaning.
You have not been keeping track of global anything for more than 140 years because of the difficulty associated with measuring anything globally. I go to the web page and I see an impressive graph. Too bad there's no indication of where the data comes from. Even if I stipulate to your data, then the crucial question becomes "Why is it happening?" You seem to have concluded that it is due to humanity's release of carbon dioxide into the atmosphere. Defend that conclusion.
In defense of the other side, I point out that heat transfer away from the surface of the earth relies more on convection, which is not affected by the quantity of carbon dioxide in the atmosphere, than on radiation, which is where the "greenhouse effect" comes in. I also point out that water molecules are, on a molecule-by-molecule basis, at least as efficient at blocking infrared radiation as carbon dioxide, and that there are two orders of magnitude more of them in a typical sample of air than there are of carbon dioxide molecules. That means that the most important greenhouse "gas" in the atmosphere isn't a gas at all, it's water vapor. Indeed, that can be seen in the recorded experiences of people in the desert from the Roman legions onward.
So, why is a trace element supposed to cause the bulk of the effect? Perhaps there is a simple explanation. Do you know what it is?
The invention of the airplane is generally credited to Sir George Cayley because it was he who realized that you didn't need flapping wings to build a heavier than air flying machine. The whole "four force" concept (lift, weight, drag, thrust) was his idea. The Wrights basically built upon his concept.
What the Wright Brothers did do is build the first successful, controllable airplane. The controllability is the key because they were the first folks to really work out how to make an airplane go where you want it to go. They also figured out that it was going to take some practice for the pilot to become proficient in flying it. They also built propellors whose efficiency wasn't bettered for decades and along the way they laid the foundation of the whole theory of propellors.
In fact, like the telephone, the airplane is a perfect example of one of those things whose creation is inevitable once the supporting technology is available. There were many, many folks working on the solution to powered flight once small and lightweight engines were available to power the craft. The groundwork had been laid more than a century before with Cayley's conceptual leaps all it took was somebody to work out the details perhaps with a leap or two of their own.
As a practical matter, history records that the aileron was invented by Glenn Curtiss in an attempt to get around the Wright patent on the airplane. History also records that it's not that difficult to get a newspaper reporter to write a story even if it's only printed in one paper. When people put forth the claim that the Wrights built a successful flying machine and the date on which it was done, they produce a photograph of their machine flying and a dated telegram with the details of the flights.
On the Website talking Mr. Pearse's claim, there is nothing of the sort. The lack of evidence that the machine flew is explained with "he didn't realize the historic importance of the flights". What crap! Flight had been a human dream for thousands of years. Wouldn't fulfilling that dream seem to you to be of some historic importance? Shouldn't it have occured to one of the numerous witnesses to mention something to somebody or to write it in a diary or something? Everyone else working on heavier than air flight seemed to realize they were solving a momentous problem, why didn't anyone in Waitohi, New Zealand?
Functions in the scanf family are truly useful, if you can grok them. However, error recovery can be a pain, so it's not generally advisable to use scanf() or fscanf(). Instead, one practice that I use is to read lines into a buffer (using fgets so that you'll be less likely to have a buffer overflow) and then run sscanf on it.
You should always check the return value from any functions in the scanf family against the number of items whose value you want to set. That will tell you whether or not you've read everything you expected. Also, bear in mind that it can be tricky to set up the format string properly. I can't think of any examples of nonobvious behavior off the top of my head, but I find I use a lot character sets when I use scanf.
Actually, most of the increase in user base recently is from people who don't contribute software. That means that I need Linux's popularity to be a tiny fraction of what it is in order to be happy with it. That's why I was happy with it even back in 1992.
It may come as a shock to you, but I don't particularly give a damn about 'my' market share. I don't need zillions of people choosing what I choose to make it the right choice for me. I use what I like and I will continue use Linux as long as I like it.
I am fortunate that my employers think Linux is worth using, and it's nice for Linux to have enough popularity that the software tools I need get developed, but they would be there even if the Linux usage had peaked in 1992. Maybe there will be a revolution. Perhaps it's even under way right now. Who knows? I'm not in it for your revolution. It isn't important to me.
If the FSF folks are, like you, so confident that their way is the only right way to do things, then why do they care about what other people do? What difference does it make if you're going to "win" in the end? If you're really right, then everyone else will eventually realize it and do what you do of their own volition.
If I were to ask the question, I wouldn't be interested in Macintosh compatability. Can you give me a quick link to somewhere I can find retailers for those PowerPC motherboards? A few google searches didn't result in anything like some place I could buy one of these.
When I looked in to this a while back (has it really been 15 years?) there was a requirement that FM broadcast stations put out at least 3000 watts. There was an exception for "educational" stations where "educational" is defined by the state in which the station is licensed but generally includes schools and sometimes churches. Since then ISTR that a rule was passed allowing low power FM stations, but it seems that low power FM stations aren't quite what you're looking for. (See The FCC's page on the subject.)
The first step in the process is to find out the rules that apply. That takes some research. Speaking with an attorney that specializes in communications is one way of doing the research. Alternatively, you can get copies of the rule books. You're looking for CFR Title 47 as well as the relevant FAA regulations if you're going to have a tower structure. These are available from the Government Printing Office Web Site They're also available at many libraries.
Frankly, I can't see the point of just breaking the law in public. In my opinion, a better course of action would be to set up the conditions for a test case that could be won in the courts. That will likely require some public lawbreaking, but will also require there be something about the lawbreaking that demonstrates how the law in question isn't reasonable. Simply showing how easy it is to violate said law isn't going to do that.
I've been programming professionally for the last decade or so and the largest programming team I've been on numbered six developers, one dedicated tester, and the manager. Before that, the largest team I had worked on was two. Just because something is commercial doesn't mean that there's going to be 1700 programmers working on it. In no case did we have any particular need to use a version control system. Indeed, it was widely viewed by the developers that the benefits of a VCS were outweighed by the cost of the changes required to make use of one.
However, I now do some things with CVS, even if I'm the only programmer working on a project becaues it means that I can work on that project on any computer I happen to be using without worrying (much) about which computer I last used to work on it.
For things I don't use CVS for, I use "make source" which I have set up to create a version-tagged zip file of the source in the current directory tree. (Crude, but effective, inexpensive, and quite easy to do. I also put all those zip files, including the old revisions, on the installation CDROM something that's already saved my posterior twice.)
Am I sure those are really antibiotics? Well, the last time it happened, the prescription said amoxicillin. (Which, a google search reveals, is a "semisynthetic antibiotic, an analog of ampicillin, with a broad spectrum of bactericidal activity.")
Now, please note that, for the case in question, I did not go to the doctor in order to get some sort of magic pill. I was trying to determine whether or not I'm going to get better without treatment. It is the doctor that assumed that I want some sort of magic pill. I suppose that it's too much trouble to explain that it's really best to let some ailments run their course and then to ask me what sort of treatment I would consider appropriate. I certainly would choose to not take medication that isn't likely to help, and I'm not alone in this view.
If diagnosis is the easy part of medicine, then how come it's so often wrong? I mean, I've seen perhaps four diagnoses in my life that were of any real importance. By that I mean I have been in or around four situations where I wasn't suffering from a runny nose or other similar thing where the appropriate thing to do was to tell me that I would get better and to send me home. (Note that I have never gotten that treatment. You doctors usually prescribe antibiotics and a decongestant for runny noses. Stop that, it's counterproductive. The runny nose is most likely a viral infection or an allergic reaction to something so antibiotics aren't indicated and I don't like the side effects most decongestants have so I won't take them. That is, in fact, why I've long since stopped seeing doctors because of runny noses.)
Anyway, of those four diagnoses, three were wrong. Based on that, it sure doesn't sound like diagnoses are easy to me! Add to that the fact that I'm pretty good at troubleshooting and I'm one of the few that I know of. Most people flail about trying things at random and, as far as I know, training isn't much help for most of those people. Yes, it's easy to memorize a few pat answers to the most common problems, which is why many people who visit the "doctor" wind up seeing a PA, who forwards to the doctor only those cases whose diagnosis is in doubt, but that's exactly why this sort of thing is important. As time goes on, doctors are going to be less and less likely to see the simple cases.
You mention psychiatric diagnosis, so I'll talk about those. A quick check of my local DMDA chapter shows that some 70% of those suffering from some serious mental illness were misdiagnosed at least once. I can't help but think that a computer program that prompts the asking questions about typical symptoms of mania and schizophrenia would reduce that because most of the misdiagnoses start as a diagnosis of depression because it's what people complain about. I know the doctors don't ask those questions because in the sample that I have (8 so far) none have asked the right questions to make what we (that is, myself and the ill person) now believe is the correct diagnosis.
In any case, since visiting a doctor (and I spend a lot of my time in doctors' waiting rooms so I know this quite well) is something like an hour waiting to see the doctor followed by maybe 10 minutes of answering questions I don't understand, filling out the forms while I wait can't do me any harm even if the diagnosis is not remotely in doubt or irrelevant to the treatment, can it? I mean, it becomes part of the patient history just like the temperature and blood pressure check you're going to do whether I come in with a fever or with a splinter, right?
Heck, I suggest you put terminals tied into that database in the ER waiting room so there'll be something to do while you're waiting the 4-6 hours (on average) it'll take to get to the head of the line.
Morse code was never the problem.
on
Field Day 2002
·
· Score: 1
I think that Bruce Perens is an idiot for blaming the problems of ham radio on the morse code. The problems that ham radio has of attracting new people to the hobby stems from the emphasis that many people (including Mr. Perens) place on amateur radio as an activity where machines talk to other machines.
Packet radio and APRS are the results of what I'm talking about here. I get lots of "wow, don't you think this is cool?" from the ham gadget freaks, and lots of "why would I care?" from everybody else.
Actually, yes I can argue against extending that copyright, if it means that I have to extend the copyright of all works of similar vintage.
The problem with the current system is that copyright law serves to un-publish works. You can't make copies of a work because it's protected by copyright, but it's out of print because the publisher (and, make no mistake, the publisher has all of the rights in the current system) has decided that the market for that work isn't large enough to be able to keep it in print.
Working against that are examples of very long-lived works such as To Kill A Mockingbird or "Peter Pan". The vast majority of all works don't have nearly the longevity of "Peter Pan". So, extending the copyright for all of those works simply because one of those works might be continuously in print for many decades leads to works that are long forgotten. We are basically destroying our cultural heritage by not allowing works to fall into the public domain.
The suggestion I once made, that copyright holders be required to periodically pay for copyright in order to keep it, was met with cries that I was trying to prevent people from becoming authors and that I was trying to starve current authors' grandchildren. This idea would allow works like "Peter Pan" to stay in copyright basically
forever, (meaning "until the fees aren't paid,") and still become public domain as soon as interest flags sufficiently for it to fall out of print.
The only part of my idea that I have trouble defending is the fact that I'm spending a lot of time thinking about making sure that people can still have access to works that they didn't care enough about when they were first published to keep them in print. Why should anyone care about a zillion B-movies that no one is ever going to re-release?
The 100m limit is there in order to make it possible to do collision detection on half-duplex links. That's also why Ethernet has a minimum packet size. If you're running full-duplex links, you can run the cable much farther because there'll never be any collisions. I don't know what the limit is with CAT-5, but you're electrically limited if you're running full-duplex. I would expect several kilometers, at the very least.
Sure. The company I work for deployed Linux in an embedded product and it was a no-brainer. It certainly was easier than the Embedded-NT that was chosen for another part of the application. The secret to success was the fact that we use all PC-compatable parts.
The fact is, the farther you get from a desktop computer, (and, at least with Linux, the farther you get from X86 PC's, or so I've been told,) the more trouble you're going to run into trying to use a desktop operating system in an embedded device. Most embedded devices don't look very much like the general-purpose computers most people are used to.
Of course, the other experiences I've had with commercial RTOSes is that they require considerable fiddling just to get them to start and start your application. Also, remember that WinCE isn't really considered an embedded operating system by many embedded people. Something like psOS or Lynx or QNX would be more like what's considered a "real" embedded OS.
When I looked at MIMEDefang for use on the mailing-list computer I'm setting up for a friend, it appeared to be sendmail-only. Is that true?
Re:I'm trying it, need to figure out the tools
on
Firebird Goes Gold
·
· Score: 1
The SQL engines that I normally use come with simple programs that allow me to type SQL at the back end and look at the results. The next step should probably be to get a book on SQL (I highly recommend The Practical SQL Handbook) and use a program like that to define some tables, stuff some data into those tables, perform queries on that data, and so forth.
Once you're there, you're probably going to want to replace those handwritten queries with a friendlier interface, so you'll be looking at developing applications that use those queries internally. However, giving specific advice about that is hard because I don't know what kinds of applications you'd be interested in doing.
Well, that all depends upon what you mean by "enter a market".
Microsoft has had products initially fail any number of times, but it's not clear that those products were new entries to the market. Further, Microsoft did not always back development of those products until they were a success. Lots of people talk about Bob, but do you remember QuickPascal? How about the MSX standard? I haven't been able to find very many "Natural Keyboards" recently.
Yes, with Windows and with their general office software, they initially had difficulties which they overcame, but it's not clear that those were entries into a new market for them. For example, Windows can be considered an operating system and Microsoft was already in the operating system business, so it's possible to argue that they weren't entering a market when they released Windows 1.0.
As near as I can tell, Microsoft has little or no track record of entering into new markets so I don't think one can claim that they have a history of backing, in the long term, entries into new markets that do not have a lot of initial success. They seem to do a pretty good job of picking the battles to fight that they actually need to win in order to maintain their market domination, but it's not now clear that the X-Box will be viewed as such a strategic product by the powers that be at Microsoft.
It just so happens that many of these sorts of simulations have runtimes that exceed the mean time between failures of the hardware they run on. I have been told that figuring out how to deal with that (and other failures, like kernel panics and whatnot) along with ancillary decisions (like whether or not it would be better to store an intermediate result or re-calculate it) occupy an awful lot of the time of those who write such things.
There doesn't seem to be an awful lot of off-the-shelf application software for clusters. This is probably because of the nature of the problems these clusters get used to solve. That, and the fact that software development for a nontrivial application usually takes quite a bit of time, means that the amount of effort required to put the application together is going to swamp the amount of time required to set up even the Beowulf option.
Further, given the vagaries of software development, a showstopper that isn't immediately obvious to someone who has read the documentation isn't likely to be found until well into the development cycle. That means that if there are lurking problems, a complete scrap-and-restart cycle is going to be needed after a lot of resources have already been committed. That encourages a close reading of the documentation for the base software and other, more general, clustering documentation, and a good look at more general concepts in the design of software that is intended to be clustered in hopes of finding potential pitfalls before time and money is spent on setting anything up.
Frankly, I don't think it would be a good idea to go out and buy (or acquire through some other means) a collection of Macintoshes and set it up as a Pooch-based cluster without doing quite a bit of reading on the subject including everything you can get your hands on about all of the available options.
Not to mention the fact that it hasn't been demonstrated to my satisfaction that it is easier to set up a Pooch-based cluster than a Beowulf cluster. Frankly, I think a lot of people would spend a lot more than an afternoon setting up the initial requirement for Pooch, a collection of networked Macs, if they didn't have one already. Of course, it's not any less difficult to set up a network of Linux machines, but it seems to me that putting the machines in place, making sure they all work, configuring the network settings, and terminating, running, and testing all the cables is going to easily consume an afternoon for all but the smallest purpose-built clusters.
One last thing, kind of off the point. While I generally agree with the statement that Free Software is only free of cost if you're not paying for people's time, my time is worth exactly what someone else is willing to pay for it, and for much of the day that is exactly nothing. That means I can gain all kinds of knowledge for absolute free and then apply that knowledge to what I get paid to do and reduce their costs because I can afford to do it on my own time rather than theirs.
Slashdot Mirror
They're not exactly running out of room where "the corporations" usually put their satellites. Look, geosynchronous orbit is about 40 Mm from the center of the earth. That means that there's about 240 Mm of linear space in geosynchronous orbit. I say "linear" because they all want to be in a circular orbit in the plane of the equator. Since each satellite has to be about 3 degrees from its neighbor because of the beamwidth of the signals being sent to it, that means that there are only about 120 active satellites in geosynchronous orbit at any one time. Of course, there are dead satellites and spares, but each slot is 2000 kilometers wide. That could easily soak up ten satellites or so and they'd still be so far apart that you wouldn't be able to seen one from another. In other words, they're not exactly running out of real estate.
I once read a wonderful cure for insomnia that was a NASA report on the odds of colliding with space junk. The upshot was this: The odds of a significant collision was highest in those orbits closest to the earth. There are two reasons for this. First, even though lower orbits decay so there's a kind of a natural cleaning process, there's a lot more junk close to the earth and the fact that there's less volume near the earth, what with the volume of a thin shell of a particular thickness being roughly proportional to the square of the radius, so the density of stuff is a lot higher. Second, lower orbits tend to be inclined with the equator. That means that the closing speeds tend to be much higher and so the potential damage is much larger.
Now, if space travel to geosynchronous was routine, it seems likely that there would be an effort to salvage dead satellites, which would, in my opinion, be beneficial to many people, but NASA's big thing is that only "steely-eyed missle men" get to fly into space. Ragmen need not apply, so it'll never happen while NASA holds the keys to low earth orbit.
From my perspective, the difference between the "experimental" kernels and the "stable" kernels lies in a difference in direction and emphasis. With the "stable" kernels, changes are usually more about fixing defects than about changing or adding features. With "experimental" kernels, the emphasis is more on changing or adding features with less concern for the people who might try to use the kernel to do real work. However, that doesn't mean that a "stable" kernel will work, for fixes may beget other problems and it doesn't mean that an "unstable" kernel might not be solid as a rock, for the intent is to make the new features work well.
Early "stable" releases are probably best approached with caution, and not just because the kernel itself may be unstable. In my case, the problems I ran into with 2.6.0-testN kernels had to do with the fact that there's some pretty big changes in the way things are supposed to be done so a lot of the supporting software had to be fixed. (See, for example, some of the defect logs for Debian's initrd-tools package, where my email address appears a few times as I reported problems so that Mr. Xu could fix them.) Once I got the kernels installed and bootable, I didn't have issues with stability. Not at the house.
On the other hand, the SCSI driver from 2.6.1 refused to function at all on my work machine, but it did work on similar hardware at home. I never nailed that problem down because 2.6.2 appears to work on both machines.
This is why I can't take the IPV6 folks seriously. Demand for addresses comes from the leaves, not the root. So what if every backbone provider has native IPv6 routing throughout they're backbones? They're not the ones who use addresses by the ton!
I've got an IPv6 tunnel and addresses from a TLA, but I can't get native IPv6 access because neither the cable modem that I use nor the equipment it talks to upstream knows anything about IPv6. In fact, there is little, if any, end user WAN equipment that speaks IPv6 natively. Availability of that kind of equipment is necessary before a "global service launch" has any kind of meaning.
In defense of the other side, I point out that heat transfer away from the surface of the earth relies more on convection, which is not affected by the quantity of carbon dioxide in the atmosphere, than on radiation, which is where the "greenhouse effect" comes in. I also point out that water molecules are, on a molecule-by-molecule basis, at least as efficient at blocking infrared radiation as carbon dioxide, and that there are two orders of magnitude more of them in a typical sample of air than there are of carbon dioxide molecules. That means that the most important greenhouse "gas" in the atmosphere isn't a gas at all, it's water vapor. Indeed, that can be seen in the recorded experiences of people in the desert from the Roman legions onward.
So, why is a trace element supposed to cause the bulk of the effect? Perhaps there is a simple explanation. Do you know what it is?
What the Wright Brothers did do is build the first successful, controllable airplane. The controllability is the key because they were the first folks to really work out how to make an airplane go where you want it to go. They also figured out that it was going to take some practice for the pilot to become proficient in flying it. They also built propellors whose efficiency wasn't bettered for decades and along the way they laid the foundation of the whole theory of propellors.
In fact, like the telephone, the airplane is a perfect example of one of those things whose creation is inevitable once the supporting technology is available. There were many, many folks working on the solution to powered flight once small and lightweight engines were available to power the craft. The groundwork had been laid more than a century before with Cayley's conceptual leaps all it took was somebody to work out the details perhaps with a leap or two of their own.
As a practical matter, history records that the aileron was invented by Glenn Curtiss in an attempt to get around the Wright patent on the airplane. History also records that it's not that difficult to get a newspaper reporter to write a story even if it's only printed in one paper. When people put forth the claim that the Wrights built a successful flying machine and the date on which it was done, they produce a photograph of their machine flying and a dated telegram with the details of the flights.
On the Website talking Mr. Pearse's claim, there is nothing of the sort. The lack of evidence that the machine flew is explained with "he didn't realize the historic importance of the flights". What crap! Flight had been a human dream for thousands of years. Wouldn't fulfilling that dream seem to you to be of some historic importance? Shouldn't it have occured to one of the numerous witnesses to mention something to somebody or to write it in a diary or something? Everyone else working on heavier than air flight seemed to realize they were solving a momentous problem, why didn't anyone in Waitohi, New Zealand?
You should always check the return value from any functions in the scanf family against the number of items whose value you want to set. That will tell you whether or not you've read everything you expected. Also, bear in mind that it can be tricky to set up the format string properly. I can't think of any examples of nonobvious behavior off the top of my head, but I find I use a lot character sets when I use scanf.
Actually, most of the increase in user base recently is from people who don't contribute software. That means that I need Linux's popularity to be a tiny fraction of what it is in order to be happy with it. That's why I was happy with it even back in 1992.
I am fortunate that my employers think Linux is worth using, and it's nice for Linux to have enough popularity that the software tools I need get developed, but they would be there even if the Linux usage had peaked in 1992. Maybe there will be a revolution. Perhaps it's even under way right now. Who knows? I'm not in it for your revolution. It isn't important to me.
If the FSF folks are, like you, so confident that their way is the only right way to do things, then why do they care about what other people do? What difference does it make if you're going to "win" in the end? If you're really right, then everyone else will eventually realize it and do what you do of their own volition.
Morning for every time zone. Meteors are most likely to enter the atmosphere near where the time is just before sunrise.
If I were to ask the question, I wouldn't be interested in Macintosh compatability. Can you give me a quick link to somewhere I can find retailers for those PowerPC motherboards? A few google searches didn't result in anything like some place I could buy one of these.
The first step in the process is to find out the rules that apply. That takes some research. Speaking with an attorney that specializes in communications is one way of doing the research. Alternatively, you can get copies of the rule books. You're looking for CFR Title 47 as well as the relevant FAA regulations if you're going to have a tower structure. These are available from the Government Printing Office Web Site They're also available at many libraries.
Frankly, I can't see the point of just breaking the law in public. In my opinion, a better course of action would be to set up the conditions for a test case that could be won in the courts. That will likely require some public lawbreaking, but will also require there be something about the lawbreaking that demonstrates how the law in question isn't reasonable. Simply showing how easy it is to violate said law isn't going to do that.
However, I now do some things with CVS, even if I'm the only programmer working on a project becaues it means that I can work on that project on any computer I happen to be using without worrying (much) about which computer I last used to work on it.
For things I don't use CVS for, I use "make source" which I have set up to create a version-tagged zip file of the source in the current directory tree. (Crude, but effective, inexpensive, and quite easy to do. I also put all those zip files, including the old revisions, on the installation CDROM something that's already saved my posterior twice.)
Now, please note that, for the case in question, I did not go to the doctor in order to get some sort of magic pill. I was trying to determine whether or not I'm going to get better without treatment. It is the doctor that assumed that I want some sort of magic pill. I suppose that it's too much trouble to explain that it's really best to let some ailments run their course and then to ask me what sort of treatment I would consider appropriate. I certainly would choose to not take medication that isn't likely to help, and I'm not alone in this view.
Anyway, of those four diagnoses, three were wrong. Based on that, it sure doesn't sound like diagnoses are easy to me! Add to that the fact that I'm pretty good at troubleshooting and I'm one of the few that I know of. Most people flail about trying things at random and, as far as I know, training isn't much help for most of those people. Yes, it's easy to memorize a few pat answers to the most common problems, which is why many people who visit the "doctor" wind up seeing a PA, who forwards to the doctor only those cases whose diagnosis is in doubt, but that's exactly why this sort of thing is important. As time goes on, doctors are going to be less and less likely to see the simple cases.
You mention psychiatric diagnosis, so I'll talk about those. A quick check of my local DMDA chapter shows that some 70% of those suffering from some serious mental illness were misdiagnosed at least once. I can't help but think that a computer program that prompts the asking questions about typical symptoms of mania and schizophrenia would reduce that because most of the misdiagnoses start as a diagnosis of depression because it's what people complain about. I know the doctors don't ask those questions because in the sample that I have (8 so far) none have asked the right questions to make what we (that is, myself and the ill person) now believe is the correct diagnosis.
In any case, since visiting a doctor (and I spend a lot of my time in doctors' waiting rooms so I know this quite well) is something like an hour waiting to see the doctor followed by maybe 10 minutes of answering questions I don't understand, filling out the forms while I wait can't do me any harm even if the diagnosis is not remotely in doubt or irrelevant to the treatment, can it? I mean, it becomes part of the patient history just like the temperature and blood pressure check you're going to do whether I come in with a fever or with a splinter, right?
Heck, I suggest you put terminals tied into that database in the ER waiting room so there'll be something to do while you're waiting the 4-6 hours (on average) it'll take to get to the head of the line.
Packet radio and APRS are the results of what I'm talking about here. I get lots of "wow, don't you think this is cool?" from the ham gadget freaks, and lots of "why would I care?" from everybody else.
The problem with the current system is that copyright law serves to un-publish works. You can't make copies of a work because it's protected by copyright, but it's out of print because the publisher (and, make no mistake, the publisher has all of the rights in the current system) has decided that the market for that work isn't large enough to be able to keep it in print.
Working against that are examples of very long-lived works such as To Kill A Mockingbird or "Peter Pan". The vast majority of all works don't have nearly the longevity of "Peter Pan". So, extending the copyright for all of those works simply because one of those works might be continuously in print for many decades leads to works that are long forgotten. We are basically destroying our cultural heritage by not allowing works to fall into the public domain.
The suggestion I once made, that copyright holders be required to periodically pay for copyright in order to keep it, was met with cries that I was trying to prevent people from becoming authors and that I was trying to starve current authors' grandchildren. This idea would allow works like "Peter Pan" to stay in copyright basically forever, (meaning "until the fees aren't paid,") and still become public domain as soon as interest flags sufficiently for it to fall out of print.
The only part of my idea that I have trouble defending is the fact that I'm spending a lot of time thinking about making sure that people can still have access to works that they didn't care enough about when they were first published to keep them in print. Why should anyone care about a zillion B-movies that no one is ever going to re-release?
The 100m limit is there in order to make it possible to do collision detection on half-duplex links. That's also why Ethernet has a minimum packet size. If you're running full-duplex links, you can run the cable much farther because there'll never be any collisions. I don't know what the limit is with CAT-5, but you're electrically limited if you're running full-duplex. I would expect several kilometers, at the very least.
The fact is, the farther you get from a desktop computer, (and, at least with Linux, the farther you get from X86 PC's, or so I've been told,) the more trouble you're going to run into trying to use a desktop operating system in an embedded device. Most embedded devices don't look very much like the general-purpose computers most people are used to.
Of course, the other experiences I've had with commercial RTOSes is that they require considerable fiddling just to get them to start and start your application. Also, remember that WinCE isn't really considered an embedded operating system by many embedded people. Something like psOS or Lynx or QNX would be more like what's considered a "real" embedded OS.
When I looked at MIMEDefang for use on the mailing-list computer I'm setting up for a friend, it appeared to be sendmail-only. Is that true?
Once you're there, you're probably going to want to replace those handwritten queries with a friendlier interface, so you'll be looking at developing applications that use those queries internally. However, giving specific advice about that is hard because I don't know what kinds of applications you'd be interested in doing.
Microsoft has had products initially fail any number of times, but it's not clear that those products were new entries to the market. Further, Microsoft did not always back development of those products until they were a success. Lots of people talk about Bob, but do you remember QuickPascal? How about the MSX standard? I haven't been able to find very many "Natural Keyboards" recently.
Yes, with Windows and with their general office software, they initially had difficulties which they overcame, but it's not clear that those were entries into a new market for them. For example, Windows can be considered an operating system and Microsoft was already in the operating system business, so it's possible to argue that they weren't entering a market when they released Windows 1.0.
As near as I can tell, Microsoft has little or no track record of entering into new markets so I don't think one can claim that they have a history of backing, in the long term, entries into new markets that do not have a lot of initial success. They seem to do a pretty good job of picking the battles to fight that they actually need to win in order to maintain their market domination, but it's not now clear that the X-Box will be viewed as such a strategic product by the powers that be at Microsoft.
It just so happens that many of these sorts of simulations have runtimes that exceed the mean time between failures of the hardware they run on. I have been told that figuring out how to deal with that (and other failures, like kernel panics and whatnot) along with ancillary decisions (like whether or not it would be better to store an intermediate result or re-calculate it) occupy an awful lot of the time of those who write such things.
There doesn't seem to be an awful lot of off-the-shelf application software for clusters. This is probably because of the nature of the problems these clusters get used to solve. That, and the fact that software development for a nontrivial application usually takes quite a bit of time, means that the amount of effort required to put the application together is going to swamp the amount of time required to set up even the Beowulf option.
Further, given the vagaries of software development, a showstopper that isn't immediately obvious to someone who has read the documentation isn't likely to be found until well into the development cycle. That means that if there are lurking problems, a complete scrap-and-restart cycle is going to be needed after a lot of resources have already been committed. That encourages a close reading of the documentation for the base software and other, more general, clustering documentation, and a good look at more general concepts in the design of software that is intended to be clustered in hopes of finding potential pitfalls before time and money is spent on setting anything up.
Frankly, I don't think it would be a good idea to go out and buy (or acquire through some other means) a collection of Macintoshes and set it up as a Pooch-based cluster without doing quite a bit of reading on the subject including everything you can get your hands on about all of the available options.
Not to mention the fact that it hasn't been demonstrated to my satisfaction that it is easier to set up a Pooch-based cluster than a Beowulf cluster. Frankly, I think a lot of people would spend a lot more than an afternoon setting up the initial requirement for Pooch, a collection of networked Macs, if they didn't have one already. Of course, it's not any less difficult to set up a network of Linux machines, but it seems to me that putting the machines in place, making sure they all work, configuring the network settings, and terminating, running, and testing all the cables is going to easily consume an afternoon for all but the smallest purpose-built clusters.
One last thing, kind of off the point. While I generally agree with the statement that Free Software is only free of cost if you're not paying for people's time, my time is worth exactly what someone else is willing to pay for it, and for much of the day that is exactly nothing. That means I can gain all kinds of knowledge for absolute free and then apply that knowledge to what I get paid to do and reduce their costs because I can afford to do it on my own time rather than theirs.