If you look in the reference books, you'll see something curious about glucose: It is listed as having two forms, a chain and a ring. The ring form does look like benzene, but it doesn't have any double bonds. For some reason, the 6-C ring with an H and an OH on each C is treated nearly identically as the chain form (which obviously needs two more Hs). They obviously aren't exactly the same chemically, but apparently they're close enough that our biochemistry sees them as interchangable.
I remember being somewhat startled and puzzled the first time I noticed this. I think maybe it's some sort of subtle joke that the universe is playing on us. The cosmos is just messing with the heads of students in their organic chem classes.
In any case, a double bond between two C atoms is less stable than a single bond, and molecules with C=C bonds behave a lot differently from the version with a C-C bond. As soon as you get even one C=C bond, you get something that you probably don't want to snack on without first checking out exactly what it'll do to you.
I've also seen a comment that many of the bigger poly sugars do the same sort of thing. They're usually drawn as a chain. But apparently most of them can lose an H from each end and replace them with a C-C bond, and metabolically they're the same. Later on, the enzymes come along that chop up the complex sugars, and they don't seem to care if the sugar is a ring. They just chop at the right place, and get simpler sugars. In the case of the ring form of glucose, they can chop any place, since it doesn't matter.
One problem that I've noticed is the fact the build from source people tend to install things in a way that's completely different than anyone else.
While I'd agree with you in general, I've found one curious case where I've learned to install from the source to make all my machines the same: apache.
For some reason, every vendor (and sometimes every release;-) seems to have apache installed in a clever way that's different from everyone else. They put the pieces in different directories; they munge the config files in gratuitous ways; they even change what a user's public_html directory is called. In a mixed network, figuring out where the web server is hidden on a disk can be a real nightmare.
So I just grab the latest stable source kit from apache.org, and compile it. That takes maybe 10 minutes on current hardware. I spend 5 minutes or so munging httpd.conf, changing only what I know has to be changed. I get as close to a default install in/usr/local/apache as I can. I run the command it tells me to run to fire up the server. If it works, I copy that command to wherever it belongs in some boot script. Total time, usually about 20 minutes, far less that what I'd waste repeatedly trying to figure out the installation on a collection of machines.
I had a fun case a few years back on a project where the management had decreed a Netscape web server. Nobody could get it to run right. The usual reason was that it was installed and managed entirely through a web interface. Sounds fine, right? Yeah, until someone misconfigged something slightly, and the web server was a zombie. Now the managment interface is dead, and all you can do is reinstall from scratch.
One day I said "The hell with this", grabbed the current kit from apache.org, and 20 minutes later I had a live server. The developers could go crazy building their site.
Occasionally, I'd say "You know, we really should work on that Netscape server that we're supposed to be running." The reaction was always along the lines of "Yeah; we should, but in the meantime apache is running fine and we have stuff to do today." I talked to them recently, and they're still running apache. They also have an unofficial policy of always installing it from source, because they've been frustrated by the packaged installs that they find on linux distros.
The apache gang has done a fine job of making an install from source fast and painless. In such cases, a package usually just makes your life more difficult. If you take the defaults whenever possible, you end up in a better situation than with most packages. All your installations for all vendors come out the same, and managing a lot of machines is very easy.
Re:Apply Shrinkage to the Buckyballs
on
Buckyballs Kill Fish
·
· Score: 4, Informative
Suppose you reduce the buckyball to the minimal number of C atoms, 1. You'd then get things like CO2, CH4, CH3OH, etc. None of these is biologically inert, though some are more active than others.
Or don't go to such an extreme. Reduce the BB to just a single C ring, say to one of the 6 C atom. If each takes on one water molecule, you get an H and an OH attached to each atom. This is a form of glucose, which is also biologically active.
If you take a piece of a BB that is one hex ring and an adjacent penta ring, and attach simple radicals to the dangling bonds, you get all sorts of interesting molecules, most of which are biologically active.
In general, clumps of C atoms smaller than a buckyball are rarely biologically inert. They have dangling or unstable bonds that interact with nearby molecules.
If you want to convert fullerenes to an inactive form, you need to make them much larger. Then they start to look locally like graphite. But graphite, while stable, isn't inert. Google for "graphite" and "catalyst", and you'll learn a lot about the subject. Graphite is a very common industrial catalyst, with small amounts of various atoms or small molecules attached to the C atoms.
One way I've seen this explained for non-physicists is to notice that in all these multi-carbon forms, each 6-C ring has three single and three double bonds. A double bond is less stable than a single bond (and a triple bond even less stable). So the C atoms on each end of a double bond are likely to break one of those bonds, and bond instead to passing atoms or molecules. Often the difference in bond strength isn't large, so it's easy for other passing molecules to steal away the attached clump of atoms, and the C then reverts to the double bond.
This is a "biochemistry for dummies" explanation of how carbon takes part in such a huge range of chemical reactions. But it gets across the idea that, when you see a ring of carbon atoms with a few double bonds, you are looking at a diagram of a molecule that is likely to interact with many other molecules in its vicinity. The underlying C ring will probably be fairly stable, but it has excess electron bonds that want to connect to something.
Well, maybe, but calling these asteroids moons of Earth is basically silly. This new one has an orbit that goes out past Mars, and merely passes within a few million km of the Earth's orbit at one point. Cruithne's orbit extends from about Mars' orbit into near Mercury's orbit, and doesn't ever really get close to Earth. Both of these orbits are significantly inclined to the ecliptic.
Suggesting that they are somehow related to Earth is basically a sign that the writers are utterly clueless. Next we're going to read that Mars is a moon of Earth.
Can't we get reports that are bit more informed than this?
Some years back, I read an interesting historical example of this approach. The topic was the ways in which, during the early 40's, various countries conquered by the Nazis resisted them.
One example was the Danes. When the order came out that all Jew were to wear distinctive armband, the result was that most of the Danish population started wearing those armbands. They figured, rightly, that the Germans would understand what was happening, and wouldn't kill the entire Danish population who declared themselves Jews.
In contrast to this was the Bulgarians. Their approach was to cheerfully agree with the demand that they turn in all the Jews. The Bulgarians took this as an opportunity to finger everyone who they thought might be a collaborator. The Germans quickly eradicated most of their friends in the country, and except for a few cities, the country was essentially ungovernable.
There is a history of this in the US. Every witch hunt that we've had has mostly caught non-witches, because people figure out pretty quickly that this is a good way to make life difficult for people that they don't like. The actual subversives (whoever they might be that decade) have been hardly affected, of course.
This attempt to get companies to turn in terrorists will have the same effect. Is there someone you don't like, or who is a serious competitor at work? Just get them reported to the feds as a potential terrorist. Then sit back and wait.
This reminds me of a comment that I read some years ago about programmer productivity. Whoever it was (I've forgotten) remarked that older programmers usually take just as long to build something as novice programmers. The difference was that the experienced programmers produced much better software, that handled more than the obvious cases and didn't crash when handed bizarre input.
Part of this was the observation that if you could walk through a company's software development areas, you could quickly get a good idea of the quality of their software without talking to anyone. Just look at the ages of the programmers. Lots of young people meant software that didn't run well in your (bizarre, unusual) situation. Lots of older programmers meant software that ran well, though it might not be up to current commercial glitz standards. Best was a wide range of ages, so their attitudes and experience could rub off on each other.
Myself, I was a bit skeptical. I've generally found that age itself doesn't mean that much. There are plenty of 20-year-old fuddy-duddies, and there are plenty of radical, innovative 65-year-olds. But it was something to think about.
If the customer can't do the acceptance test, who can?
I've found myself on the developer side of this on numerous projects. The other side was either the QA people or actual customers. What I've done that works pretty well is: I code up tests for everything I can think of. I stick them in a directory called "test", and run everything there (in alphabetical order;-) as a regression test after a change seems to be working.
When the QA or customer rep makes the usual complaints about how complex and poorly defined their job is, I introduce them to my test suite. They are usually overjoyed that I've done their job for them. Then I tell them that I don't think I've done their job. My test suite is almost certainly incomplete. I give them a list of the things that I have tests for, and tell them to go off and think about it. I expect them to make suggestions for new tests.
Their suggestions are sometimes good. Other times they are vague and unprogrammable. So I try to get them to clarify. And so on. It's somewhat random, but usually a lot of useful ideas come out of it.
Eventually they have not just a "product", but a set of validation tests that show what it can do. And the contents of the test directory constitute a HOWTO for their own programmers stuck with the task of using the stuff.
The one major problem with this is that it's difficult to program tests for features that require user interaction. Simulating a human is not easy. If that human is hidden behind a WIMP interface, it's nearly impossible, no matter what people will tell you when they don't have to do it themselves.
[A]ll you have to do is register your copyright *before* you sue.
Yeah; that's what it looks like. If so, I have another modest proposal for a business plan: Write a lot of code that does an assortment of basic, obvious tasks in a very generic fashion. Put a copyright notice in the code. Don't publish any of it anywhere. Watch the Open-Source archives. When you see something that is a good match for any of your code fragments, quickly register your copyright, and file a lawsuit against the Open-Source "infringers".
If I can have a valid copyright on secret text, you have no defense at all against this. Furthermore, if I were even more unscrupulous, I wouldn't even bother writing the code. I'd take your code, make a few changes to variable names, back-date it, and register that. Faking an earlier date is fairly easy inside computers.
Of course, I may have a problem here, in that SCO has probably registered a patent on this business model...
With a few minor exceptions, you cannot sue in court until you have registered your copyright with the USCO.
Anyone know if this is really true? If so, it could have an interesting impact on the SCO charges of ingringing code in linux.
The reason is that, as part of the registration, you do have to give the USCO a copy of your copyrighted text, right?
The big problem the linux crowd has with the SCO charge is that SCO's code is secret. But if they've registered the copyright, then the USCO should have a copy of their code in some vault somewhere. We can just subpoena that, run a match program, and recode any close matches.
The really big problem with current copyright laws is that something can be unavailable to writers and still protected by copyright. So when you write something, you can't possibly know whether it's infriinging someone's copyright. All you can do is publish your work, and wait to see whether someone sues you.
You'd think this would be a clear violation of the US Constitutions "progress of the useful arts and sciences" idea, since it obvious puts fear of prosecution into the mind of anyone writing what they think is original text. But apparently not.
Still, it would be useful to know whether the USCO has a copy of SCO's copyrighted unix code. If not, is the code actually protected from infringement by people who can't see it? Or is it really possible to register a copyright on a secret text?
Suppose these poems this guy wrote described in intimate detail what it's like to have sex with his wife.... And if he does this, what would get it into public view?
Hey, ask Paris Hilton about that!;-)
Though we might note that her owning the copyright hasn't been too effective at blocking distribution. But chances are that she could get a pretty good settlement for "actual damages" if she presses infringement charges.
OTOH, the court might observe that she is a professional actor, and as such, the term is "publicity", not "damages". Then the judge would order her to pay distribution fees.
Dunno about tinfoil hats, but you might want to get in the habit of carefully checking the receipt before you pay, and also get a lot more careful about checking all the charges on your charge cards.
One of the properties of RFIDs is that they can be read at a distance of a meter or so. This includes not just the RFID tags in your shopping cart; it also includes all the RFID tags hidden inside anything you're wearing or carrying.
So as you go though the checkout line, the reader will be reading the RFID of every item of clothing you're wearing. If any of them match something that the store sells, there's a really good chance that they'll attempt to charge you for it.
Not doing this is a bit of technical challenge. One of the failure modes of RFID readers is that they will read any tag within range, and they can't tell whether it's an item you own or an item you're buying. Even if their records show that you bought that item recently, how can the computer tell whether it's the one you bought or a new one? They're not going to have a unique id for every sock ever sold.
This will probably get sorted out eventually, somehow. But for at least a decade or so, we can expect a lot of attempts by stores to charge you for things you already own.
Also, keep all your receipts. Demand that the receipt show an exact description of the items you're buying. And don't be shy about taking them to small claims court if they give you a hassle over the issue. In many cases, they won't be doing this innocently. If they think they can get away with it, well, hey, it's free income, right?
Myself, I think I'll try to find out which stores have RFID readers, and not shop there for the next 10 years or so while they get this figured out. Many years ago, I got stopped for shoplifting, for an item (a cap that I had in my pocket when I went in, and put on my head as I walked out into the weather) that I'd bought there a couple of months earlier. I don't want to go through that experience again. And with RFID, it's inevitable.
No, I want a way to discover that it's your work, and ask you if I can use it (possibly paying you a royalty). Under the current system, this information is usually impossible to find.
Case in point: I've sent letters to a number of music publishers saying that I have a tune in my mind, and I'd like to know if it's a tune that they own the copyright to. I ask them how I can discover if they own a particular tune.
So far, the answer from every one of them is that they'd be happy to sell me copies of all their music publications.
That's right; the only way I have, according to the publishers, is to purchase a copy of everything they publish and search it. This could easily cost several million dollars, and require a large warehouse to hold the books. The publishers consider this a reasonable answer to the question. And note that this isn't even possible with books that are out of print but still covered by copyright, since you usually can't buy them at all.
Actually, of course, there is one effective way that a musician can discover if a tune is copyrighted. You start performing it in public, or start selling a recording. If it is copyrighted, you'll be hit with a lawsuit; if not, you won't. (Well, maybe you will anyway; fraudulent copyright claims are common.;-)
As far as I can tell, this is the only practical way to discover if a tune is copyrighted.
It would be really useful if there were a registry where I could submit a tune and be told whether it's a close match to anything in the registry. It would be especially useful if the registry could give me contact info for the copyright holder. We have the technology to do this now, but there's no such registry.
I don't see how unconditional copyright creates a violation of free speech.
Well, under the current rules, this text is copyrighted as soon as I hit the Submit button. However, I don't know whether I am the copyright holder. It's possible (and on/. quite likely;-) that someone else has "published" the same words previously, and they are the copyright holder.
So I may be the copyright holder, or I may be a criminal engaged in copyright iinfringement. How can I know which I am? Right - I can't. There is no way that I can possibly know whether someone already owns the copyright to these words. This is because they don't have to register their copyright, and they don't have to publish there words any place that I can find them.
So I have two choices: I can take the risk of infringing someone else's copyright. Or I can keep quiet and not worry.
This seems as clear a violation of free speech as you could possibly have. Any words that I "publish" entail a serious threat of civil and/or criminal charges for copyright infringement, and my only way to avoid this is to say nothing.
Let's see; should I hit the Submit button and take the risk? Ah, why the hell not?...
Actually, the center of mass of the earth-moon system is about 1000 km down from the Earth's surface. But it's much closer to the surface than to the center.
Despite this, many astronomers still classify the earth-moon system as a double planet rather than as a primary+satellite. This is partly because, as Alan Stern argues, they basically do use the self-gravity rule to define "planet". Another line of reasoning is that the moon's orbit is everywhere concave to the sun, so technically it isn't orbiting the earth. Rather, both are in the same orbit about the sun, and are doing the "orbital dance" that two bodies in a common orbit do.
In any case, words like "planet" are human concepts. The universe doesn't have to supply objects that fit nicely into our classificational bins, and in this case, it doesn't.
Two individuals are generally recognized to be of the same species if they can mate and produce fertile offspring.
I remember back in high school I caused a bit of a fuss when the teacher came out with this definition. I held up my hand, and pointed out that, according to that definition, he and I were not the same species. (It may not be obvious in this forum, but I'm male.;-)
Funny thing was that he was flustered for a bit, and didn't quite know how to answer. He obviously hadn't ever thought about it, and really hadn't noticed that this definition misses something really important.
Myself, I like the astronomer's daughter's reasoning. Similarly, most of the supposed countries in the world shouldn't be considered countries at all. Who can remember them all? They just cause problems for school children who have to deal with tests that ask about them. Only countries that I remember should be allowed to exist.
I've discovered that paying off your balance is the last thing you want to do to build a good credit history... our financial analyst said if we'd paid off all our credit cards, our credit rating would have gone down instead of up.
Right. A few years ago, the news got out here in the US that the credit industry has a term for people who pay off their credit cards every month: deadbeats.
This is because when you do that, you're "stealing" profits from the credit companies. They make their living by charging interest on credit loans, and they don't much like people who don't take out loans by running a credit balance.
It's interesting what qualifies as "theft" these days. We've had a few discussions here of the way the recording, movie and publishing industries use the term, applying it to cases where the theft victim still has everything that they did before the claimed theft. And the credit industry does the same thing. They consider that interest charge as rightfully theirs. When you cheat the system by avoiding paying interest, you aren't keeping your own money; you are taking money out of the paychecks of hard-working corporate executives.
So much for the old advice to never a borrower or lender be. (Attributed to both Shakespeare and Ben Franklin. They probably both borrowed it from someone else.;-)
Yeah, getting a 5 billion dollar appropriation out of Congress... s no problem at all.
Hey, we folks here in the Boston area successfully got about three times that from Congress just to lower a few km of highway about 15 meters. Ask anyone from the area about the "Big Dig".
In comparison, getting a mere few billion to build practical, useful software should be easy.
Hmmm... except that Congress might not understand concepts like "practical" and "useful". I know; we can push it as the basis of new game-playing software.
Huh? Then you're totally off topic. Look back at the messages. The top level is about Lawrence Lessig's recent comments to the effect that the "IP" story isn't about culture, but is about businesses using the growing idiocy of the copyright and patent laws to throttle innovation and cretivity. Someone commented about the finite limit to possible (copyrightable) tunes, so I facetiously proposed a business model based on copyrighting all possible tune fragments.
This is totally about law and crap. The crap is, of course, the growing concept of making money via IP lawsuits.
If you're not talking about that, you're off topic.
And yes, I do understand your musical comments. I play a lot of music like you described. I'd bet that a lot of others here do, too. But, interesting as this is, it isn't really relevant to a discussion of businesses using copyright and patent law to stifle creativity.
I wonder if it's possible to get a discussion of music theory going here on/.? If you can think of a way, try to submit a story on the topic.
All very true, and legally irrelevant. If you were to, say, take a major-mode tune, convert it to minor, and claim it's original, there's not a court in the world that would find you not guilty of copyright infringement. Similarly, replacing a single note with a repeated note, or vice versa, or changing the length of notes (especially at the end of a phrase ) would not produce a new tune in anyone's mind.
All those things may be musically very important. But small changes don't make for a new tune, and won't get you out of a copyright infringement charge. You have to make significant changes, so that the tunes don't sound similar to the ear of whatever judge you happen to end up with.
Just ask George Harrison. Oh, wait, we can't...;-)
... given that there exist a provably finite number of melodies,...
OK, here's a business model: First, find out what the courts have decided is too short a melody fragment to be copyrightable. This probably isn't quite the same in every jurisdiction, but it's likely under 10 notes. Let this be N. For example, the distinguished first phrase of Happy Birthday is just 5 notes, all the same length, so N is at most 4.
Next, write a little program to generate all the tune fragments from N+1 to, say, 2*N notes. We probably can limit the melody to just an 8-note octave, or maybe make it 12 notes to be on the safe side. OTOH, we might want to include variants of the fragments with various notes of double length, to take into account rhythmic variations.
Apply for copyright on all of these fragments. Whether they all get processed and approved is irrelevant; you'll have applied so you can claim them.
When a new hit song comes out, compare it with your archive of melody fragments. It will match some of them. File suit for copyright infringement....
...purchasing a silent track isn't as ridiculous as it sounds (pun intended).
Back in the 1970's, there was a '45' put out that was several minutes of silence. Many bars and restaurants reported that it was the most-played selection on their juke box. Lots of people were willing to put out a nickel or dime or whatever to prevent any music being played for a few minutes.
I wonder if that "recording" had a copyright notice?
While the story may well be bogus, it's interesting to think of the implications of this publicity. The primary lesson that most managers will learn is "That's what happens when you trust software from American companies." So, whether it's intentional or not, the main effect of this story will be to hurt American companies' software sales.
It's interesting that this seems to be coming from American government sources at a time when the US government is more and more run by those big corporations.
In particular, this story probably casts a real shadow over Microsoft's sales outside the US. And MS is now one of George W's big contributors.
I wonder what's going on behind the scenes that permitted this to happen?
Whether bacteria in the rest of the Solar System can be traced to Earth is a bit tricky. If they can be identified as close relatives of Earthly bacteria, the conclusion is obvious. But since the other planets, comets, etc. don't have very Earth-like conditions, we would expect that anything actually living elsewhere would rapidly diverge from any Earth relatives. We then have the problem of whether their ancestors came from Earth, or from outside the Solar System.
And the "panspermia" hypothesis suggests the latter. The galaxy is roughly three times older than the Earth, so if life evolved on Earth, it had three times as long to evolve elsewhere. Those places have also been spraying their bacterial spores (thinly) across the galaxy. The idea then is that life may not have originated on Earth at all, but may have come here in the form of bacterial spores. They were in the cloud that condensed to form the Solar System, the proto-Earth had lots of them, and those that didn't get baked in the interior of the forming planet were the ancestors of our current biosphere.
Getting evidence one way or the other is not easy. Conclusive evidence will probably have to wait until we can collect enough samples of dust that's crossing the Solar System at speeds too great to be in orit about the Sun. Then, if the panspermia idea is correct, we'd predict that some tiny fraction of these dust particles will be (or contain) bacterial spores.
If we can take those spores apart and study their biochemistry, we may find some that are unlike anything on Earth. That would strongly support the hypothesis. Our kind of life evolved somewhere (maybe Earth, maybe not), other kinds evolved other places, and they're all drifting around just a-lookin' for a home.
On the other hand, if they are all based on the same RNA/DNA/amino-acid units as Earth life, things get trickier. Maybe there was only one origin of life in the galaxy. Maybe there's only one way that life can work. Maybe the Earth is the only planet with life, and we've picked up our own ancient bacteria on their tour of the galaxy.
Or maybe we won't find any viable spores on the extra-system dust particles, but a few or a lot of dead spore fragments. This would mean that they really can't survive millennia in space, and panspermia doesn't work.
It's going to be a while before we can collect the evidence.
Slashdot Mirror
If you look in the reference books, you'll see something curious about glucose: It is listed as having two forms, a chain and a ring. The ring form does look like benzene, but it doesn't have any double bonds. For some reason, the 6-C ring with an H and an OH on each C is treated nearly identically as the chain form (which obviously needs two more Hs). They obviously aren't exactly the same chemically, but apparently they're close enough that our biochemistry sees them as interchangable.
I remember being somewhat startled and puzzled the first time I noticed this. I think maybe it's some sort of subtle joke that the universe is playing on us. The cosmos is just messing with the heads of students in their organic chem classes.
In any case, a double bond between two C atoms is less stable than a single bond, and molecules with C=C bonds behave a lot differently from the version with a C-C bond. As soon as you get even one C=C bond, you get something that you probably don't want to snack on without first checking out exactly what it'll do to you.
I've also seen a comment that many of the bigger poly sugars do the same sort of thing. They're usually drawn as a chain. But apparently most of them can lose an H from each end and replace them with a C-C bond, and metabolically they're the same. Later on, the enzymes come along that chop up the complex sugars, and they don't seem to care if the sugar is a ring. They just chop at the right place, and get simpler sugars. In the case of the ring form of glucose, they can chop any place, since it doesn't matter.
One problem that I've noticed is the fact the build from source people tend to install things in a way that's completely different than anyone else.
;-) seems to have apache installed in a clever way that's different from everyone else. They put the pieces in different directories; they munge the config files in gratuitous ways; they even change what a user's public_html directory is called. In a mixed network, figuring out where the web server is hidden on a disk can be a real nightmare.
/usr/local/apache as I can. I run the command it tells me to run to fire up the server. If it works, I copy that command to wherever it belongs in some boot script. Total time, usually about 20 minutes, far less that what I'd waste repeatedly trying to figure out the installation on a collection of machines.
While I'd agree with you in general, I've found one curious case where I've learned to install from the source to make all my machines the same: apache.
For some reason, every vendor (and sometimes every release
So I just grab the latest stable source kit from apache.org, and compile it. That takes maybe 10 minutes on current hardware. I spend 5 minutes or so munging httpd.conf, changing only what I know has to be changed. I get as close to a default install in
I had a fun case a few years back on a project where the management had decreed a Netscape web server. Nobody could get it to run right. The usual reason was that it was installed and managed entirely through a web interface. Sounds fine, right? Yeah, until someone misconfigged something slightly, and the web server was a zombie. Now the managment interface is dead, and all you can do is reinstall from scratch.
One day I said "The hell with this", grabbed the current kit from apache.org, and 20 minutes later I had a live server. The developers could go crazy building their site.
Occasionally, I'd say "You know, we really should work on that Netscape server that we're supposed to be running." The reaction was always along the lines of "Yeah; we should, but in the meantime apache is running fine and we have stuff to do today." I talked to them recently, and they're still running apache. They also have an unofficial policy of always installing it from source, because they've been frustrated by the packaged installs that they find on linux distros.
The apache gang has done a fine job of making an install from source fast and painless. In such cases, a package usually just makes your life more difficult. If you take the defaults whenever possible, you end up in a better situation than with most packages. All your installations for all vendors come out the same, and managing a lot of machines is very easy.
Suppose you reduce the buckyball to the minimal number of C atoms, 1. You'd then get things like CO2, CH4, CH3OH, etc. None of these is biologically inert, though some are more active than others.
Or don't go to such an extreme. Reduce the BB to just a single C ring, say to one of the 6 C atom. If each takes on one water molecule, you get an H and an OH attached to each atom. This is a form of glucose, which is also biologically active.
If you take a piece of a BB that is one hex ring and an adjacent penta ring, and attach simple radicals to the dangling bonds, you get all sorts of interesting molecules, most of which are biologically active.
In general, clumps of C atoms smaller than a buckyball are rarely biologically inert. They have dangling or unstable bonds that interact with nearby molecules.
If you want to convert fullerenes to an inactive form, you need to make them much larger. Then they start to look locally like graphite. But graphite, while stable, isn't inert. Google for "graphite" and "catalyst", and you'll learn a lot about the subject. Graphite is a very common industrial catalyst, with small amounts of various atoms or small molecules attached to the C atoms.
One way I've seen this explained for non-physicists is to notice that in all these multi-carbon forms, each 6-C ring has three single and three double bonds. A double bond is less stable than a single bond (and a triple bond even less stable). So the C atoms on each end of a double bond are likely to break one of those bonds, and bond instead to passing atoms or molecules. Often the difference in bond strength isn't large, so it's easy for other passing molecules to steal away the attached clump of atoms, and the C then reverts to the double bond.
This is a "biochemistry for dummies" explanation of how carbon takes part in such a huge range of chemical reactions. But it gets across the idea that, when you see a ring of carbon atoms with a few double bonds, you are looking at a diagram of a molecule that is likely to interact with many other molecules in its vicinity. The underlying C ring will probably be fairly stable, but it has excess electron bonds that want to connect to something.
Well, maybe, but calling these asteroids moons of Earth is basically silly. This new one has an orbit that goes out past Mars, and merely passes within a few million km of the Earth's orbit at one point. Cruithne's orbit extends from about Mars' orbit into near Mercury's orbit, and doesn't ever really get close to Earth. Both of these orbits are significantly inclined to the ecliptic.
Suggesting that they are somehow related to Earth is basically a sign that the writers are utterly clueless. Next we're going to read that Mars is a moon of Earth.
Can't we get reports that are bit more informed than this?
Some years back, I read an interesting historical example of this approach. The topic was the ways in which, during the early 40's, various countries conquered by the Nazis resisted them.
One example was the Danes. When the order came out that all Jew were to wear distinctive armband, the result was that most of the Danish population started wearing those armbands. They figured, rightly, that the Germans would understand what was happening, and wouldn't kill the entire Danish population who declared themselves Jews.
In contrast to this was the Bulgarians. Their approach was to cheerfully agree with the demand that they turn in all the Jews. The Bulgarians took this as an opportunity to finger everyone who they thought might be a collaborator. The Germans quickly eradicated most of their friends in the country, and except for a few cities, the country was essentially ungovernable.
There is a history of this in the US. Every witch hunt that we've had has mostly caught non-witches, because people figure out pretty quickly that this is a good way to make life difficult for people that they don't like. The actual subversives (whoever they might be that decade) have been hardly affected, of course.
This attempt to get companies to turn in terrorists will have the same effect. Is there someone you don't like, or who is a serious competitor at work? Just get them reported to the feds as a potential terrorist. Then sit back and wait.
This reminds me of a comment that I read some years ago about programmer productivity. Whoever it was (I've forgotten) remarked that older programmers usually take just as long to build something as novice programmers. The difference was that the experienced programmers produced much better software, that handled more than the obvious cases and didn't crash when handed bizarre input.
Part of this was the observation that if you could walk through a company's software development areas, you could quickly get a good idea of the quality of their software without talking to anyone. Just look at the ages of the programmers. Lots of young people meant software that didn't run well in your (bizarre, unusual) situation. Lots of older programmers meant software that ran well, though it might not be up to current commercial glitz standards. Best was a wide range of ages, so their attitudes and experience could rub off on each other.
Myself, I was a bit skeptical. I've generally found that age itself doesn't mean that much. There are plenty of 20-year-old fuddy-duddies, and there are plenty of radical, innovative 65-year-olds. But it was something to think about.
If the customer can't do the acceptance test, who can?
;-) as a regression test after a change seems to be working.
I've found myself on the developer side of this on numerous projects. The other side was either the QA people or actual customers. What I've done that works pretty well is: I code up tests for everything I can think of. I stick them in a directory called "test", and run everything there (in alphabetical order
When the QA or customer rep makes the usual complaints about how complex and poorly defined their job is, I introduce them to my test suite. They are usually overjoyed that I've done their job for them. Then I tell them that I don't think I've done their job. My test suite is almost certainly incomplete. I give them a list of the things that I have tests for, and tell them to go off and think about it. I expect them to make suggestions for new tests.
Their suggestions are sometimes good. Other times they are vague and unprogrammable. So I try to get them to clarify. And so on. It's somewhat random, but usually a lot of useful ideas come out of it.
Eventually they have not just a "product", but a set of validation tests that show what it can do. And the contents of the test directory constitute a HOWTO for their own programmers stuck with the task of using the stuff.
The one major problem with this is that it's difficult to program tests for features that require user interaction. Simulating a human is not easy. If that human is hidden behind a WIMP interface, it's nearly impossible, no matter what people will tell you when they don't have to do it themselves.
[A]ll you have to do is register your copyright *before* you sue.
...
Yeah; that's what it looks like. If so, I have another modest proposal for a business plan: Write a lot of code that does an assortment of basic, obvious tasks in a very generic fashion. Put a copyright notice in the code. Don't publish any of it anywhere. Watch the Open-Source archives. When you see something that is a good match for any of your code fragments, quickly register your copyright, and file a lawsuit against the Open-Source "infringers".
If I can have a valid copyright on secret text, you have no defense at all against this. Furthermore, if I were even more unscrupulous, I wouldn't even bother writing the code. I'd take your code, make a few changes to variable names, back-date it, and register that. Faking an earlier date is fairly easy inside computers.
Of course, I may have a problem here, in that SCO has probably registered a patent on this business model
With a few minor exceptions, you cannot sue in court until you have registered your copyright with the USCO.
Anyone know if this is really true? If so, it could have an interesting impact on the SCO charges of ingringing code in linux.
The reason is that, as part of the registration, you do have to give the USCO a copy of your copyrighted text, right?
The big problem the linux crowd has with the SCO charge is that SCO's code is secret. But if they've registered the copyright, then the USCO should have a copy of their code in some vault somewhere. We can just subpoena that, run a match program, and recode any close matches.
The really big problem with current copyright laws is that something can be unavailable to writers and still protected by copyright. So when you write something, you can't possibly know whether it's infriinging someone's copyright. All you can do is publish your work, and wait to see whether someone sues you.
You'd think this would be a clear violation of the US Constitutions "progress of the useful arts and sciences" idea, since it obvious puts fear of prosecution into the mind of anyone writing what they think is original text. But apparently not.
Still, it would be useful to know whether the USCO has a copy of SCO's copyrighted unix code. If not, is the code actually protected from infringement by people who can't see it? Or is it really possible to register a copyright on a secret text?
Suppose these poems this guy wrote described in intimate detail what it's like to have sex with his wife. ... And if he does this, what would get it into public view?
;-)
Hey, ask Paris Hilton about that!
Though we might note that her owning the copyright hasn't been too effective at blocking distribution. But chances are that she could get a pretty good settlement for "actual damages" if she presses infringement charges.
OTOH, the court might observe that she is a professional actor, and as such, the term is "publicity", not "damages". Then the judge would order her to pay distribution fees.
Dunno about tinfoil hats, but you might want to get in the habit of carefully checking the receipt before you pay, and also get a lot more careful about checking all the charges on your charge cards.
One of the properties of RFIDs is that they can be read at a distance of a meter or so. This includes not just the RFID tags in your shopping cart; it also includes all the RFID tags hidden inside anything you're wearing or carrying.
So as you go though the checkout line, the reader will be reading the RFID of every item of clothing you're wearing. If any of them match something that the store sells, there's a really good chance that they'll attempt to charge you for it.
Not doing this is a bit of technical challenge. One of the failure modes of RFID readers is that they will read any tag within range, and they can't tell whether it's an item you own or an item you're buying. Even if their records show that you bought that item recently, how can the computer tell whether it's the one you bought or a new one? They're not going to have a unique id for every sock ever sold.
This will probably get sorted out eventually, somehow. But for at least a decade or so, we can expect a lot of attempts by stores to charge you for things you already own.
Also, keep all your receipts. Demand that the receipt show an exact description of the items you're buying. And don't be shy about taking them to small claims court if they give you a hassle over the issue. In many cases, they won't be doing this innocently. If they think they can get away with it, well, hey, it's free income, right?
Myself, I think I'll try to find out which stores have RFID readers, and not shop there for the next 10 years or so while they get this figured out. Many years ago, I got stopped for shoplifting, for an item (a cap that I had in my pocket when I went in, and put on my head as I walked out into the weather) that I'd bought there a couple of months earlier. I don't want to go through that experience again. And with RFID, it's inevitable.
I want your work, but I want it for free.
;-)
No, I want a way to discover that it's your work, and ask you if I can use it (possibly paying you a royalty). Under the current system, this information is usually impossible to find.
Case in point: I've sent letters to a number of music publishers saying that I have a tune in my mind, and I'd like to know if it's a tune that they own the copyright to. I ask them how I can discover if they own a particular tune.
So far, the answer from every one of them is that they'd be happy to sell me copies of all their music publications.
That's right; the only way I have, according to the publishers, is to purchase a copy of everything they publish and search it. This could easily cost several million dollars, and require a large warehouse to hold the books. The publishers consider this a reasonable answer to the question. And note that this isn't even possible with books that are out of print but still covered by copyright, since you usually can't buy them at all.
Actually, of course, there is one effective way that a musician can discover if a tune is copyrighted. You start performing it in public, or start selling a recording. If it is copyrighted, you'll be hit with a lawsuit; if not, you won't. (Well, maybe you will anyway; fraudulent copyright claims are common.
As far as I can tell, this is the only practical way to discover if a tune is copyrighted.
It would be really useful if there were a registry where I could submit a tune and be told whether it's a close match to anything in the registry. It would be especially useful if the registry could give me contact info for the copyright holder. We have the technology to do this now, but there's no such registry.
I don't see how unconditional copyright creates a violation of free speech.
/. quite likely ;-) that someone else has "published" the same words previously, and they are the copyright holder.
...
Well, under the current rules, this text is copyrighted as soon as I hit the Submit button. However, I don't know whether I am the copyright holder. It's possible (and on
So I may be the copyright holder, or I may be a criminal engaged in copyright iinfringement. How can I know which I am? Right - I can't. There is no way that I can possibly know whether someone already owns the copyright to these words. This is because they don't have to register their copyright, and they don't have to publish there words any place that I can find them.
So I have two choices: I can take the risk of infringing someone else's copyright. Or I can keep quiet and not worry.
This seems as clear a violation of free speech as you could possibly have. Any words that I "publish" entail a serious threat of civil and/or criminal charges for copyright infringement, and my only way to avoid this is to say nothing.
Let's see; should I hit the Submit button and take the risk? Ah, why the hell not?
Actually, the center of mass of the earth-moon system is about 1000 km down from the Earth's surface. But it's much closer to the surface than to the center.
Despite this, many astronomers still classify the earth-moon system as a double planet rather than as a primary+satellite. This is partly because, as Alan Stern argues, they basically do use the self-gravity rule to define "planet". Another line of reasoning is that the moon's orbit is everywhere concave to the sun, so technically it isn't orbiting the earth. Rather, both are in the same orbit about the sun, and are doing the "orbital dance" that two bodies in a common orbit do.
In any case, words like "planet" are human concepts. The universe doesn't have to supply objects that fit nicely into our classificational bins, and in this case, it doesn't.
Two individuals are generally recognized to be of the same species if they can mate and produce fertile offspring.
;-)
I remember back in high school I caused a bit of a fuss when the teacher came out with this definition. I held up my hand, and pointed out that, according to that definition, he and I were not the same species. (It may not be obvious in this forum, but I'm male.
Funny thing was that he was flustered for a bit, and didn't quite know how to answer. He obviously hadn't ever thought about it, and really hadn't noticed that this definition misses something really important.
Myself, I like the astronomer's daughter's reasoning. Similarly, most of the supposed countries in the world shouldn't be considered countries at all. Who can remember them all? They just cause problems for school children who have to deal with tests that ask about them. Only countries that I remember should be allowed to exist.
Why do people think that reporters are the "good guys" when they attempt to take a gun through airport security to prove how insecure things are?
Maybe because in such cases, the reporters are the good guys.
I'd much rather have them do this and report on the results than some criminal or terrorist do it.
Both will probably cause increased airport security. But the reporters won't kill anyone.
I've discovered that paying off your balance is the last thing you want to do to build a good credit history... our financial analyst said if we'd paid off all our credit cards, our credit rating would have gone down instead of up.
;-)
Right. A few years ago, the news got out here in the US that the credit industry has a term for people who pay off their credit cards every month: deadbeats.
This is because when you do that, you're "stealing" profits from the credit companies. They make their living by charging interest on credit loans, and they don't much like people who don't take out loans by running a credit balance.
It's interesting what qualifies as "theft" these days. We've had a few discussions here of the way the recording, movie and publishing industries use the term, applying it to cases where the theft victim still has everything that they did before the claimed theft. And the credit industry does the same thing. They consider that interest charge as rightfully theirs. When you cheat the system by avoiding paying interest, you aren't keeping your own money; you are taking money out of the paychecks of hard-working corporate executives.
So much for the old advice to never a borrower or lender be. (Attributed to both Shakespeare and Ben Franklin. They probably both borrowed it from someone else.
Yeah, getting a 5 billion dollar appropriation out of Congress ... s no problem at all.
... except that Congress might not understand concepts like "practical" and "useful". I know; we can push it as the basis of new game-playing software.
Hey, we folks here in the Boston area successfully got about three times that from Congress just to lower a few km of highway about 15 meters. Ask anyone from the area about the "Big Dig".
In comparison, getting a mere few billion to build practical, useful software should be easy.
Hmmm
Pages that require HTTP_REFERER are already broken.
;-)
(And why do I always read that as HTTP_REEFER?
I wasn't even talking about law and crap.
/.? If you can think of a way, try to submit a story on the topic.
Huh? Then you're totally off topic. Look back at the messages. The top level is about Lawrence Lessig's recent comments to the effect that the "IP" story isn't about culture, but is about businesses using the growing idiocy of the copyright and patent laws to throttle innovation and cretivity. Someone commented about the finite limit to possible (copyrightable) tunes, so I facetiously proposed a business model based on copyrighting all possible tune fragments.
This is totally about law and crap. The crap is, of course, the growing concept of making money via IP lawsuits.
If you're not talking about that, you're off topic.
And yes, I do understand your musical comments. I play a lot of music like you described. I'd bet that a lot of others here do, too. But, interesting as this is, it isn't really relevant to a discussion of businesses using copyright and patent law to stifle creativity.
I wonder if it's possible to get a discussion of music theory going here on
All very true, and legally irrelevant. If you were to, say, take a major-mode tune, convert it to minor, and claim it's original, there's not a court in the world that would find you not guilty of copyright infringement. Similarly, replacing a single note with a repeated note, or vice versa, or changing the length of notes (especially at the end of a phrase ) would not produce a new tune in anyone's mind.
... ;-)
All those things may be musically very important. But small changes don't make for a new tune, and won't get you out of a copyright infringement charge. You have to make significant changes, so that the tunes don't sound similar to the ear of whatever judge you happen to end up with.
Just ask George Harrison. Oh, wait, we can't
... given that there exist a provably finite number of melodies, ...
...
OK, here's a business model: First, find out what the courts have decided is too short a melody fragment to be copyrightable. This probably isn't quite the same in every jurisdiction, but it's likely under 10 notes. Let this be N. For example, the distinguished first phrase of Happy Birthday is just 5 notes, all the same length, so N is at most 4.
Next, write a little program to generate all the tune fragments from N+1 to, say, 2*N notes. We probably can limit the melody to just an 8-note octave, or maybe make it 12 notes to be on the safe side. OTOH, we might want to include variants of the fragments with various notes of double length, to take into account rhythmic variations.
Apply for copyright on all of these fragments. Whether they all get processed and approved is irrelevant; you'll have applied so you can claim them.
When a new hit song comes out, compare it with your archive of melody fragments. It will match some of them. File suit for copyright infringement.
Profit!
...purchasing a silent track isn't as ridiculous as it sounds (pun intended).
Back in the 1970's, there was a '45' put out that was several minutes of silence. Many bars and restaurants reported that it was the most-played selection on their juke box. Lots of people were willing to put out a nickel or dime or whatever to prevent any music being played for a few minutes.
I wonder if that "recording" had a copyright notice?
While the story may well be bogus, it's interesting to think of the implications of this publicity. The primary lesson that most managers will learn is "That's what happens when you trust software from American companies." So, whether it's intentional or not, the main effect of this story will be to hurt American companies' software sales.
It's interesting that this seems to be coming from American government sources at a time when the US government is more and more run by those big corporations.
In particular, this story probably casts a real shadow over Microsoft's sales outside the US. And MS is now one of George W's big contributors.
I wonder what's going on behind the scenes that permitted this to happen?
Whether bacteria in the rest of the Solar System can be traced to Earth is a bit tricky. If they can be identified as close relatives of Earthly bacteria, the conclusion is obvious. But since the other planets, comets, etc. don't have very Earth-like conditions, we would expect that anything actually living elsewhere would rapidly diverge from any Earth relatives. We then have the problem of whether their ancestors came from Earth, or from outside the Solar System.
And the "panspermia" hypothesis suggests the latter. The galaxy is roughly three times older than the Earth, so if life evolved on Earth, it had three times as long to evolve elsewhere. Those places have also been spraying their bacterial spores (thinly) across the galaxy. The idea then is that life may not have originated on Earth at all, but may have come here in the form of bacterial spores. They were in the cloud that condensed to form the Solar System, the proto-Earth had lots of them, and those that didn't get baked in the interior of the forming planet were the ancestors of our current biosphere.
Getting evidence one way or the other is not easy. Conclusive evidence will probably have to wait until we can collect enough samples of dust that's crossing the Solar System at speeds too great to be in orit about the Sun. Then, if the panspermia idea is correct, we'd predict that some tiny fraction of these dust particles will be (or contain) bacterial spores.
If we can take those spores apart and study their biochemistry, we may find some that are unlike anything on Earth. That would strongly support the hypothesis. Our kind of life evolved somewhere (maybe Earth, maybe not), other kinds evolved other places, and they're all drifting around just a-lookin' for a home.
On the other hand, if they are all based on the same RNA/DNA/amino-acid units as Earth life, things get trickier. Maybe there was only one origin of life in the galaxy. Maybe there's only one way that life can work. Maybe the Earth is the only planet with life, and we've picked up our own ancient bacteria on their tour of the galaxy.
Or maybe we won't find any viable spores on the extra-system dust particles, but a few or a lot of dead spore fragments. This would mean that they really can't survive millennia in space, and panspermia doesn't work.
It's going to be a while before we can collect the evidence.