I generally can't stand the reality TV show scheme; however, I happened to catch half an episode of the first Joe Schmoe season at a friend's house and was instantly hooked. I've now seen the whole season.
It was a remarkably humane and thoughtful program. It not only celebrated humanity at its best but managed to tackle complicated ethical questions with more depth and integrity than any other program on television. (Albeit with the occasional gratuitous tit flash.)
Although it's not clear to me the producers *intended* (and it sure as hell isn't the program that was advertised in commercial spots), they managed to create a something genuine and moving in a field dominated by the cruel and the stupid. The result is closer to Studs Terkel than to Survivor.
I know nothing about the astronaut show except the slashdot blurb (TFA won't cooperate with my browser at present), but there's certainly the potential there to create something which isn't mean spirited and cruel. Tone down the competition, hire people who aren't selfish and stupid, and it's possible to create something truly worthwhile in reality TV.
I do hope they eventually straighten participants out about the bad physics they've had to feed them, though. Pointing out to the audience that their gravity explanation is total bullocks would seem an ethical must as well.
>You can't use the non-gui >version of emacs? Why not? >If you know the keybindings, >emacs should be perfectly >usable on the console.
I'm always amazed by the number of unix savvy people who don't know that emacs will run on the console. I suspect it's the primary reason emacs has a reputation for being slow and clunky.
If I were in charge, I'd replace the "-nw" flag with an "-x" flag. Making the default behavior slow and ugly is a questionable strategy. (As it is, I run a non-X build of emacs at home, just for the petty and senseless pleasure of saving a few extra kB of ram.)
>to be fair, emacs keybindings >aren't easy to learn either
True. But, if you haven't learned them, there's the great advantage that you can almost always guess the name of the thing you want to do and cause it to happen anyway.
If you ask me, meta-x + function name (along with tab completion) is the second most useful and least advertised emacs feature. Being able to do things that you only need to do twice a year without having to look up some crazy key binding sure is nice. I haven't got a clue what the key binding for kill-rectangle or toggle-truncate-lines are, and probably never will. But I know those features exist and can easily cause them to happen every time I need them.
As an aside, I claim the very most useful and least advertised feature is the "-nw" flag. I've been astounded by the number of aged twenty-something unix savvy folks who've never heard of it and force themselves to learn VI because they don't want to run X apps over a slow network connection. (I'm not going to claim there aren't perfectly good reasons that a person would want to use VI for it's own sake. I'm talking strictly about the "I hate VI but I have no choice but to use it" crowd.)
Sounds like fun. (As an aside, I'm shocked by the number of people arguing against having after hours fun with such toys. Is the slashdot readership really so cowardly and unimaginative? Sure, one has to be careful and should avoid pissing off the bosses during the first few weeks at work. And it goes without saying that if the electron microscope happens to break while you're using it for personal projects, bad things are likely to happen to you, unless you have or happen to be a particularly cool boss. But, the risk may well be worth it.)
If you've got access to a scanning electron microscope, any sample should be fun. Around here (a multi-group academic facility) the machine is jealously guarded by a dedicated staff person and we get charged rather a lot of money for each use, so I haven't done any recreational microscopy. But, just looking at the stuff we're supposed to look at is overwhelmingly nifty. (Obviously you should stop and think before putting foreign objects into either the miscroscope itself or a sputtering chamber.)
With die cutting, ceramics, and electroplating, you could certainly make some beautiful cases for homebrew projects. If you go in for a retro look, try to cook up some faux-bakelite. (Or real bakelite, for that matter, if you can get your hands on the stuff.)
Another possibility would be tinkering with electrostatic levitation. Suspended objects are always neat.
You've also got the ingredients for making homebrew optics toys. With lapping and plating gear, you might be able to make your own optical quality mirrors for homebrew telescope parts / lasers / holography setups / etc. Anything else involving precision ground metal parts and custom ceramics is an obvious candidate: home made particle detectors / geiger muller tubes, for example.
And there's always the obvious option of making really big capacitors, charging them to really high voltages, and zapping things. (As described, for example, here http://www.amasci.com/amateur/capexpt.html )
>They are a subsidiary of the >Warner empire (as are their >record, film, and book divisions) >but they are not a record company.
Actually both Warner/Chappell Music and Warner Records are now part of a separate company, the Warner Music Group. They were sold off by the rest of the Warner empire a couple years ago.
Of course that doesn't affect the point of your post. Their publishing house is now just a small part of a less giant company that also owns recording companies. But, with only two hands left, it is perhaps even harder to understand why they aren't looking out for each other's interests.
>If some researcher needs to be in orbit >for some research, they pay NASA x dollars >for room and board on the station >(appropriately subsidized by the American >taxpayer--x may be zero).
In which case the researcher writes out a grant proposal to NASA/NSF/DOE and asks for money to be used in paying NASA for room and board? Doesn't seem particularly effective, except possibly as a way to force through manned spaceflight at the expense of other programs if funding is ever cut. (Asking non-US researchers to pay might work, although it isn't clear how much demand there would be for seats from nations that aren't already ISS partners.)
On the other hand, if the ISS were forced to live off of money tied to specific research projects, funding agencies would pretty quickly realize how very little research there is that requires placing humans in low earth orbit for months. The cost of experiments that we've conducted in the recent decades are only remotely reasonable if you ignore the billions of dollars spent subsidizing the shuttle and station. As a way of killing off the ISS, it might work.
That's not to say that I'm particularly opposed to the shuttle, the ISS, or manned spaceflight in general. But it is and for the forseable future will be a public works project rather than a scientific endeavor.
That's a fine thing in itself, but it won't work as a for-profit enterprise. The only financially worthwhile thing you can do with humans in low earth orbit is to charge rich tourists to take them there. Given a choice between sending up some useless astronauts to hang about taking pictures of cool weightless stuff and sending up some useless right guys to hang about taking pictures of cool weightless stuff, I'd choose the former. At least the cool pictures will be public domain afterward.
>we are talking about a >construction made of early >forms of cement - which was >made from straw and mud if I >am not mistaken.
I'm not expert, but all the pyramids I know of (in either Egypt or Latin America) are made from carved stone, which is an all together heartier substance. The same is true for almost any ancient archaeological monument you can name that isn't either underground, a bare foundation, or easily mistaken for a small hill.
There are adobe structures that have survived in some recognizable form for a couple thousand years (eg. the Bam Citadel), but considering how many buildings have been built that way, a rather small number of them are in good shape, and most of those have been more or less continuously maintained by people.
Certainly our abandoned cities would include an awful lot of obviously artificial materials, but it certainly isn't obvious to me that they'd be recognizable to you and I as cities in a few centuries. Structural steel and concrete don't weather particularly well once they've been exposed.
Ha! Good point. I read it, accepted it, and found it funny without pausing to consider that. Amazing. It actually makes no sense at all, except as a play on the wacky habit of naming things based on what they're not. (To the posters credit, that habit certainly does correlate with recursive naming, which must be why the joke seems funny anyway.)
Perhaps the poster meant "DEBUNTU is not Ubuntu" where the acronym "DEBUNTU" is composed entirely of. . well. . . the first word of the whole phrase. Not a particularly good way of forming an acronym, but sadly not the worst ever suggested in public.
Or, on further reflection, the poster never actually claimed it *was* recursive. He simply said, "take that, you crazy recursive-naming bastards!" Which could be interpreted to mean, "To those who create recursive acronyms, I hate you, and I hurl a very badly formed non-recursive acronym at you and hope it offends you." Not the obvious interpretation, perhaps, but none the less a valid one.
Or, perhaps he's just an TSURA. (TSURAs screw up recursive acronyms.)
I remember a now long dead contender from the ICQ heydays called Jabber that included text to speech in the client.
As I recall they included lots of uniformly annoying things as well that were a pain to disable, like loud cartoonish sound effects and "click click" noises to accompany keystrokes.
I though the text to speech feature would be handy. (And still could be.) In the end, though, I was never able to convince another human being to use it regularly. Most tried it for about a minute and ran off in search of a client that didn't go out of its way to irritate with a default install.
I suspect it's one of my top 5 most often used keystrokes.
Funny how different people behave so differently when performing the same sort of tasks.
I've got the screen key bound to backtick, since I almost never use it for anything, and on those occasions when I do use it, it's always a conscious decision.
And, at the risk of being both off topic and redundant, I have to ask: where the hell are all the extra buttons we ought to have on our keyboards?
There's room for another 20-30 touch typable buttons along the margins of a standard keyboard, and at least 5-10 nice big control key sort of buttons. Why are we stuck with a twenty year old interface + a "windows" key?
The chording gymnastics we have to go through in order to make a set of letters, numbers, and a half dozen control keys do everything is just crazy. I want big, sloppy thumbswitches under the space bar labeled ctrl2 and ctrl3, and a row of sticky mod buttons to the left of the tab row. And don't even get me started on the insanity of making both "caps lock" and "\" easier to hit than ctrl on most modern keyboards. I mean, come on - caps lock? Why not just rip out the key entirely and put a little note that say, "wouldn't it be nice to have a useful key in this incredibly handy place?" (Yeah, I know you can remap keys, and I do on every machine I control. . . but sometimes I have to use other people's machines, and in some of the more primitive operating systems there's no user specific way to remap keys.)
The most common for me (but also probably the least common generally and least relevant to this discussion) is to have a terminal up in which I'm tinkering with data and a couple of small plotting windows in which to display it. I find it handy to use a single full screen terminal (usually broken into internal columns in screen or emacs) and then to let the plotting windows float over the top of it in a place that doesn't get in the way much. That minimizes line wrapping and maximizes screen real estate. It also makes it possible to have dozens of plotting windows sitting in the background and to only stick those you are interested in at a particular moment in front of the main window without having to cycle through them hunting for the right ones.
The second is that there are lots of console programs out there that go nutso if you give them anything smaller than an 80xN terminal and don't treat resizing well. Sometimes it's handy to be able to keep an eye on a small section of one of those terminals without having to see the whole thing. Admittedly, this is probably kind of rare. I can only think of one or two examples.
The third is that one occasionally has to copy data from one place to another, as others have mentioned. True, it would usually be possible to do that in a tiled environment, although with small monitors one is limited to vertical columns to avoid lots of badly wrapped lines.
That said, I'm actually something of a fan of tiling window managers, particularly larswm. If it weren't for the first case (which unfortunately includes something like 50-60% of the time I spend at a computer), I'd switch over to it entirely.
. . . as far as I know, you can only tile them with no overlapping parts, and sometimes funny things happen when resizing terminals.
If all you want to do is stack two or three terminals in a vertical column without rearranging borders often, then it works great.
I've never been able to get anything more complicated to work reliably.
(And then again, half my terminals are usually logged into remote machines and running screen anyway. Even with compatible key-bindings, running screens within screens is something my brain just can't handle without constant goofups. So, keeping multiple xterms around is hard to avoid.)
>Okay, okay, I exaggerated. Slackware >has evolved. I was just trying to >make the point that newer, sexier, >more ricey distributions are eating up >its market share.
I suspect you're right that the fraction of linux users using Slackware has decreased in recent years.
But, I'd also suggest that the absolute number of Slackware users has probably increased. (No, I don't have any numbers handy. But, anecdotally, I've heard an awful lot of people talk about switching *to* slackware in the last few years, and very few talk about switching *from* slackware.)
If you ask me, that's okay. So long as we've got enough different distributions around to dissuade hardware and software makers from enforcing distribution-specific requirements, and so long as there are enough of us Slackers around to offer technical advice to each other and pay a couple people to maintain the distribution, then I'm content. I'm a big slackware fan, but I couldn't care less if Slack drops from the top 10, or the top 100.
Market share isn't a particularly useful metric when comparing non-commercial entities. The goal isn't for your particular distribution to gather as many users as possible, but instead it is for all the distributions taken together to be maximally useful. (Or at least, that's what the goal *should* be, if you ask me.)
If other distributions are doing something that most people like better than slackware, and especially if it draws in new linux users, than that's a good thing. It would be silly for Slack to try to compete with them, if you ask me. Let them serve the people who like what they're doing, and let them get the fame they deserve for it.
Slackware is exactly what I and other Slackers want it to be. (And an awful lot of us have come to it largely because it lacks the features of other distros that we dislike. Adding almost anything would damage it, if you ask me.) I'd rather have a distro that meets my needs perfectly than a very popular one that includes things I dislike. And so, I suspect, would most people.
Better a hundred specialized distros so that you can choose the one that does *exactly* what you want than one giant compromise that doesn't quite satisfy anyone at all.
>Nothing is stoping you from downloading >a source tarball and installing it the >same way you would in slackware, in fedora.
True, except that when you do install something from source, you have to either jump through all sorts of crazy hoops to get your package manager to know about it or else simply resign yourself to never again using the package manager for anything meaningful. At least, that was what I took away from a few abortive trials with Debian and RPM-centric distributions.
Inevitably, one runs into a piece of software or a library for which one needs a newer version than the official package. One is then faced with two options: remove the old version and break your package manager dependencies forever, or install the new version in parallel with the original and spend hours trying to force other software to use your new versions. If you're unlucky enough to need a new version of a library that is used by lots of managed software, it's a nightmare. (I gave up on RPMs forever after spending one of the most frustrating weekends of my life trying to force Mandrake to play nice with cvs versions of Alsa components.)
In slackware, it's easy: just remove the offending package and drop in the new version in the usual place. Your package manager won't know the difference, and all your software uses the new version without a hiccup, and you can continue to install and maintain all your other packages as before. And, when the time comes, you can still smoothly upgrade to the new version using the ordinary package manager tools without any funny business.
I'm sure there's some way to force other package managers to smoothly deal with situations like this. But, as doing so requires far more distribution-specific knowledge and effort than abandoning a dependency tracking package manager altogether, it's a non starter if you ask me. (Unless something had dramatically changed in the last couple years, in which case I happily withdraw my complaint.)
I'm glad RPM and apt exist, since a lot of people find them useful. But making a dependency tracking package manager an integral part of a system makes the whole system a lot less valuable, if you ask me. It doesn't just add options, it also adds complications.
>OK, and the antennas used in the MIT article >did not have a resonance at 2.6 GHz, but were >wideband antennas? Please give me the source of >these antennas, they might be exactly what >I have been looking for for a very long time...
They never claimed that. (And I certainly didn't.)
But they did do a sweep with the tin foil hat on and a sweep with the tin foil hat off and divide one by the other to generate their plot. Assuming their signal was well above the noise everywhere, it shouldn't make much difference what their antenna response looks like.
The only caveat is that from the photos they seem to be in the very near field, and coupling between the antenna and the tin foil hat itself is probably significant. (And, for that matter, even coupling to the guy leaning over it to press buttons on the network analyzer.) To really do it right, one would want to use a source located far away from the head in the open and explore a range of orientations and polarizations.
Some sections of track, especially in the tunnel and a few of the underground bits in Oakland are painfully loud. It's really pretty astounding, and far worse than any other subway I've encountered.
The NIST site that hosts the article you mentioned has some tips on specific media types, but trying to buy them in small retail quantities isn't alway easy. There's rarely any guarantee that on opening two identically packaged boxes from a media company you'll end up with identical media.
For some anecdotal info (and links to even more anecdotal info), check out the section 7 of the CD-Recordable FAQ
To add one more statistically questionable story to the pile, I know several recording studio techs who swear by Mitsui. They're a little more expensive than generics, but you can buy 50 or 100 disc pack from the company itself (or an official distributor) and be reasonably sure of what you're getting. I've had only good experiences with them myself.
But then I've had very few bad experiences so far with any media, and all of those have involved generics with gummy printable labels applied to them, and all were given to me by other people. (My own paranoid technique is to label disks only with a non-alcohol based felt tipped pen.)
On the other hand, if you're goal is archiving the irreplaceable (rather than just stuff that will be expensive to replace), it's hard to beat a pair of hard drives which contain flac (or, if disk space is cheaper than processor time wav) files and checksums for every file. Every year or two you plug in each drive and make sure all the files are good, and when it starts to become hard to find systems that will interface with your old drives, you transfer everything to new ones. When you can pick up a 200 gig ata drive and a USB hard drive enclosure for well under a hundred bucks, it's hard to argue against that sort of strategy. You could do the same with DVDs or even CDs, of course, but checking them becomes a manual hassle.
In any event, make two copies of everything so that if one goes bad, you are likely to have a backup. Keeping one somewhere other than your house doesn't hurt either.
>Wasn't the whole point of the MIT article that aluminum amplifies and tin degrades signals?
No.
The point of the article was that a hemisphere of thin foil with something sort of like a head in it doesn't block RF very well, and that in a couple specific frequency ranges and geometries it slightly increases the signal measured in the "head."
Wrapping an object in several layers of material and leaving no opening for RF to get in is a very different situation, and a perfectly sound strategy to defeat an RFID reader.
At least in the physical sciences, 2/3 of the audience have their laptops open and are busy at work finishing up their own presentations, sending email, and doing any number of other random things. I've never seen a classroom with anywhere *near* the density of laptops being used for tasks other than note taking that happens in a conference full of professors, post-docs, and senior grad students.
It seems to me the reason is simple: a lot of what speakers say isn't useful, even in the case of good speakers and voluntary audience. Even when it *is* useful, the rate at which information is presented is usually an order of magnitude slower than the rate at which the audience can absorb it, with huge gaps of dead time between important statements. So, at lectures people spend an hour sitting in their seats in order to catch a minute or two of really useful information.
As someone who hasn't taken a course without lots of equations and diagrams in a long time, I've never had an excuse to bring a laptop to class. Instead, I have to spend all that dead time thinking about other things on my own without the benefit of a technological distraction.
The problem with laptops in the classroom is that it hurts the feelings of lecturers, who are forced to confront the fact that most of their audience isn't paying attention to most of what they say most of the time.
There's no reason *not* to be confused by the article. It's a pretty subtle phenomenon, described in an astoundingly sloppy writeup. Hard to believe it took three people to write something which is neither complete nor coherent, and which doesn't even give you enough key words to search for more information.
The Gravity Probe B homepage has a far better introduction to the experiment. (Go to "classroom" -> "story of GPB" for a concise intro.) http://einstein.stanford.edu/
In short, general relativity predicts that a massive rotating object (like the earth) distorts the space around it in such a way that nearby objects that are locally at rest are actually rotating slightly when compared to distant stars. (Locally at rest means that, for example, if you put some guy in a box with any measurement apparatus he could imagine, his measurements would show that the box isn't rotating.) This doesn't happen in Newtonian physics, and Gravity Probe B should be able to measure it and compare it to what one predicts using GR.
The effect is usually called "frame dragging," or the "Lense-Thirring Effect."
MP3 is ubiquitous and great for almost any portable player. The only reason not to use it is because you like some other format and want to forcibly promote it. (Which assumes that your audience knows your work well enough to consider installing new software in order to hear you... which may or may not be true.) At least when using LAME, you can fine tune the compressing for the speaker to reduce file size pretty considerably.
Ogg vorbis is great for both philosophical and practical reasons, but you'll limit your audience since only a few portable player companies include support for it. If your background is quiet, you can really crank up the compression on a speach-only vorbis file before noticeably impacting quality.
Never used the speex, but there's no harm in offering it. Chances are few will use it.
>When the cow is asleep, it is not >consciously adjusting for it's balence, >if you run at it and give it a hard shove, >it falls over pretty easily. I have seen it >myself.
Sounds plausible.
This isn't really a statics problem at all. No one is going to tip a cow by walking up to it and trying to shove it from a dead stop.
A better model is one in which the person doesn't push on the cow at all, but rather collides with it and imparts momentum to it at shoulder height. The real question isn't "with how much force can a person continuously shove a cow," it is "how fast can you run at a cow without it noticing you and take evasive action?" (Although, even with the questionable assumptions that went into the article, they still manage to conclude that two people could tip a cow. That hardly sounds like a debunking to me.)
Can't wait for the followup: "Chopping wood: another rural myth debunked. If a person can push on an axe with a force equal to their own weight. . . "
One can only hope the student in this article isn't an indication of the caliber of RCMP forensics. (Unless one happens to be a Canadian criminal, that is.)
Slashdot Mirror
You make a very good point.
I generally can't stand the reality TV show scheme; however, I happened to catch half an episode of the first Joe Schmoe season at a friend's house and was instantly hooked. I've now seen the whole season.
It was a remarkably humane and thoughtful program. It not only celebrated humanity at its best but managed to tackle complicated ethical questions with more depth and integrity than any other program on television. (Albeit with the occasional gratuitous tit flash.)
Although it's not clear to me the producers *intended* (and it sure as hell isn't the program that was advertised in commercial spots), they managed to create a something genuine and moving in a field dominated by the cruel and the stupid. The result is closer to Studs Terkel than to Survivor.
I know nothing about the astronaut show except the slashdot blurb (TFA won't cooperate with my browser at present), but there's certainly the potential there to create something which isn't mean spirited and cruel. Tone down the competition, hire people who aren't selfish and stupid, and it's possible to create something truly worthwhile in reality TV.
I do hope they eventually straighten participants out about the bad physics they've had to feed them, though. Pointing out to the audience that their gravity explanation is total bullocks would seem an ethical must as well.
>You can't use the non-gui
>version of emacs? Why not?
>If you know the keybindings,
>emacs should be perfectly
>usable on the console.
I'm always amazed by the number of unix savvy people who don't know that emacs will run on the console. I suspect it's the primary reason emacs has a reputation for being slow and clunky.
If I were in charge, I'd replace the "-nw" flag with an "-x" flag. Making the default behavior slow and ugly is a questionable strategy. (As it is, I run a non-X build of emacs at home, just for the petty and senseless pleasure of saving a few extra kB of ram.)
>to be fair, emacs keybindings
>aren't easy to learn either
True. But, if you haven't learned them, there's the great advantage that you can almost always guess the name of the thing you want to do and cause it to happen anyway.
If you ask me, meta-x + function name (along with tab completion) is the second most useful and least advertised emacs feature. Being able to do things that you only need to do twice a year without having to look up some crazy key binding sure is nice. I haven't got a clue what the key binding for kill-rectangle or toggle-truncate-lines are, and probably never will. But I know those features exist and can easily cause them to happen every time I need them.
As an aside, I claim the very most useful and least advertised feature is the "-nw" flag. I've been astounded by the number of aged twenty-something unix savvy folks who've never heard of it and force themselves to learn VI because they don't want to run X apps over a slow network connection. (I'm not going to claim there aren't perfectly good reasons that a person would want to use VI for it's own sake. I'm talking strictly about the "I hate VI but I have no choice but to use it" crowd.)
Sounds like fun. (As an aside, I'm shocked by the number of people arguing against having after hours fun with such toys. Is the slashdot readership really so cowardly and unimaginative? Sure, one has to be careful and should avoid pissing off the bosses during the first few weeks at work. And it goes without saying that if the electron microscope happens to break while you're using it for personal projects, bad things are likely to happen to you, unless you have or happen to be a particularly cool boss. But, the risk may well be worth it.)
If you've got access to a scanning electron microscope, any sample should be fun. Around here (a multi-group academic facility) the machine is jealously guarded by a dedicated staff person and we get charged rather a lot of money for each use, so I haven't done any recreational microscopy. But, just looking at the stuff we're supposed to look at is overwhelmingly nifty. (Obviously you should stop and think before putting foreign objects into either the miscroscope itself or a sputtering chamber.)
With die cutting, ceramics, and electroplating, you could certainly make some beautiful cases for homebrew projects. If you go in for a retro look, try to cook up some faux-bakelite. (Or real bakelite, for that matter, if you can get your hands on the stuff.)
Another possibility would be tinkering with electrostatic levitation. Suspended objects are always neat.
You've also got the ingredients for making homebrew optics toys. With lapping and plating gear, you might be able to make your own optical quality mirrors for homebrew telescope parts / lasers / holography setups / etc. Anything else involving precision ground metal parts and custom ceramics is an obvious candidate: home made particle detectors / geiger muller tubes, for example.
And there's always the obvious option of making really big capacitors, charging them to really high voltages, and zapping things. (As described, for example, here http://www.amasci.com/amateur/capexpt.html )
>They are a subsidiary of the
>Warner empire (as are their
>record, film, and book divisions)
>but they are not a record company.
Actually both Warner/Chappell Music and Warner Records are now part of a separate company, the Warner Music Group. They were sold off by the rest of the Warner empire a couple years ago.
Of course that doesn't affect the point of your post. Their publishing house is now just a small part of a less giant company that also owns recording companies. But, with only two hands left, it is perhaps even harder to understand why they aren't looking out for each other's interests.
>If some researcher needs to be in orbit
>for some research, they pay NASA x dollars
>for room and board on the station
>(appropriately subsidized by the American
>taxpayer--x may be zero).
In which case the researcher writes out a grant proposal to NASA/NSF/DOE and asks for money to be used in paying NASA for room and board? Doesn't seem particularly effective, except possibly as a way to force through manned spaceflight at the expense of other programs if funding is ever cut. (Asking non-US researchers to pay might work, although it isn't clear how much demand there would be for seats from nations that aren't already ISS partners.)
On the other hand, if the ISS were forced to live off of money tied to specific research projects, funding agencies would pretty quickly realize how very little research there is that requires placing humans in low earth orbit for months. The cost of experiments that we've conducted in the recent decades are only remotely reasonable if you ignore the billions of dollars spent subsidizing the shuttle and station. As a way of killing off the ISS, it might work.
That's not to say that I'm particularly opposed to the shuttle, the ISS, or manned spaceflight in general. But it is and for the forseable future will be a public works project rather than a scientific endeavor.
That's a fine thing in itself, but it won't work as a for-profit enterprise. The only financially worthwhile thing you can do with humans in low earth orbit is to charge rich tourists to take them there. Given a choice between sending up some useless astronauts to hang about taking pictures of cool weightless stuff and sending up some useless right guys to hang about taking pictures of cool weightless stuff, I'd choose the former. At least the cool pictures will be public domain afterward.
>we are talking about a
>construction made of early
>forms of cement - which was
>made from straw and mud if I
>am not mistaken.
I'm not expert, but all the pyramids I know of (in either Egypt or Latin America) are made from carved stone, which is an all together heartier substance. The same is true for almost any ancient archaeological monument you can name that isn't either underground, a bare foundation, or easily mistaken for a small hill.
There are adobe structures that have survived in some recognizable form for a couple thousand years (eg. the Bam Citadel), but considering how many buildings have been built that way, a rather small number of them are in good shape, and most of those have been more or less continuously maintained by people.
Certainly our abandoned cities would include an awful lot of obviously artificial materials, but it certainly isn't obvious to me that they'd be recognizable to you and I as cities in a few centuries. Structural steel and concrete don't weather particularly well once they've been exposed.
>That's not recursive.
Ha! Good point. I read it, accepted it, and found it funny without pausing to consider that. Amazing. It actually makes no sense at all, except as a play on the wacky habit of naming things based on what they're not. (To the posters credit, that habit certainly does correlate with recursive naming, which must be why the joke seems funny anyway.)
Perhaps the poster meant "DEBUNTU is not Ubuntu" where the acronym "DEBUNTU" is composed entirely of. . well. . . the first word of the whole phrase. Not a particularly good way of forming an acronym, but sadly not the worst ever suggested in public.
Or, on further reflection, the poster never actually claimed it *was* recursive. He simply said, "take that, you crazy recursive-naming bastards!" Which could be interpreted to mean, "To those who create recursive acronyms, I hate you, and I hurl a very badly formed non-recursive acronym at you and hope it offends you." Not the obvious interpretation, perhaps, but none the less a valid one.
Or, perhaps he's just an TSURA. (TSURAs screw up recursive acronyms.)
Doh!
Jabber is, of course, a perfectly viable and great protocal available today. (And my favorite.)
I was thinking of something called PowWow.
Sorry!
I remember a now long dead contender from the ICQ heydays called Jabber that included text to speech in the client.
As I recall they included lots of uniformly annoying things as well that were a pain to disable, like loud cartoonish sound effects and "click click" noises to accompany keystrokes.
I though the text to speech feature would be handy. (And still could be.) In the end, though, I was never able to convince another human being to use it regularly. Most tried it for about a minute and ran off in search of a client that didn't go out of its way to irritate with a default install.
If you're a linux user, check out ixbiff:
http://ixbiff.sourceforge.net/
It maps keyboard LEDS to mail directories.
Pretty handy, at least if you use your scroll-lock LED to check on your scroll lock status as often as I do.
Hmmm. ctrl-z is rarely used on the command line?
I suspect it's one of my top 5 most often used keystrokes.
Funny how different people behave so differently when performing the same sort of tasks.
I've got the screen key bound to backtick, since I almost never use it for anything, and on those occasions when I do use it, it's always a conscious decision.
And, at the risk of being both off topic and redundant, I have to ask: where the hell are all the extra buttons we ought to have on our keyboards?
There's room for another 20-30 touch typable buttons along the margins of a standard keyboard, and at least 5-10 nice big control key sort of buttons. Why are we stuck with a twenty year old interface + a "windows" key?
The chording gymnastics we have to go through in order to make a set of letters, numbers, and a half dozen control keys do everything is just crazy. I want big, sloppy thumbswitches under the space bar labeled ctrl2 and ctrl3, and a row of sticky mod buttons to the left of the tab row. And don't even get me started on the insanity of making both "caps lock" and "\" easier to hit than ctrl on most modern keyboards. I mean, come on - caps lock? Why not just rip out the key entirely and put a little note that say, "wouldn't it be nice to have a useful key in this incredibly handy place?" (Yeah, I know you can remap keys, and I do on every machine I control. . . but sometimes I have to use other people's machines, and in some of the more primitive operating systems there's no user specific way to remap keys.)
Ah well, enough ranting, I guess.
There are three reasons I do it.
The most common for me (but also probably the least common generally and least relevant to this discussion) is to have a terminal up in which I'm tinkering with data and a couple of small plotting windows in which to display it. I find it handy to use a single full screen terminal (usually broken into internal columns in screen or emacs) and then to let the plotting windows float over the top of it in a place that doesn't get in the way much. That minimizes line wrapping and maximizes screen real estate. It also makes it possible to have dozens of plotting windows sitting in the background and to only stick those you are interested in at a particular moment in front of the main window without having to cycle through them hunting for the right ones.
The second is that there are lots of console programs out there that go nutso if you give them anything smaller than an 80xN terminal and don't treat resizing well. Sometimes it's handy to be able to keep an eye on a small section of one of those terminals without having to see the whole thing. Admittedly, this is probably kind of rare. I can only think of one or two examples.
The third is that one occasionally has to copy data from one place to another, as others have mentioned. True, it would usually be possible to do that in a tiled environment, although with small monitors one is limited to vertical columns to avoid lots of badly wrapped lines.
That said, I'm actually something of a fan of tiling window managers, particularly larswm. If it weren't for the first case (which unfortunately includes something like 50-60% of the time I spend at a computer), I'd switch over to it entirely.
. . . as far as I know, you can only tile them with no overlapping parts, and sometimes funny things happen when resizing terminals.
If all you want to do is stack two or three terminals in a vertical column without rearranging borders often, then it works great.
I've never been able to get anything more complicated to work reliably.
(And then again, half my terminals are usually logged into remote machines and running screen anyway. Even with compatible key-bindings, running screens within screens is something my brain just can't handle without constant goofups. So, keeping multiple xterms around is hard to avoid.)
>Okay, okay, I exaggerated. Slackware
>has evolved. I was just trying to
>make the point that newer, sexier,
>more ricey distributions are eating up
>its market share.
I suspect you're right that the fraction of linux users using Slackware has decreased in recent years.
But, I'd also suggest that the absolute number of Slackware users has probably increased. (No, I don't have any numbers handy. But, anecdotally, I've heard an awful lot of people talk about switching *to* slackware in the last few years, and very few talk about switching *from* slackware.)
If you ask me, that's okay. So long as we've got enough different distributions around to dissuade hardware and software makers from enforcing distribution-specific requirements, and so long as there are enough of us Slackers around to offer technical advice to each other and pay a couple people to maintain the distribution, then I'm content. I'm a big slackware fan, but I couldn't care less if Slack drops from the top 10, or the top 100.
Market share isn't a particularly useful metric when comparing non-commercial entities. The goal isn't for your particular distribution to gather as many users as possible, but instead it is for all the distributions taken together to be maximally useful. (Or at least, that's what the goal *should* be, if you ask me.)
If other distributions are doing something that most people like better than slackware, and especially if it draws in new linux users, than that's a good thing. It would be silly for Slack to try to compete with them, if you ask me. Let them serve the people who like what they're doing, and let them get the fame they deserve for it.
Slackware is exactly what I and other Slackers want it to be. (And an awful lot of us have come to it largely because it lacks the features of other distros that we dislike. Adding almost anything would damage it, if you ask me.) I'd rather have a distro that meets my needs perfectly than a very popular one that includes things I dislike. And so, I suspect, would most people.
Better a hundred specialized distros so that you can choose the one that does *exactly* what you want than one giant compromise that doesn't quite satisfy anyone at all.
>Nothing is stoping you from downloading
>a source tarball and installing it the
>same way you would in slackware, in fedora.
True, except that when you do install something from source, you have to either jump through all sorts of crazy hoops to get your package manager to know about it or else simply resign yourself to never again using the package manager for anything meaningful. At least, that was what I took away from a few abortive trials with Debian and RPM-centric distributions.
Inevitably, one runs into a piece of software or a library for which one needs a newer version than the official package. One is then faced with two options: remove the old version and break your package manager dependencies forever, or install the new version in parallel with the original and spend hours trying to force other software to use your new versions. If you're unlucky enough to need a new version of a library that is used by lots of managed software, it's a nightmare. (I gave up on RPMs forever after spending one of the most frustrating weekends of my life trying to force Mandrake to play nice with cvs versions of Alsa components.)
In slackware, it's easy: just remove the offending package and drop in the new version in the usual place. Your package manager won't know the difference, and all your software uses the new version without a hiccup, and you can continue to install and maintain all your other packages as before. And, when the time comes, you can still smoothly upgrade to the new version using the ordinary package manager tools without any funny business.
I'm sure there's some way to force other package managers to smoothly deal with situations like this. But, as doing so requires far more distribution-specific knowledge and effort than abandoning a dependency tracking package manager altogether, it's a non starter if you ask me. (Unless something had dramatically changed in the last couple years, in which case I happily withdraw my complaint.)
I'm glad RPM and apt exist, since a lot of people find them useful. But making a dependency tracking package manager an integral part of a system makes the whole system a lot less valuable, if you ask me. It doesn't just add options, it also adds complications.
>Hashes are one way.
>There are an infinite
>number of files that
>have the same hash. The
>tricky part is finding them.
But if we limit ourselves to files of a specific size, then we get away with a finite (but unpleasantly large) number of files.
In that case, all we have to do is design a quantum computer able to pick the one that is best described by the liner notes.
Granted, it's a bit harder than just buying some extra media in the first place, but certainly a lot more fun.
>OK, and the antennas used in the MIT article
>did not have a resonance at 2.6 GHz, but were
>wideband antennas? Please give me the source of
>these antennas, they might be exactly what
>I have been looking for for a very long time...
They never claimed that. (And I certainly didn't.)
But they did do a sweep with the tin foil hat on and a sweep with the tin foil hat off and divide one by the other to generate their plot. Assuming their signal was well above the noise everywhere, it shouldn't make much difference what their antenna response looks like.
The only caveat is that from the photos they seem to be in the very near field, and coupling between the antenna and the tin foil hat itself is probably significant. (And, for that matter, even coupling to the guy leaning over it to press buttons on the network analyzer.) To really do it right, one would want to use a source located far away from the head in the open and explore a range of orientations and polarizations.
Yup.
I always wear earplugs on BART.
Some sections of track, especially in the tunnel and a few of the underground bits in Oakland are painfully loud. It's really pretty astounding, and far worse than any other subway I've encountered.
The NIST site that hosts the article you mentioned has some tips on specific media types, but trying to buy them in small retail quantities isn't alway easy. There's rarely any guarantee that on opening two identically packaged boxes from a media company you'll end up with identical media.
For some anecdotal info (and links to even more anecdotal info), check out the section 7 of the CD-Recordable FAQ
http://www.cdrfaq.org/
To add one more statistically questionable story to the pile, I know several recording studio techs who swear by Mitsui. They're a little more expensive than generics, but you can buy 50 or 100 disc pack from the company itself (or an official distributor) and be reasonably sure of what you're getting. I've had only good experiences with them myself.
But then I've had very few bad experiences so far with any media, and all of those have involved generics with gummy printable labels applied to them, and all were given to me by other people. (My own paranoid technique is to label disks only with a non-alcohol based felt tipped pen.)
On the other hand, if you're goal is archiving the irreplaceable (rather than just stuff that will be expensive to replace), it's hard to beat a pair of hard drives which contain flac (or, if disk space is cheaper than processor time wav) files and checksums for every file. Every year or two you plug in each drive and make sure all the files are good, and when it starts to become hard to find systems that will interface with your old drives, you transfer everything to new ones. When you can pick up a 200 gig ata drive and a USB hard drive enclosure for well under a hundred bucks, it's hard to argue against that sort of strategy. You could do the same with DVDs or even CDs, of course, but checking them becomes a manual hassle.
In any event, make two copies of everything so that if one goes bad, you are likely to have a backup. Keeping one somewhere other than your house doesn't hurt either.
>Wasn't the whole point of the MIT article that aluminum amplifies and tin degrades signals?
No.
The point of the article was that a hemisphere of thin foil with something sort of like a head in it doesn't block RF very well, and that in a couple specific frequency ranges and geometries it slightly increases the signal measured in the "head."
Wrapping an object in several layers of material and leaving no opening for RF to get in is a very different situation, and a perfectly sound strategy to defeat an RFID reader.
At least in the physical sciences, 2/3 of the audience have their laptops open and are busy at work finishing up their own presentations, sending email, and doing any number of other random things. I've never seen a classroom with anywhere *near* the density of laptops being used for tasks other than note taking that happens in a conference full of professors, post-docs, and senior grad students.
It seems to me the reason is simple: a lot of what speakers say isn't useful, even in the case of good speakers and voluntary audience. Even when it *is* useful, the rate at which information is presented is usually an order of magnitude slower than the rate at which the audience can absorb it, with huge gaps of dead time between important statements. So, at lectures people spend an hour sitting in their seats in order to catch a minute or two of really useful information.
As someone who hasn't taken a course without lots of equations and diagrams in a long time, I've never had an excuse to bring a laptop to class. Instead, I have to spend all that dead time thinking about other things on my own without the benefit of a technological distraction.
The problem with laptops in the classroom is that it hurts the feelings of lecturers, who are forced to confront the fact that most of their audience isn't paying attention to most of what they say most of the time.
There's no reason *not* to be confused by the article. It's a pretty subtle phenomenon, described in an astoundingly sloppy writeup. Hard to believe it took three people to write something which is neither complete nor coherent, and which doesn't even give you enough key words to search for more information.
The Gravity Probe B homepage has a far better introduction to the experiment. (Go to "classroom" -> "story of GPB" for a concise intro.)
http://einstein.stanford.edu/
In short, general relativity predicts that a massive rotating object (like the earth) distorts the space around it in such a way that nearby objects that are locally at rest are actually rotating slightly when compared to distant stars. (Locally at rest means that, for example, if you put some guy in a box with any measurement apparatus he could imagine, his measurements would show that the box isn't rotating.) This doesn't happen in Newtonian physics, and Gravity Probe B should be able to measure it and compare it to what one predicts using GR.
The effect is usually called "frame dragging," or the "Lense-Thirring Effect."
Yup. "All of them" is the way to go.
MP3 is ubiquitous and great for almost any portable player. The only reason not to use it is because you like some other format and want to forcibly promote it. (Which assumes that your audience knows your work well enough to consider installing new software in order to hear you... which may or may not be true.) At least when using LAME, you can fine tune the compressing for the speaker to reduce file size pretty considerably.
Ogg vorbis is great for both philosophical and practical reasons, but you'll limit your audience since only a few portable player companies include support for it. If your background is quiet, you can really crank up the compression on a speach-only vorbis file before noticeably impacting quality.
Never used the speex, but there's no harm in offering it. Chances are few will use it.
>When the cow is asleep, it is not
>consciously adjusting for it's balence,
>if you run at it and give it a hard shove,
>it falls over pretty easily. I have seen it
>myself.
Sounds plausible.
This isn't really a statics problem at all. No one is going to tip a cow by walking up to it and trying to shove it from a dead stop.
A better model is one in which the person doesn't push on the cow at all, but rather collides with it and imparts momentum to it at shoulder height. The real question isn't "with how much force can a person continuously shove a cow," it is "how fast can you run at a cow without it noticing you and take evasive action?" (Although, even with the questionable assumptions that went into the article, they still manage to conclude that two people could tip a cow. That hardly sounds like a debunking to me.)
Can't wait for the followup:
"Chopping wood: another rural myth debunked. If a person can push on an axe with a force equal to their own weight. . . "
One can only hope the student in this article isn't an indication of the caliber of RCMP forensics. (Unless one happens to be a Canadian criminal, that is.)