It isn't a question of how long. It is more a question of dollars.
It is often said that 20 years ago, the physics community estimated that they could have reactors working in 20 years. People usually ignore that this was only the first half of the estimate -- the other half was the level of funding needed to achieve the result. Needless to say, they received a small (and still shrinking) fraction of the funding they said was necessary, and the result is unsurprising.
Everyone laments the expense of large scale research in creating new basic technology. Bare in mind, however, that the production cost of the movie Spiderman 2 was a bit larger than the budget for the whole US fusion program the year it was released.
The big question is this -- what would she like to be able to do at the end of the day? Does she just want to program for the intellectual challenge of it? Or does she want something to tinker with? Or, does she want to learn a skill? (I really mean a skill, in this case, not a vocation.)
If she's interested in programming as an intellectual challenge, I think PLT-Scheme'sDrScheme is your tool of choice. I've never used HtDP, but just following the built-in tutorials that come with DrScheme is very instructive. This is especially true if you are willing to set aside the goal of building some particular program. When I get bored, I will occasionally do a problem or two from the DrScheme tutorial. It's much more fun than dumb desktop games like Solitare. Personally, I think of it more as solving puzzles than as "coding." (It's also a Debian package, so life is good.)
If she wants something to tinker with, I suggest getting a little colo box (a UML box isn't too expensive) and hacking around a little with apache, perl, python, shell and web-ish stuff. It's not always strictly programming, but it's a lot of fun and results can be seen without a huge amount of time/knowledge/effort, and she'll have something tangible for her efforts. Well, as tangible as anything in computing can be, anyway.
What the hell is the point of an RSS feed with naught but a few sentences of the story? And no images? The above story looks like this on Slashdot's RSS feed:
RSS Web-Feeds, The Next Big Thing? Sun, 29 Feb 17:36:00
mi writes "Yahoo! carries an Associated Press editorial about RSS-based news feeds, and how they are pushing the spam-ridden e-mail and advertising-ridden...
Creator: CmdrTaco
Why bother with that? Boooooo Slashdot! Include the whole article! If the web sucks because of popups (I haven't seen one in years, thanks to Mozilla), ads (haven't seen one of those in years, thanks to AdZapper and squid), and spyware (haven't seen any of that, thanks to running an OS that actually has a security model) then RSS sucks because of crummy crippled feeds like the one from Slashdot.
Have you tried using the Linux Test Project? I administer a cluster of Linux machines, and use LTP as a pretty comprehensive test framework. Many of the tests are software related, but you can shut those off if they're not useful to you.
I suggest you make a Knoppix CD with LTP installed. With a little configuration, that will take care of all of all your tests for the memory, disk, IO, and CPU. You might want to install America's Army or something to test the video subsystem.
If you put a little effort into it, you'll have a test suite as good as, and likely better than, anything you could pay money for. If you want to buy something, you can make a donation to the LTP and Knoppix projects.
There are also simpler tools, like Memtest86. I find this tool to be invaluable when I try to salvage old hardwar. I can't begin to tell you how much time it's saved me that I would have spent aimlessly swapping components around.
With one caviat: You are describing the Slashdot ecology, not the Linux community in general. People come to Slashdot because they want to rant about something. For reasons I don't understand (tradition? the moderation system? Hemos?), it attracts all the hotheads in, well, just about every software community I can think of.
The wider *NIX community is actually pretty mellow, with a few well-documented exceptions. Theo de Raadt, for instance. He's a well known dickhead, but even he admits that. (And in his case, his acidic personality actually serves to make OpenBSD more secure, I think.)
Slashdot is fun, but don't take it too seriously, or as even as representative.
Stand up and be counted. SCO can only afford so much in legal expenses, and there are a lot of us. I'll happily sit around in a courtroom while they piss away their cash.
To: bstowell@sco.com Subject: a request From: Russell Neches
Hi there --
I've used Linux since 1998. I own and operate a variety of machines of various architectures on which I have installed the GNU/Linux operating system. Furthermore, I use a several Linux machines at school (as well as working as a volunteer administrator of said machines). Furthermore, I use and enjoy a large number of products and services that depend on Linux. This includes, but is not limited to, Akami hosting, Weta Digital movies, Merrill Lynch financial services, New York Times Co. media products, and T-Mobile wireless services (who happen to run my distribution of choice, Debian).
If you plan to continue your pursuit of licencing fees from Linux users, I am hereby declaring my intent to refuse to pay such fees. Please add me to your list of people against whom you wish to file suit.
I may be reached at xxxxxxx@xxx.xxx.xxx, or through my (Debian-using) cell phone service provider at xxx-xxx-xxxx. I shall furnish a mailing address once I have established one.
All right, I wasn't looking at the desk. It could be plywood and cinderblocks for all I care. Just goes to show that you can sell anything if you put a pretty enough girl next to it, and I'm only slightly ashamed to admit that it works on me.
Before anyone jumps on me for this, I do in fact own a PII 500. It's a prototype that Intel wisely never foisted on the market.
What can I say. It was free and it runs Linux like a champ, aside from a needing trivial patch to the kernel source to prevent it from thinking it's running on a PIII. And it does the job.
Get a laptop. An old laptop. Install the weirdest OS you can find that has a networking stack. Make sure you have a couple of battaries that hold a charge so you can take it to the library, coffee shop or lobby while your roommate is busy contracting and spreading chlamydia, or whatever STD is popular on your campus.
Here's the reasoning: you want to make sure that you cannot play games on your computer. You know as well as I do that if you can play games, you will. Intead of doing your homework. I know whole Counter Strike clans that failed out of expensive private universty educations. You must avoid this fate at all costs.
Sound lame? Yeah, it is. But think of it this way. You (or your parents, or the government) is/are paying tens of thousands dollars a year to send you to a place where you can aquire an education. It's very likely that this is the only shot you're going to get, and that if you screw up bad enough, you've got a rewarding carrear in burger flipping.
That doesn't mean that you shouldn't have fun; on the contrary, you should have as much fun as you can. But, keep in mind that you are packed into a tiny, grubby place with thousands of other people your age, some of whom are worth getting to know. Keep in mind that there are proffesors and staff who've dedicated their lives to educating punks like you. Keep in mind that there is probably an interesting city or town to explore. Keep in mind that there is probably a gym that's flat-out better than any fitness company you could find that you can just use, for free. And you're probably miserably out of shape. Keep in mind that there is probably a world-class library crammed with books you should have already read by now. Exploit all of these things to the maximum extent permitted by hours in the day and callories in your diet, and maybe you'll get your money's worth.
As much as I like video games, they are mutually exclusive with these goals.
So, get an old laptop. Resist the urge to splurge on anything more ostentatious than a Pentium II 500. Your friends will laugh at it. Tell them you're poor, and that they should fuck off. Instead of playing games, amuse yourself with your creaky old hardware by hacking cool software. Or whatever you like, so long as you're creating something. You don't need fancy-pants graphics to run vim, screen, ssh, gcc, mutt, LaTeX and xterm. You might need a little more oomph for javac, or mzscheme, perl, or the like if your classes need 'em. Gaim, naim, or ICQ if it improves your social life. xmms, but don't go nuts on the P2P networks. It's a waste of your time. If your roommate wants to waste their time, mooch of of him or her.
Trust me. If you think you need anything else, you need to re-evaluate your goals.
And if you're lazy and/or adventurous, you can turn on NFS over TCP in your kernel and tunnel it over ssh or ppp/ssh. I've never tried it, but it ought to work. I understand that NFS over TCP is relatively untested, but is reputed to work rather well. Doing weird things like this would be a pretty good way to test the NFS over TCP code, and I'm sure the developers would be interested to hear how it goes. Particularly if you run a lot of data over it for a long time, and have a good way of verifying that all is well. Or, better still - if all is not well, and you have a good way of articulating what went wrong.
Of course, that doesn't mean it's a good idea. I think your solution with IPSec is much more elegant. Unfortunately, I happen to need to get through a heavily packet-shaped network that massively favors port 80, and drops random packets everywhere else. Not IPSec friendly at all. I avoid this by running multiple ppp/ssh tunnels through the retarded parts of the network and letting my gateway balance between them. Unfortunately, this requires privileged accounts on many, many boxes in odd places.
By the way, 10 points to any Northeastern University students who send polite, well considered complaints to Network Services. Not RESNet - they exist only to prevent you from talking to Network Services. Don't bother yelling at them - they exist specifically for that purpose. RESNet has no authority whatsoever to, for instance, allow CVS to work when Network Services decides to to drop 90 percent of packets on port 2401. This is for your benifit - I'm perfectly happy with my tunnels.
People might be interested in looking at the article Lessing is talking about, since the Red Herring article does not link to it. enjoy
Permanant I/O error while reading Polish backwards
on
Gnarly Error Messages
·
· Score: 2
It kind of makes sense, oddly enough. This error was thrown by a preprocessor on PDP/11, if I remember the tale correctly. It was induced when you screwed up your data types (float, int, +, or something like that). In order to turn this into a valid assembly instruction, you have to reverse the order of the elements (add $r1 $r2). If the preprocessor didn't have a function for generating the proper assembly, it would hang. The program waiting for its output would then report an unrecoverable IO error... while reading reverse Polish notation backwards.
Well, it'll be a great candidate once they figure out the itunesdb and other issues. There's been some progress, though. So, the answer is no for now. Uneless you consider OS X to belong to the set {Linux, OpenBSD}, which most people do not.
Build a gigantic secret base under an island somewhere. Equip with bad 70's decor, absurd super-weapons, henchmen, girls in swimsuits that inexplicably produce large caliber handguns when required, speed boats, hilicopters, and so forth. Put security cameras everywhere, and pay a midget or a guy with claws or something to monitor them. Obtain an absurd exotic pet, like a singing tapir or a farting oscilot. Train said pet to run errands about the island. Develop some mildly disturbing hobby, interest or passtime, such as bobbing for bat guano or hunting opera singers with a crossbow. The secret plans themselves, of course, should be encoded somehow in a fantastically ugly peice of fo-futurostic art. Where else?
Soft, by Rupert Thompson (ISBN 0-7475-3670-8). A beverage company uses subliminal messages, and "embassadors" to pitch their product. Very creepy book. Then again, it's a creepy idea.
Hack together something that lets you work around your window manager's key bindings.
Post your patch to the SDL mailing list
So what if you're not a big-time programmer? Read the documentation, the man pages, and the FAQs. If you still need guidance, politely solicit assistance from the appropriate SDL mailing list. If you're coherant, polite and willing to learn, you will find that people will help you. I can't overemphasize how important it is to communicate like an adult. Every project wants helpful, compotent and motivated developers. Make it your goal to become such a person, and you'll receive a remarkable amount of help. (In contrast, if you whine and bitch on the mailing list, you will be ignored at best.)
Yes, your first patch will probably suck. Big deal. Think of it as a "proof of concept." If it turns out to be a Good Idea, someone with more experience may clean it up a bit. Maybe that someone will be you, assuming you keep hacking. If all goes well, then your patch will be included in a release. Put it on your resume, pat yourself on the back, and enjoy gaming on your platform of choice.
Non-ionizing radiation may disrupt or alter inter-cellular signalling and thereby interfere with those mechanisms.
Well, clearly there are many different ways of triggering cancer. What
I'd like to know specifically how a 1.9 GHz signal can interact
with parts of a cell, or any chemical used for signaling. If you can
explain the mechanism by which a photon with a wavelength of 4 cm can
interact directly with the chemical function of a cell, I'll be
very impressed. Of course, the biology is quite complicated - but the
physics isn't.
Wavelengths of that size don't have enough energy to
disrupt even a weak chemical bond. Photons don't leave cancer-causing mojo just by passing through - the only way they interact with matter (for our purposes) is by absorbtion. So, if it's going to cause damage, it has to get absorbed by something. The only structures in the human body large enough to absorb that kind of signal are tissues, and the consiquence of absorbing such a photon would be heat.
All right, I suppose if you had a powerful enough signal, you
could cook yourself. But at 0.75 watts, you're going to have to wait a
long, long time. I suppose that raising the temperature in your head by a fraction of a degree could have an impact on intercellular (no pun intended) signalling, but I don't see how it would be different from a very slight fever, or sitting by a campfire. If you focused that 0.75 watts into a very, very small area, you might be able to cause a tiny radiation burn (which would also denature the neurotransmitters, almost certainly rendering them useless), but cell phones have omnidirectional antennae.
No matter how you look at it, the effect of this kind of radiation on the human body is going to be within the range of normal operation. It has to be : for a cell phone to give you cancer, the human body would have have to be more chemically unstable than nitroglycerine.
There may be other dangers from exposure to low
frequency radiation (bad driving, rudeness, use of annoying ringtones... ^_^ ), but I'm pretty sure cancer isn't one of them.
I've heard this kind of stuff over and over again - and alarmingly
often from people who ought to know better (physics teachers,
engineers, et cetera). The next time someone starts to tell you about
cell phones giving you cancer, here's what you should tell them.
My cell phone (the ubiquitous Nokia 3360) is a TDMA phone that
operates in the IS-54 (800 MHz) and IS-136 (1.9 GHz) bands. Now, 1.9
GHz sounds like a big, scary cancer-causing number. So let's see if it
really is.
First of all, we need to know how radiation causes cancer. We'll just
assume it's electromagnetic radiation, since cell phones definitely do
not emit anti-protons, neutrons, muons and other shit like
that. There's no way in hell a battery the size of a Triscut can
generate reaction energies high enough to produce hadrons or leptons,
so we can forget about them. (Well, actually, with a big capacitor you might get a few, but you're already getting showered with cosmic rays, and the pathetic little fart of hadrons you'd get out of a cell phone battery wouldn't count for didly squat.) The cancer-causing mechanism for
electromagnetic radiation is fairly simple. In order to be dangerous,
a photon (the electromagnetic force carrier particle) needs to carry
enough energy to ionize (chemistry parlance for "fuck up") something
important. It doesn't really matter how many photons you're slinging
around, since it's the frequency that determines the energy of a
single quanta.
So, what is our hypothetical candidate cancer-causing quanta going to
have to inonize to do the deed? Well, DNA of course. It's going to
have to cause a genetic mutation. Because of the way photons interact
with matter, they are most likely to be absorbed by
electromagnetically contiguous objects of sizes roughly equal to their
wavelength. The reasons are deeper than this, but suffice it to say
that a photon is "smeared" over an area about the size of its
wavelength. Since you can't absorb part of a quanta (that's why
they're called quanta, after all), you have to have a thing big enough
to soak up a whole particle about the size of the wavelength. In this
way, everything is, or is made of, antennae. To cause a mutation, you
have to have a photon whose wavelength is about equal to diameter of a
DNA molecule. Actually, the ideal length of an antenna is a quarter
the wavelength of its intended optimal frequency, so we'll say the
wavelength we're looking for is four times the diameter.
So, as I said, my cell phone operates at 1.9 GHz, or 1.9 billion
cycles per second. What's the wavelength? Well, wavelength is the
period times the speed of light. The period is the the inverse of
frequency, so :
3*10^8 / 1.9*10^9 ~= 0.16 M
That's about the length of your hand, give or take a thumb. One
quarter of that is about 4 cm - about the length of your thumb, give or take a nail. Now ask
yourself this question: How big is your DNA?
If your DNA is built out of atoms the size of rasins, you might have
something to worry about. The diameter of the DNA helix is 2 nm and
the vertical rise per base pair is 0.34 nm. If you want a photon that
will be able to reliably zap DNA, it needs to have a wavelength
_smaller_ than 8 nm. The probability that a photon will be absorbed by
a given object decreases with respect to the difference between the
size of the object and the wavelength of the photon according to the
standard deviation. So what's the probability that a given photon
spewing out of my cell phone is going to fry some of my DNA? Well,
we're a factor of five million away from the optimal wavelength. I'd
say it's pretty fucking unlikely.
But wait a second - what's kind of radiation has a wavelength of 8
namometers? Well, we do the opposite to find the frequency :
3*10^8 / 8*10-9 = 3.7*10^16
That's in the ultraviolet range. Surprise, surprise!
So, what can we conclude from this? Well, since a cell phone has a
transmission power of less than a watt and a wavelength the size of
your thumb, it's not going to do jack shit to your DNA. Nada. Zilch.
In other words, THERE IS NO WAY CELL PHONE RADIATION CAN GIVE YOU
CANCER!!! I'd be more inclined to beleive that the plastic in the
earpiece causes cancer.
You're several orders of magnitude more likely to contract cancer as a
result of proximity to a 100 watt incandecant light bulb. It's got a
much, much higher output, and its frequency range is thousands of
times higher.
So relax, enjoy your wireless technology, and wear your SPF-30.
You know, if you're going to argue with someone, it helps if you don't refute your own points.
I was (rather obviosly) speaking about the historical state of mplayer as it pertains to this latest release.
Of course support for binary only drivers is the responsibility of their respective developers. That's why I was drawing the comparison. The mplayer developers don't have time to deal with people's random binary distributions of their project. The kernel developers don't have time to deal with people's binary-only drivers. For essentially the same reasons.
Yes, current versions have runtime CPU detection. I'm glad you were able to read the.90pre1 changelog.
To everyone else - I apologize for feeding the troll.
Their web site explains the licencing problems they've had pretty clearly.
Part of the reason is because they can't distribute binary packages. There are legal reasons, but at this stage of develpment it doesn't make sense for them to worry about bugs in binary distributions. Linux kernel developers will give you similar rhetoric about supporting binary drivers. It's technically doable, but it's so much more work than dealing with source that it's not worth their time.
There are also technical reasons (like a lack of runtime CPU detection, so you kinda have to compile it locally).
It also appears that they needed to use non-GPLed codecs in their earlier releases. I think they are either removing them in the.90 release, or they are migrating to GPL-friendly codecs.
Read their website - it explains pretty much everything.
A long, long time ago (like, 1992 or thereabouts) I scribbled out a design for a CD-ROM that I calculated could do 100x fairly easily, although that wasn't my intention in the design. I was trying to find a way to prevent portable CD players from skipping, eating batteries, and otherwise sucking.
Basically, you hold the CD still in a little bracket, and spin a tiny little curved mirror around at the center. Since the laser will bounce erratically off the surface of the CD, you would read from the disk by placing a thin glass or plastic cover over the CD with a few photosensors sensors around its edge. The returning laser (carrying the data) could strike the cover at any pount, and the internal reflection of the cover would get enough of it to the photosensors to read the data. The laser will zip all over the place, so you'd use timing to ignore the data from non-contiguous parts of the disk. The mirror could be as small as the diameter of the laser, so you could spin it much, much faster than the CD iself could withstand.
The only problem I counld think of for such a device is that I don't think normal optical media will work as expected if you read it at a low angle.
Clearly, since no one seems to have done it, it's not that great of an idea. There's probably something wrong with it that I didn't think about at the time. Oh well - I was 12, and I just wanted to listen to Paul Simon without having to worry about bumping the desk while I was doing my cursed multiplication tables.
I don't mean to sound contentious, but Parrot is the virtual machine
that Perl 6 will target. It is
not, in fact, the language itself.
In this way, I suppose, Perl 6 will be kinda sorta like Java in that
it will target a VM, rather than a CPU. The Parrot VM is not exclusive
to Perl - from what I understand, idea is that eventually Python and
Perl will share a common VM. It's conceivable, I suppose, that some
sick bastard could write a Java compiler that targets Parrot.
Parrot isn't finished yet, but there are a few "toy" languages that
target it (Jako and Cola, and more recently BASIC).
The nice thing about current motherboard fab processes is that they are cheap. Populating both side of the board is possible, but rather expensive. It requires more layers (which make the board more expensive to make and more difficult to debug). Also, it requires very different physical manufacturing facilities, since the presence of components on the reverse side makes putting pressure on the topside difficult. When the robots punch down components on the top, they could squash, crack or bend componets on the other if the board isn't sitting on a specially designed caddy. As you might imagine, it's possible to do the same thing when you have the board at home when installing componets. (Zero insertion force my foot!) There are also issues of cooling, since a the board would require airflow on both sides (which could mean a bigger case in a worst-case situation!).
Of course, if these issues can be addressed with a sane form factor standard, I think everyone would be happy to have smaller computers. Smaller is better, so long as it doesn't apply to things that would require you to buy an expensive car to compensate.
Personally, I think one of the major limitations of the ATX standard is the breakout. If there were a standard IO block that connected to the motherboard with a cable, you would have a lot more freedom when it came to positioning the board inside the case. Old AT motherboards did this for everything except the keyboard connector, I seem to recall. It would be a little more complicated to install, but I think it would be better.
Or, we could just make mass market PC104 stacks for everything.
Slashdot Mirror
It isn't a question of how long. It is more a question of dollars.
It is often said that 20 years ago, the physics community estimated that they could have reactors working in 20 years. People usually ignore that this was only the first half of the estimate -- the other half was the level of funding needed to achieve the result. Needless to say, they received a small (and still shrinking) fraction of the funding they said was necessary, and the result is unsurprising.
Everyone laments the expense of large scale research in creating new basic technology. Bare in mind, however, that the production cost of the movie Spiderman 2 was a bit larger than the budget for the whole US fusion program the year it was released.
If she's interested in programming as an intellectual challenge, I think PLT-Scheme's DrScheme is your tool of choice. I've never used HtDP, but just following the built-in tutorials that come with DrScheme is very instructive. This is especially true if you are willing to set aside the goal of building some particular program. When I get bored, I will occasionally do a problem or two from the DrScheme tutorial. It's much more fun than dumb desktop games like Solitare. Personally, I think of it more as solving puzzles than as "coding." (It's also a Debian package, so life is good.)
If she wants something to tinker with, I suggest getting a little colo box (a UML box isn't too expensive) and hacking around a little with apache, perl, python, shell and web-ish stuff. It's not always strictly programming, but it's a lot of fun and results can be seen without a huge amount of time/knowledge/effort, and she'll have something tangible for her efforts. Well, as tangible as anything in computing can be, anyway.
I suggest you make a Knoppix CD with LTP installed. With a little configuration, that will take care of all of all your tests for the memory, disk, IO, and CPU. You might want to install America's Army or something to test the video subsystem.
If you put a little effort into it, you'll have a test suite as good as, and likely better than, anything you could pay money for. If you want to buy something, you can make a donation to the LTP and Knoppix projects.
There are also simpler tools, like Memtest86. I find this tool to be invaluable when I try to salvage old hardwar. I can't begin to tell you how much time it's saved me that I would have spent aimlessly swapping components around.
As a long-time Linux user, I agree completely.
With one caviat: You are describing the Slashdot ecology, not the Linux community in general. People come to Slashdot because they want to rant about something. For reasons I don't understand (tradition? the moderation system? Hemos?), it attracts all the hotheads in, well, just about every software community I can think of.
The wider *NIX community is actually pretty mellow, with a few well-documented exceptions. Theo de Raadt, for instance. He's a well known dickhead, but even he admits that. (And in his case, his acidic personality actually serves to make OpenBSD more secure, I think.)
Slashdot is fun, but don't take it too seriously, or as even as representative.
Stand up and be counted. SCO can only afford so much in legal expenses, and there are a lot of us. I'll happily sit around in a courtroom while they piss away their cash.
To: bstowell@sco.com
Subject: a request
From: Russell Neches
Hi there --
I've used Linux since 1998. I own and operate a variety of machines of various architectures on which I have installed the GNU/Linux operating system. Furthermore, I use a several Linux machines at school (as well as working as a volunteer administrator of said machines). Furthermore, I use and enjoy a large number of products and services that depend on Linux. This includes, but is not limited to, Akami hosting, Weta Digital movies, Merrill Lynch financial services, New York Times Co. media products, and T-Mobile wireless services (who happen to run my distribution of choice, Debian).
If you plan to continue your pursuit of licencing fees from Linux users, I am hereby declaring my intent to refuse to pay such fees. Please add me to your list of people against whom you wish to file suit.
I may be reached at xxxxxxx@xxx.xxx.xxx, or through my (Debian-using) cell phone service provider at xxx-xxx-xxxx. I shall furnish a mailing address once I have established one.
Russell Neches
...I think.
All right, I wasn't looking at the desk. It could be plywood and cinderblocks for all I care. Just goes to show that you can sell anything if you put a pretty enough girl next to it, and I'm only slightly ashamed to admit that it works on me.
Good thing I don't have any money.
What can I say. It was free and it runs Linux like a champ, aside from a needing trivial patch to the kernel source to prevent it from thinking it's running on a PIII. And it does the job.
Get a laptop. An old laptop. Install the weirdest OS you can find that has a networking stack. Make sure you have a couple of battaries that hold a charge so you can take it to the library, coffee shop or lobby while your roommate is busy contracting and spreading chlamydia, or whatever STD is popular on your campus.
Here's the reasoning: you want to make sure that you cannot play games on your computer. You know as well as I do that if you can play games, you will. Intead of doing your homework. I know whole Counter Strike clans that failed out of expensive private universty educations. You must avoid this fate at all costs.
Sound lame? Yeah, it is. But think of it this way. You (or your parents, or the government) is/are paying tens of thousands dollars a year to send you to a place where you can aquire an education. It's very likely that this is the only shot you're going to get, and that if you screw up bad enough, you've got a rewarding carrear in burger flipping.
That doesn't mean that you shouldn't have fun; on the contrary, you should have as much fun as you can. But, keep in mind that you are packed into a tiny, grubby place with thousands of other people your age, some of whom are worth getting to know. Keep in mind that there are proffesors and staff who've dedicated their lives to educating punks like you. Keep in mind that there is probably an interesting city or town to explore. Keep in mind that there is probably a gym that's flat-out better than any fitness company you could find that you can just use, for free. And you're probably miserably out of shape. Keep in mind that there is probably a world-class library crammed with books you should have already read by now. Exploit all of these things to the maximum extent permitted by hours in the day and callories in your diet, and maybe you'll get your money's worth.
As much as I like video games, they are mutually exclusive with these goals.
So, get an old laptop. Resist the urge to splurge on anything more ostentatious than a Pentium II 500. Your friends will laugh at it. Tell them you're poor, and that they should fuck off. Instead of playing games, amuse yourself with your creaky old hardware by hacking cool software. Or whatever you like, so long as you're creating something. You don't need fancy-pants graphics to run vim, screen, ssh, gcc, mutt, LaTeX and xterm. You might need a little more oomph for javac, or mzscheme, perl, or the like if your classes need 'em. Gaim, naim, or ICQ if it improves your social life. xmms, but don't go nuts on the P2P networks. It's a waste of your time. If your roommate wants to waste their time, mooch of of him or her.
Trust me. If you think you need anything else, you need to re-evaluate your goals.
And if you're lazy and/or adventurous, you can turn on NFS over TCP in your kernel and tunnel it over ssh or ppp/ssh. I've never tried it, but it ought to work. I understand that NFS over TCP is relatively untested, but is reputed to work rather well. Doing weird things like this would be a pretty good way to test the NFS over TCP code, and I'm sure the developers would be interested to hear how it goes. Particularly if you run a lot of data over it for a long time, and have a good way of verifying that all is well. Or, better still - if all is not well, and you have a good way of articulating what went wrong.
Of course, that doesn't mean it's a good idea. I think your solution with IPSec is much more elegant. Unfortunately, I happen to need to get through a heavily packet-shaped network that massively favors port 80, and drops random packets everywhere else. Not IPSec friendly at all. I avoid this by running multiple ppp/ssh tunnels through the retarded parts of the network and letting my gateway balance between them. Unfortunately, this requires privileged accounts on many, many boxes in odd places.
By the way, 10 points to any Northeastern University students who send polite, well considered complaints to Network Services. Not RESNet - they exist only to prevent you from talking to Network Services. Don't bother yelling at them - they exist specifically for that purpose. RESNet has no authority whatsoever to, for instance, allow CVS to work when Network Services decides to to drop 90 percent of packets on port 2401. This is for your benifit - I'm perfectly happy with my tunnels.
People might be interested in looking at the article Lessing is talking about, since the Red Herring article does not link to it. enjoy
It kind of makes sense, oddly enough. This error was thrown by a
preprocessor on PDP/11, if I remember the tale correctly. It was
induced when you screwed up your data types (float, int, +, or
something like that). In order to turn this into a valid assembly
instruction, you have to reverse the order of the elements (add $r1
$r2). If the preprocessor didn't have a function for generating the
proper assembly, it would hang. The program waiting for its output
would then report an unrecoverable IO error... while reading reverse
Polish notation backwards.
How's that?
Well, it'll be a great candidate once they figure out the itunesdb and other issues. There's been some progress, though. So, the answer is no for now. Uneless you consider OS X to belong to the set {Linux, OpenBSD}, which most people do not.
Build a gigantic secret base under an island somewhere. Equip with bad 70's decor, absurd super-weapons, henchmen, girls in swimsuits that inexplicably produce large caliber handguns when required, speed boats, hilicopters, and so forth. Put security cameras everywhere, and pay a midget or a guy with claws or something to monitor them. Obtain an absurd exotic pet, like a singing tapir or a farting oscilot. Train said pet to run errands about the island. Develop some mildly disturbing hobby, interest or passtime, such as bobbing for bat guano or hunting opera singers with a crossbow. The secret plans themselves, of course, should be encoded somehow in a fantastically ugly peice of fo-futurostic art. Where else?
Soft, by Rupert Thompson (ISBN 0-7475-3670-8). A beverage company uses subliminal messages, and "embassadors" to pitch their product. Very creepy book. Then again, it's a creepy idea.
- Check the source out of CVS
- Hack together something that lets you work around your window manager's key bindings.
- Post your patch to the SDL mailing list
So what if you're not a big-time programmer? Read the documentation, the man pages, and the FAQs. If you still need guidance, politely solicit assistance from the appropriate SDL mailing list. If you're coherant, polite and willing to learn, you will find that people will help you. I can't overemphasize how important it is to communicate like an adult. Every project wants helpful, compotent and motivated developers. Make it your goal to become such a person, and you'll receive a remarkable amount of help. (In contrast, if you whine and bitch on the mailing list, you will be ignored at best.)Yes, your first patch will probably suck. Big deal. Think of it as a "proof of concept." If it turns out to be a Good Idea, someone with more experience may clean it up a bit. Maybe that someone will be you, assuming you keep hacking. If all goes well, then your patch will be included in a release. Put it on your resume, pat yourself on the back, and enjoy gaming on your platform of choice.
Well, clearly there are many different ways of triggering cancer. What I'd like to know specifically how a 1.9 GHz signal can interact with parts of a cell, or any chemical used for signaling. If you can explain the mechanism by which a photon with a wavelength of 4 cm can interact directly with the chemical function of a cell, I'll be very impressed. Of course, the biology is quite complicated - but the physics isn't.
Wavelengths of that size don't have enough energy to disrupt even a weak chemical bond. Photons don't leave cancer-causing mojo just by passing through - the only way they interact with matter (for our purposes) is by absorbtion. So, if it's going to cause damage, it has to get absorbed by something. The only structures in the human body large enough to absorb that kind of signal are tissues, and the consiquence of absorbing such a photon would be heat.
All right, I suppose if you had a powerful enough signal, you could cook yourself. But at 0.75 watts, you're going to have to wait a long, long time. I suppose that raising the temperature in your head by a fraction of a degree could have an impact on intercellular (no pun intended) signalling, but I don't see how it would be different from a very slight fever, or sitting by a campfire. If you focused that 0.75 watts into a very, very small area, you might be able to cause a tiny radiation burn (which would also denature the neurotransmitters, almost certainly rendering them useless), but cell phones have omnidirectional antennae.
No matter how you look at it, the effect of this kind of radiation on the human body is going to be within the range of normal operation. It has to be : for a cell phone to give you cancer, the human body would have have to be more chemically unstable than nitroglycerine.
There may be other dangers from exposure to low frequency radiation (bad driving, rudeness, use of annoying ringtones... ^_^ ), but I'm pretty sure cancer isn't one of them.
My cell phone (the ubiquitous Nokia 3360) is a TDMA phone that operates in the IS-54 (800 MHz) and IS-136 (1.9 GHz) bands. Now, 1.9 GHz sounds like a big, scary cancer-causing number. So let's see if it really is.
First of all, we need to know how radiation causes cancer. We'll just assume it's electromagnetic radiation, since cell phones definitely do not emit anti-protons, neutrons, muons and other shit like that. There's no way in hell a battery the size of a Triscut can generate reaction energies high enough to produce hadrons or leptons, so we can forget about them. (Well, actually, with a big capacitor you might get a few, but you're already getting showered with cosmic rays, and the pathetic little fart of hadrons you'd get out of a cell phone battery wouldn't count for didly squat.) The cancer-causing mechanism for electromagnetic radiation is fairly simple. In order to be dangerous, a photon (the electromagnetic force carrier particle) needs to carry enough energy to ionize (chemistry parlance for "fuck up") something important. It doesn't really matter how many photons you're slinging around, since it's the frequency that determines the energy of a single quanta.
So, what is our hypothetical candidate cancer-causing quanta going to have to inonize to do the deed? Well, DNA of course. It's going to have to cause a genetic mutation. Because of the way photons interact with matter, they are most likely to be absorbed by electromagnetically contiguous objects of sizes roughly equal to their wavelength. The reasons are deeper than this, but suffice it to say that a photon is "smeared" over an area about the size of its wavelength. Since you can't absorb part of a quanta (that's why they're called quanta, after all), you have to have a thing big enough to soak up a whole particle about the size of the wavelength. In this way, everything is, or is made of, antennae. To cause a mutation, you have to have a photon whose wavelength is about equal to diameter of a DNA molecule. Actually, the ideal length of an antenna is a quarter the wavelength of its intended optimal frequency, so we'll say the wavelength we're looking for is four times the diameter.
So, as I said, my cell phone operates at 1.9 GHz, or 1.9 billion cycles per second. What's the wavelength? Well, wavelength is the period times the speed of light. The period is the the inverse of frequency, so :
3*10^8 / 1.9*10^9 ~= 0.16 M
That's about the length of your hand, give or take a thumb. One quarter of that is about 4 cm - about the length of your thumb, give or take a nail. Now ask yourself this question: How big is your DNA?
If your DNA is built out of atoms the size of rasins, you might have something to worry about. The diameter of the DNA helix is 2 nm and the vertical rise per base pair is 0.34 nm. If you want a photon that will be able to reliably zap DNA, it needs to have a wavelength _smaller_ than 8 nm. The probability that a photon will be absorbed by a given object decreases with respect to the difference between the size of the object and the wavelength of the photon according to the standard deviation. So what's the probability that a given photon spewing out of my cell phone is going to fry some of my DNA? Well, we're a factor of five million away from the optimal wavelength. I'd say it's pretty fucking unlikely.
But wait a second - what's kind of radiation has a wavelength of 8 namometers? Well, we do the opposite to find the frequency :
3*10^8 / 8*10-9 = 3.7*10^16
That's in the ultraviolet range. Surprise, surprise!
So, what can we conclude from this? Well, since a cell phone has a transmission power of less than a watt and a wavelength the size of your thumb, it's not going to do jack shit to your DNA. Nada. Zilch. In other words, THERE IS NO WAY CELL PHONE RADIATION CAN GIVE YOU CANCER!!! I'd be more inclined to beleive that the plastic in the earpiece causes cancer.
You're several orders of magnitude more likely to contract cancer as a result of proximity to a 100 watt incandecant light bulb. It's got a much, much higher output, and its frequency range is thousands of times higher.
So relax, enjoy your wireless technology, and wear your SPF-30.
You know, if you're going to argue with someone, it helps if you don't refute your own points.
- I was (rather obviosly) speaking about the historical state of mplayer as it pertains to this latest release.
- Of course support for binary only drivers is the responsibility of their respective developers. That's why I was drawing the comparison. The mplayer developers don't have time to deal with people's random binary distributions of their project. The kernel developers don't have time to deal with people's binary-only drivers. For essentially the same reasons.
- Yes, current versions have runtime CPU detection. I'm glad you were able to read the
.90pre1 changelog.
To everyone else - I apologize for feeding the troll.Their web site explains the licencing problems they've had pretty clearly.
.90 release, or they are migrating to GPL-friendly codecs.
Part of the reason is because they can't distribute binary packages. There are legal reasons, but at this stage of develpment it doesn't make sense for them to worry about bugs in binary distributions. Linux kernel developers will give you similar rhetoric about supporting binary drivers. It's technically doable, but it's so much more work than dealing with source that it's not worth their time.
There are also technical reasons (like a lack of runtime CPU detection, so you kinda have to compile it locally).
It also appears that they needed to use non-GPLed codecs in their earlier releases. I think they are either removing them in the
Read their website - it explains pretty much everything.
Well, if you filed a pattent, I could always just point at my post and say "prior art!"
and put the CD on the true north pole and read it from a...
never mind. that's retarded.
Basically, you hold the CD still in a little bracket, and spin a tiny little curved mirror around at the center. Since the laser will bounce erratically off the surface of the CD, you would read from the disk by placing a thin glass or plastic cover over the CD with a few photosensors sensors around its edge. The returning laser (carrying the data) could strike the cover at any pount, and the internal reflection of the cover would get enough of it to the photosensors to read the data. The laser will zip all over the place, so you'd use timing to ignore the data from non-contiguous parts of the disk. The mirror could be as small as the diameter of the laser, so you could spin it much, much faster than the CD iself could withstand.
The only problem I counld think of for such a device is that I don't think normal optical media will work as expected if you read it at a low angle.
Clearly, since no one seems to have done it, it's not that great of an idea. There's probably something wrong with it that I didn't think about at the time. Oh well - I was 12, and I just wanted to listen to Paul Simon without having to worry about bumping the desk while I was doing my cursed multiplication tables.
In this way, I suppose, Perl 6 will be kinda sorta like Java in that it will target a VM, rather than a CPU. The Parrot VM is not exclusive to Perl - from what I understand, idea is that eventually Python and Perl will share a common VM. It's conceivable, I suppose, that some sick bastard could write a Java compiler that targets Parrot.
Parrot isn't finished yet, but there are a few "toy" languages that target it (Jako and Cola, and more recently BASIC).
The nice thing about current motherboard fab processes is that they are cheap. Populating both side of the board is possible, but rather expensive. It requires more layers (which make the board more expensive to make and more difficult to debug). Also, it requires very different physical manufacturing facilities, since the presence of components on the reverse side makes putting pressure on the topside difficult. When the robots punch down components on the top, they could squash, crack or bend componets on the other if the board isn't sitting on a specially designed caddy. As you might imagine, it's possible to do the same thing when you have the board at home when installing componets. (Zero insertion force my foot!) There are also issues of cooling, since a the board would require airflow on both sides (which could mean a bigger case in a worst-case situation!).
Of course, if these issues can be addressed with a sane form factor standard, I think everyone would be happy to have smaller computers. Smaller is better, so long as it doesn't apply to things that would require you to buy an expensive car to compensate.
Personally, I think one of the major limitations of the ATX standard is the breakout. If there were a standard IO block that connected to the motherboard with a cable, you would have a lot more freedom when it came to positioning the board inside the case. Old AT motherboards did this for everything except the keyboard connector, I seem to recall. It would be a little more complicated to install, but I think it would be better.
Or, we could just make mass market PC104 stacks for everything.