Well, except for that whole ease-of-use, intuitive thing. Like you're going to see any average business secretary firing up
LaTeX and producing anything usable...
I'll take a functional tool over a pretty one. Same choice as power tools vs. Fisher-Price tools.
BTW, don't underestimate the intelligence of "average" people. I have - quite easily - taught two secretaries to use LaTeX. They were more productive than another who was using whatever version of Word came with Windows 3.1 (this was 1995), and I don't see that Word 2000 has anything significant over that version for doing the sorts of things that secretaries do.
Many of the classical puzzles are easily defined (like this one, the ol' can't trisect and angle with a straight-edge and compass,
etc.), which makes them exciting for non-geniuses like myself to study.
Number theory will always have deceptively simple sounding conjectures that are surprisingly hard. Try Goldbach's Conjecture - every even number is the sum of two primes. That has been a rich field for crackpots - er, I mean non-geniuses like you - to study.
If you're a little more ambitious you can try the Riemann Conjecture. That has all kinds of nice implications like the distribution of primes etc.
If I were to sneak into a recording studio and insert a 20 kHz low-pass filter and inject white noise at -90dB, would anyone notice?
No.
And that's the sort of thing you should think about. Not mythical triangle and square waves (which instruments produce those?) but what's the bandwidth and the S/N ratio of the whole system? From mikes to master tapes to mixers to your home amps and speakers. All this put end-to-end, what is the bandwidth and S/N ratio? That's all. Once you have that, you know what is the lowest sampling frequency you can use, and the lowest resolution you can use.
Hint: the system has a bandwidth of less than 20 kHz and an S/N ratio of less than 70 dB.
At and around 1/2 sampling freqency, the wave forms
become basically nothing but square waves [alternating between a single high, and a single low point]
This is such bullshit!
The Nyquist theorem says that if you have a bandwidth-limited signal with a bandwidth of n Hz, then sampling at 2n Hz recovers the signal exactly.
Drill that phrase into your heads: band-limited. Each time you start to construct mythical counterexamples with square waves, think of bandwidth. Think audio, and read up on Fourier transforms.
This means that if I take a music stream and pass it through a low-pass filter of 22.05 kHz, then there is no difference in the output stream whether I give you the low-pass filtered music or I sampled at 44.1 kHz, transmitted those samples, then passed them through a low-pass filter. None. No "basically nothing but square waves".
However, for a digital format like CD, there are other variables besides sampling frequency: what is the resolution? The higher the resolution, the better the S/N ratio of the system. Low resolution means that there will be a higher noise level because of aliasing. (Any reasonable audio system will use dithering when performing the D/A step - the analog of anti-aliasing in the visual world.)
Higher bit-resolution is not just an academic issue. CDs use 16-bit resolution, which is probably adequate for music - it represents an uncertainty of 1.5e-6 or around 98 dB. Using 24 bits would give us 104 dB (each additional 8 bits add about 6dB) but it would be ridiculously hard to make good A/D converters.
Another factor is the encoding curve - how are the sound levels translated to numbers? The CD format uses a 16-bit linear curve. The choice of a linear encoding was poor. Human ears have a log response curve i.e. we can hear differences in loudness better in soft sounds than in loud sounds. This is why sound levels are measured in dB, which is logarithmic. (A difference in 3dB represents a sound twice as loud.)
[The telephone industry has understood all this much better than the audio industry - the mu-law (and A-law, used in Europe) curve is logarithmic.]
Basically if you have a bandwidth limited signal with a certain S/N ratio, there is a sampling frequency and resolution you can use that will result in a signal indistinguishable from the original.
There are a few engineering issues: a perfect "brick-wall" low-pass filter cannot be made. When recording a CD one would probably use a low-pass filter that starts falling off around 18 or 20 kHz or use over-sampling at 2x or 4x the 22.05 kHz frequency. Inaccuracies in the clock when recovering a signal add noise. Etc. etc. etc.
In the case of something like a claimed algorithm for an
NP-complete problem, the person making the claim has a relatively easy way to demonstrate credibility: just implement the
thing and use it solve some hard problems such as the RSA factoring challenges.
This is a little harder than it seems: first, factoring is not in NP-complete. Ok, so let's take graph colouring - lots of people would be happy to see a P-time algorithm for any of the NP-complete graph problems.
The problem is that a P-time algorithm is not necessarily useful. If I give you a O(n^10^100) algorithm, you can't use it for anything but the smallest problems, like maybe a graph with 4 nodes.
Proving P=NP might not be useful in a practical sense.
...I have no way to know if you are going to be brain dead, paranoid or psychotic the next time you come to work.
...I've got no idea when you are going to be hauled off to jail. At least you did not hurt anyone at my
firm, but now I have to go find another employee.
Exactly. And if I hire women, I have no idea when they might not go and get pregnant - then I'll have to re-train another employee. And if I hire anyone who enjoys a drink, they might become an alcoholic or drive drunk and kill someone. And if I hire someone who's overweight, they might have a heart attack and drop dead and now I have to go find another employee. Same thing for people who ride motorcycles. Or like scuba diving. Or playing pool - their eyes may get poked by those long pointy sticks.
And oh yes, anyone who voices any political views that are not endorsed by the mainstream - what about them? They might go and join a protest march and get killed, or even worse, they might loot and pillage and get thrown in jail. Back to the job search.
"These are practical concerns, and good reasons to refuse to hire" these kinds of people.
Now let's look at periodic on the job testing....
Everyone will be better off if you can be "fixed".
Yes! And let's not forget Thoughtcrime either. This horrible problem that will lead to the downfall of society can be stopped with a simple measure - all TVs will have little cameras also. And the TVs can never be turned off, only down - but the cameras will stay on.
Everyone is very concerned about their rights. So they should be. But let me ask this.
How could the FBI perform their wiretaps in a legal manner, without enraging us all about our 'rights'?
Individual ISP's will install monitoring software capable of complying with court-ordered email taps. They will install whatever they think agrees with their conscience; one hopes, the open source version so they know exactly what its capabilities are.
For the consumer this doesn't necessarily mean anything better; but at least I will have a choice: I can choose an ISP that I believe will respect my constitutional rights. If the Federales get to install their black boxes everywhere, I don't have that choice.
...most criminals are truly not particularly
bright (else they would not be criminals)
Or rather, the dumb criminals get caught and make TV shows etc. The smart ones - if they're really smart, people won't even know that there had been a crime committed. (Like white-collar crime.)
But if I were planning some sort of crime, I wouldn't use a cypher; not even steganography. I'd use a code. All the details would be discussed in person, then I'd call (or send email) saying "I think I'm going to McDonald's today" meaning we hit the jewellery store, and "I think I'll go to Burger King" means it's the bank we'll roll. No wiretaps or Carnivore will catch that.
...but Sun has a better OS to put on the box. One thats more mature and reliable... Solaris X86...
Ugh! Solaris may be "more mature" (chronologically) but Solaris x86 is no match for Linux or any of the BSDs. The one strength that Solaris has is strong SMP support. Everything else - forget it! On the same hardware, Solaris x86 is much slower than Linux or *BSD.
And then of course Linux and the *BSD's are much nicer to use. They come with all the Gnu tools as standard whereas you have to install them separately on Solaris; and all the system calls make much more sense. The SunOS4 (BSD-based) -> Solaris2 (SysV based) switch was the worst thing Sun could have done, in my book. Yes, I know there were some good reasons for doing so; but on the whole SysV is baroque, bloated and clunky.
In a similar situation what I did was "find" some prior art that covered the central claim.
This effectively killed the application.
The thing is, the patent attorney has to take your
word for it. In my case there wasn't really a prior publication i.e. something that covered the "invention" exactly; I made an estimation, as a skilled practitioner in the field, that the extras covered by the claim would have been obvious to other practitioners in the field.
In my case I was the only "inventor" so I only had to deal with my own conscience (and loss of the $2500 bonus that was given to patents granted). I'm sure things would be a little more intricate if there were others involved.
Basically I refused to allow the lawyer to tell me what patent law "really" meant. (The lawyer was also a friend, and privately, outside work, he agreed with what I did.)
Please respect my copyright; as a writer, it's all I've got.
To this, I paraphrase Vaclav Havel:
Work for something because it is good, not just because it stands a chance to succeed.
This applies to all art. Painting, writing, sculpture, and yes, writing software. Only do something because not doing it would hurt you. I write code because if I didn't, I would be profoundly unhappy.
This is exactly what rms has been saying all along: it makes sense to talk about ownership of physical objects, because if I give you the apple I had, I can no longer eat that apple; but if I give you an idea I have, I still have that idea.
"Intellectual property" is a recent invention and only benefits those who are already rich and powerful.
there are about a kajillion different ways to do the same thing
Nothing wrong with that as such (or perhaps I should say "per se"...)
and I've never much cared for the way variables are declared and used in Perl (scalars?). Too much use of symbols, not enough grammar.
Amen!
This is a plug for Icon. You can do everything you can in Perl in Icon - with sane and human-readable syntax. Things are slightly different there; see the article from the Linux Journal.
To download this object-oriented, very high level language with garbage collection, X11 support, POSIX stuff (networking, file/directory manipulation etc.) go to the Unicon web page. It is available for Unix (Linux, FreeBSD etc.) and Windows-NT. It is, of course, free.
Ahhh, don't you just love the Closed-Mind Collectiveness of it all? Why not just provide a standard, external, API for binary drivers?
I assume you're talking about Linux and the rest of the kernel gang? It's really quite simple: there are many reasons why closed source drivers are a bad idea. Linux would sacrifice its prized jewel: its stability. Furthermore, the kernel just isn't ready to be frozen; so the device driver interface cannot be frozen. Besides, the kind of pinhead company that refuses to release programming specs because it doesn't care about those fringe nuts who use Linux certainly wouldn't be the sort of company that would spend the money to write a Linux driver for those fringe nuts!
Why don't hardware vendors realise that they are just hurting themselves by not documenting their products? To hell with the vendors providing open source drivers; just tell us how the damned card works, and we'll write the damn driver!
Why the hell do they think that by releasing information on how someone else can use the card, they will lose their "valuable" "intellectual property"? Bullshit!
I am a sys admin/support person and what I need from Linux is a system that allows my users to get on with their jobs, easily, securely and if possible boosts their work experience. Personally the easier and more accesible Linux is the better.
From where I sit, Linux iseasier. It supports all the hardware I have. When I buy new hardware, I make sure it's supported under Linux. Easy!
What kind of users are you sysadmin'ing that use exotic unsupported hardware?
I accept the free software ethos, but for me working, closed source drivers etc. are just plain more useful.
Don't use Linux, then.
It's that simple. If Linux doesn't meet your needs, don't use it. It meets my needs (and the ideology is, for me, something that I need) so I use Linux and I'm happy. If that means that I can't go buy the latest XBidia BeForce Rage Terror Anguish 2048 XLT video card, so be it.
... but if you (or your sister) spent five hours securing, maybe you wouldn't have had to spend twenty cleaning up.
Bullshit!
If I break into your house and look through your stuff - even if I didn't damage anything, not even the locks - I bet you're going to be pissed off. And rightly so.
I agree that if you want to put a system up on the internet you should learn about security and spend 5 hours securing the box. But I'm just so fucking sick and tired of the whole "blame the victim" mentality.
I'd even be tempted to help out if any sysadmins were planning to pay a late night visit with a two-by-four to a cracker's house.
All script kiddies will eventually end up probably as computer scinetists PhD.
I am a "computer scinetist PhD". When I was a kid, I had a fascination with blowing up things; but I made the explosives myself. I didn't buy firecrackers and set them off under cars; I taught myself chemistry and almost killed myself many times with my "experiments". It would have been totally without honor to just use some sort of pre-packaged destructive device.
You know what else? The script kiddies of my youth are now pathetic losers in dead-end jobs who beat their kids.
I'm gonna encourage my kids to be script kidies.
I feel sorry for them and for you. Why not encourage them to learn about electronics and build their own hardware? And then to learn how to program, and write code for the hardware they built? (And while you're at it, make sure they learn spelling and grammar!)
Do you also plan to encourage them to break into homes and shops as long as they leave "good and fair" people alone?
Consider how many "innovations" m$ has released, intending to replace something else. Disreguarding operating systems, that is. How many of those innovations actually made the replacement?
It is to larff!
And does anyone else think it funny that for a language that slices, dices, is a floor wax and a dessert topping, the language manual is only available as an executable?
(I suppose it's a Micros**t self-extracting archive or some such.)
This is purely an issue with the FCC and the choices/rules of individual airlines. There is NOTHING in the FARs (Federal Aviation Regulations) that restricts the use of any electronics while in flight. The only restriction is what the PIC (Pilot in Command) deems will not interfere with the safe operation of the aircraft.
FAR Part 91 (general aviation) says PIC determines what electronics may be used.
FAR Part 121 (scheduled air carriers) restricts electronic devices. (121.305) The operator is the one who makes the determination, not the PIC - and the operator is the airline. These restrictions are usually in the airline's Ops. Specs., the operating manual; for a Part 121 airline, the ops specs have the full weight of FAR's. (Of course the PIC may forbid something not explicitly forbidden by the FAR's or the ops specs.)
The FCC regulation covers the AMPS cellular service; it does not cover - for instance - CDMA PCS.
well, GPS approach is becoming more widespread (Truckee-Tahoe even has one, and they don't have a tower!) and ILS isn't used as much.
There are no precision GPS approaches (yet). For a precision approach (i.e. an approach where vertical guidance is provided) GPS will require WAAS or LAAS. I bet that GPS approach into Truckee doesn't get you any lower than about 600' AGL.
do you know if GPS is as suceptable to this sort of interference?
There is a well-known problem with the King KX-155 (a common navigation/communication radio used a lot in general aviation) when tuned to certain NAV frequencies causes GPS receivers to lose satellites. The KX-165 seems to not have this problem. In the airplanes I fly no electronic devices have caused any nav/comm interference but I absolutely won't allow anyone to use them if we're IMC! [in the clouds]
The local oscillator leakage caused a false indication on one of the ILS (Instrument Landing System) instruments (I think it was called the localizer).
Localizers use frequencies around 110 MHz. Consumer FM receivers use a first intermediate frequency (IF) of 10.7 MHz. What this means is that if you tune an FM radio to KOZY 101, the radio transmits noise at 101.1 + 10.7 = 111.8 - which is bang in the middle of the navigation band.
Fortunately, all commercial jetliners now use inertial navigation systems, so this particular failure mode is much less likely.
Inertial is not used for approaches, only for en-route navigation. The problem with inertial systems is that they drift, so the longer you've been airborne, the more inaccurate they are. An error of about a mile is no problem up in the stratosphere, but you can't be a mile off the runway when you touch down...
Right now ILS is the only precision approach available in the vast majority of cases. Once the FAA puts up WAAS and LAAS augmentation for GPS, aircraft can use GPS for precision approaches.
AM is used by aircraft because the strongest carrier signal is heard, so even if someone sits on a microphone, the tower can still talk to other aircraft.
False.
FM modulation results in "signal capture" i.e. the strongest signal gets through. With AM what you get is "heterodyning" [look it up] which results in an annoying squeal for everyone listening on that frequency.
Pilots know this very well, since ATC communications are AM. Every now and then two people will speak together (referred to as "stepping on someone else") followed by someone saying "Blocked" or similar.
AM is bad for blocking interference also. You can hear thunderstorms on AM radio.
A very common problem with electronics close to each other is the front-end of a receiver getting overloaded by a strong RF signal. With a mobile phone this is unlikely but possible - and as someone else said, aircraft radios were never certified against mobile phones and their frequencies.
>
stupid stuff with no evidence of design or consistency. (And I know what > I'm talking about, I've had to write lots of VB code at work.)
> Strange, but I have written lots of VB code too yet my code is elegantly > designed, consistant, and robust.
Please work on your comprehension skills; I said the language (if you can call it that... where is the spec? MS changes it with every release of Visual Studio!) sucks, not that all code written using VB sucks. I was writing "elegant, consistant [sic], and robust code" in IBM 360 assembly and FORTRAN some 25 years ago. (Now I prefer Unicon).
BASIC sucks. VB sucks so bad it makes me fear for my society, for surely if God has any sense of aesthetics, he will not allow the society to live that allowed VB to be created.
VB is not a programming language. It is a cancerous agglutination of stupid stuff with no evidence of design or consistency. (And I know what I'm talking about, I've had to write lots of VB code at work.) Also, kids are smarter than you think. They're learning natural languages, a few funny characters like { and } doesn't faze them at all.
As far as the feared GOTO keyword, I never need to use it in VB
The goto is just a tool; use it where it makes sense. If you have three levels of nested blocks and you need to exit from the innermost, you shouldn't be writing code, son - go back to school and learn about design.
Really, only one thing needs to be said here: LOGO. It's freely available on all kinds of platforms, kids like it, and has very few stupid things about it. And UCB Logo is open source! Go look at http://http.cs.berkeley.edu/~bh/. On MS-Windows, MSW Logo.
If you really like procedural languages, have you considered Unicon, an extension of Icon that has integrated graphics, is object-oriented and has POSIX system calls? It even runs on NT! I've taught ten-year-olds to use it and they loved it. (Of course it runs under all kinds of Unix machines.)
For some reason people like writing software but find documentation a chore.
"People"? Who?
I know people who like writing documentation but couldn't write code. We need to cultivate these people. Attitudes like "people like writing software but find documentation a chore" can only hurt us.
I like writing code, and guess what? I like writing documentation too. (If any developers out there want someone to write docs, send me email. If I find your project interesting, I'd love to document it!)
Slashdot Mirror
BTW, don't underestimate the intelligence of "average" people. I have - quite easily - taught two secretaries to use LaTeX. They were more productive than another who was using whatever version of Word came with Windows 3.1 (this was 1995), and I don't see that Word 2000 has anything significant over that version for doing the sorts of things that secretaries do.
Har har! It is to larff!!
Have you looked at the HTML that MS applications generate? Why do you think programs like HTML De-Moroniser exist?
Number theory will always have deceptively simple sounding conjectures that are surprisingly hard. Try Goldbach's Conjecture - every even number is the sum of two primes. That has been a rich field for crackpots - er, I mean non-geniuses like you - to study.
If you're a little more ambitious you can try the Riemann Conjecture. That has all kinds of nice implications like the distribution of primes etc.
If I were to sneak into a recording studio and insert a 20 kHz low-pass filter and inject white noise at -90dB, would anyone notice?
No.
And that's the sort of thing you should think about. Not mythical triangle and square waves (which instruments produce those?) but what's the bandwidth and the S/N ratio of the whole system? From mikes to master tapes to mixers to your home amps and speakers. All this put end-to-end, what is the bandwidth and S/N ratio? That's all. Once you have that, you know what is the lowest sampling frequency you can use, and the lowest resolution you can use.
Hint: the system has a bandwidth of less than 20 kHz and an S/N ratio of less than 70 dB.
The Nyquist theorem says that if you have a bandwidth-limited signal with a bandwidth of n Hz, then sampling at 2n Hz recovers the signal exactly.
Drill that phrase into your heads: band-limited. Each time you start to construct mythical counterexamples with square waves, think of bandwidth. Think audio, and read up on Fourier transforms.
This means that if I take a music stream and pass it through a low-pass filter of 22.05 kHz, then there is no difference in the output stream whether I give you the low-pass filtered music or I sampled at 44.1 kHz, transmitted those samples, then passed them through a low-pass filter. None. No "basically nothing but square waves".
However, for a digital format like CD, there are other variables besides sampling frequency: what is the resolution? The higher the resolution, the better the S/N ratio of the system. Low resolution means that there will be a higher noise level because of aliasing. (Any reasonable audio system will use dithering when performing the D/A step - the analog of anti-aliasing in the visual world.) Higher bit-resolution is not just an academic issue. CDs use 16-bit resolution, which is probably adequate for music - it represents an uncertainty of 1.5e-6 or around 98 dB. Using 24 bits would give us 104 dB (each additional 8 bits add about 6dB) but it would be ridiculously hard to make good A/D converters.
Another factor is the encoding curve - how are the sound levels translated to numbers? The CD format uses a 16-bit linear curve. The choice of a linear encoding was poor. Human ears have a log response curve i.e. we can hear differences in loudness better in soft sounds than in loud sounds. This is why sound levels are measured in dB, which is logarithmic. (A difference in 3dB represents a sound twice as loud.) [The telephone industry has understood all this much better than the audio industry - the mu-law (and A-law, used in Europe) curve is logarithmic.]
Basically if you have a bandwidth limited signal with a certain S/N ratio, there is a sampling frequency and resolution you can use that will result in a signal indistinguishable from the original.
There are a few engineering issues: a perfect "brick-wall" low-pass filter cannot be made. When recording a CD one would probably use a low-pass filter that starts falling off around 18 or 20 kHz or use over-sampling at 2x or 4x the 22.05 kHz frequency. Inaccuracies in the clock when recovering a signal add noise. Etc. etc. etc.
For an excellent treatment see http://www.vgard.net/digital aud io/digitalaudio.html.
This is a little harder than it seems: first, factoring is not in NP-complete. Ok, so let's take graph colouring - lots of people would be happy to see a P-time algorithm for any of the NP-complete graph problems.
The problem is that a P-time algorithm is not necessarily useful. If I give you a O(n^10^100) algorithm, you can't use it for anything but the smallest problems, like maybe a graph with 4 nodes.
Proving P=NP might not be useful in a practical sense.
Exactly. And if I hire women, I have no idea when they might not go and get pregnant - then I'll have to re-train another employee. And if I hire anyone who enjoys a drink, they might become an alcoholic or drive drunk and kill someone. And if I hire someone who's overweight, they might have a heart attack and drop dead and now I have to go find another employee. Same thing for people who ride motorcycles. Or like scuba diving. Or playing pool - their eyes may get poked by those long pointy sticks.
And oh yes, anyone who voices any political views that are not endorsed by the mainstream - what about them? They might go and join a protest march and get killed, or even worse, they might loot and pillage and get thrown in jail. Back to the job search.
Yes! And let's not forget Thoughtcrime either. This horrible problem that will lead to the downfall of society can be stopped with a simple measure - all TVs will have little cameras also. And the TVs can never be turned off, only down - but the cameras will stay on.Let us never forget:
-sIndividual ISP's will install monitoring software capable of complying with court-ordered email taps. They will install whatever they think agrees with their conscience; one hopes, the open source version so they know exactly what its capabilities are.
For the consumer this doesn't necessarily mean anything better; but at least I will have a choice: I can choose an ISP that I believe will respect my constitutional rights. If the Federales get to install their black boxes everywhere, I don't have that choice.
Or rather, the dumb criminals get caught and make TV shows etc. The smart ones - if they're really smart, people won't even know that there had been a crime committed. (Like white-collar crime.)
But if I were planning some sort of crime, I wouldn't use a cypher; not even steganography. I'd use a code. All the details would be discussed in person, then I'd call (or send email) saying "I think I'm going to McDonald's today" meaning we hit the jewellery store, and "I think I'll go to Burger King" means it's the bank we'll roll. No wiretaps or Carnivore will catch that.
And then of course Linux and the *BSD's are much nicer to use. They come with all the Gnu tools as standard whereas you have to install them separately on Solaris; and all the system calls make much more sense. The SunOS4 (BSD-based) -> Solaris2 (SysV based) switch was the worst thing Sun could have done, in my book. Yes, I know there were some good reasons for doing so; but on the whole SysV is baroque, bloated and clunky.
The thing is, the patent attorney has to take your word for it. In my case there wasn't really a prior publication i.e. something that covered the "invention" exactly; I made an estimation, as a skilled practitioner in the field, that the extras covered by the claim would have been obvious to other practitioners in the field.
In my case I was the only "inventor" so I only had to deal with my own conscience (and loss of the $2500 bonus that was given to patents granted). I'm sure things would be a little more intricate if there were others involved.
Basically I refused to allow the lawyer to tell me what patent law "really" meant. (The lawyer was also a friend, and privately, outside work, he agreed with what I did.)
To this, I paraphrase Vaclav Havel:
This applies to all art. Painting, writing, sculpture, and yes, writing software. Only do something because not doing it would hurt you. I write code because if I didn't, I would be profoundly unhappy.
This is exactly what rms has been saying all along: it makes sense to talk about ownership of physical objects, because if I give you the apple I had, I can no longer eat that apple; but if I give you an idea I have, I still have that idea.
"Intellectual property" is a recent invention and only benefits those who are already rich and powerful.
-s
Nothing wrong with that as such (or perhaps I should say "per se"...)
Amen!
This is a plug for Icon. You can do everything you can in Perl in Icon - with sane and human-readable syntax. Things are slightly different there; see the article from the Linux Journal.
To download this object-oriented, very high level language with garbage collection, X11 support, POSIX stuff (networking, file/directory manipulation etc.) go to the Unicon web page. It is available for Unix (Linux, FreeBSD etc.) and Windows-NT. It is, of course, free.
-s
I assume you're talking about Linux and the rest of the kernel gang? It's really quite simple: there are many reasons why closed source drivers are a bad idea. Linux would sacrifice its prized jewel: its stability. Furthermore, the kernel just isn't ready to be frozen; so the device driver interface cannot be frozen. Besides, the kind of pinhead company that refuses to release programming specs because it doesn't care about those fringe nuts who use Linux certainly wouldn't be the sort of company that would spend the money to write a Linux driver for those fringe nuts!
Why don't hardware vendors realise that they are just hurting themselves by not documenting their products? To hell with the vendors providing open source drivers; just tell us how the damned card works, and we'll write the damn driver!
Why the hell do they think that by releasing information on how someone else can use the card, they will lose their "valuable" "intellectual property"? Bullshit!
From where I sit, Linux iseasier. It supports all the hardware I have. When I buy new hardware, I make sure it's supported under Linux. Easy!
What kind of users are you sysadmin'ing that use exotic unsupported hardware?
Don't use Linux, then.
It's that simple. If Linux doesn't meet your needs, don't use it. It meets my needs (and the ideology is, for me, something that I need) so I use Linux and I'm happy. If that means that I can't go buy the latest XBidia BeForce Rage Terror Anguish 2048 XLT video card, so be it.
Bullshit!
If I break into your house and look through your stuff - even if I didn't damage anything, not even the locks - I bet you're going to be pissed off. And rightly so.
I agree that if you want to put a system up on the internet you should learn about security and spend 5 hours securing the box. But I'm just so fucking sick and tired of the whole "blame the victim" mentality.
I'd even be tempted to help out if any sysadmins were planning to pay a late night visit with a two-by-four to a cracker's house.
I am a "computer scinetist PhD". When I was a kid, I had a fascination with blowing up things; but I made the explosives myself. I didn't buy firecrackers and set them off under cars; I taught myself chemistry and almost killed myself many times with my "experiments". It would have been totally without honor to just use some sort of pre-packaged destructive device.
You know what else? The script kiddies of my youth are now pathetic losers in dead-end jobs who beat their kids.
I feel sorry for them and for you. Why not encourage them to learn about electronics and build their own hardware? And then to learn how to program, and write code for the hardware they built? (And while you're at it, make sure they learn spelling and grammar!)
Do you also plan to encourage them to break into homes and shops as long as they leave "good and fair" people alone?
-s
It is to larff!
And does anyone else think it funny that for a language that slices, dices, is a floor wax and a dessert topping, the language manual is only available as an executable?
(I suppose it's a Micros**t self-extracting archive or some such.)
FAR Part 91 (general aviation) says PIC determines what electronics may be used.
FAR Part 121 (scheduled air carriers) restricts electronic devices. (121.305) The operator is the one who makes the determination, not the PIC - and the operator is the airline. These restrictions are usually in the airline's Ops. Specs., the operating manual; for a Part 121 airline, the ops specs have the full weight of FAR's. (Of course the PIC may forbid something not explicitly forbidden by the FAR's or the ops specs.)
The FCC regulation covers the AMPS cellular service; it does not cover - for instance - CDMA PCS.
There is a well-known problem with the King KX-155 (a common navigation/communication radio used a lot in general aviation) when tuned to certain NAV frequencies causes GPS receivers to lose satellites. The KX-165 seems to not have this problem. In the airplanes I fly no electronic devices have caused any nav/comm interference but I absolutely won't allow anyone to use them if we're IMC! [in the clouds]
Localizers use frequencies around 110 MHz. Consumer FM receivers use a first intermediate frequency (IF) of 10.7 MHz. What this means is that if you tune an FM radio to KOZY 101, the radio transmits noise at 101.1 + 10.7 = 111.8 - which is bang in the middle of the navigation band.
Inertial is not used for approaches, only for en-route navigation. The problem with inertial systems is that they drift, so the longer you've been airborne, the more inaccurate they are. An error of about a mile is no problem up in the stratosphere, but you can't be a mile off the runway when you touch down...Right now ILS is the only precision approach available in the vast majority of cases. Once the FAA puts up WAAS and LAAS augmentation for GPS, aircraft can use GPS for precision approaches.
False.
FM modulation results in "signal capture" i.e. the strongest signal gets through. With AM what you get is "heterodyning" [look it up] which results in an annoying squeal for everyone listening on that frequency.
Pilots know this very well, since ATC communications are AM. Every now and then two people will speak together (referred to as "stepping on someone else") followed by someone saying "Blocked" or similar.
AM is bad for blocking interference also. You can hear thunderstorms on AM radio.
A very common problem with electronics close to each other is the front-end of a receiver getting overloaded by a strong RF signal. With a mobile phone this is unlikely but possible - and as someone else said, aircraft radios were never certified against mobile phones and their frequencies.
-s
VB is not a programming language. It is a cancerous agglutination of stupid stuff with no evidence of design or consistency. (And I know what I'm talking about, I've had to write lots of VB code at work.) Also, kids are smarter than you think. They're learning natural languages, a few funny characters like { and } doesn't faze them at all.
The goto is just a tool; use it where it makes sense. If you have three levels of nested blocks and you need to exit from the innermost, you shouldn't be writing code, son - go back to school and learn about design.
Really, only one thing needs to be said here: LOGO. It's freely available on all kinds of platforms, kids like it, and has very few stupid things about it. And UCB Logo is open source! Go look at http://http.cs.berkeley.edu/~bh/. On MS-Windows, MSW Logo.
If you really like procedural languages, have you considered Unicon, an extension of Icon that has integrated graphics, is object-oriented and has POSIX system calls? It even runs on NT! I've taught ten-year-olds to use it and they loved it. (Of course it runs under all kinds of Unix machines.)
-s
"People"? Who?
I know people who like writing documentation but couldn't write code. We need to cultivate these people. Attitudes like "people like writing software but find documentation a chore" can only hurt us.
I like writing code, and guess what? I like writing documentation too. (If any developers out there want someone to write docs, send me email. If I find your project interesting, I'd love to document it!)