Two things: 1. If you can type on your iPhone as fast as you can on your laptop, then you are a slow typist, so your opinion on speed really doesn't matter. You'll probably never get 60 WPM on your iPhone, and definitely not the 100+ WPM that a "power-typists" have.
2. Predictive text is incredibly useful, but only up to a point. Unfortunately, it sucks if you aren't typing normal english words. For example, try typing this message on your iPhone: "Dear Svrisathana, the composer's name is Shostakovich, and it cannot be spelled Schostakovitch or Shostackovitsch. Will we meet at AX-2470Y or AY-2470X?"
(To say nothing of foreign languages. Try typing this on your phone: "Ni3 hao3, wo3 shi4 fei1chang2 ben4 de5 lao3wai4")
Predictive text is nearly useless for the bulk of programming -- sure, variable and function names, auto-suggest, etc. are great, but since the general keywords in languages are short, and punctuation/non-letter keys are very important, touch-typists have a huge advantage.
Yes, but think of it this way. That 170M is the pool of potential army members -- they are not all going to be fit for the job, and some will be better than others. Prune the tree. Take the best 1.7 million from those and you've got the top 1% killers, the elite. And that's a lot of fucking badass killers.
Can you actually name some commercial software that doesn't have an OSS counterpart?
Let's break this question into 2 parts.
Commercial software without any OSS counterpart.
So far as I know, Reason by Propellerhead has no counterpart. I am certain that Wenlin, the chinese language learning software, has no OSS counterpart, and unfortunately I never had the money to purchase it when I was studying Chinese...
Commercial software without any GOOD OSS counterpart.
Such as Pro Tools (come on, Ardour -- not really useable for professional work). Sibelius & Finale have no useable counterpart.
That's just a small list from my background (music arranging and multitracking), but it is a field where OSS is really lacking. You simply can't do professional level work in Linux at this moment. But even in my experience, sure there are counterparts for software like iPhoto or iTunes, but none of it is really as well done. I tend to support a decent amount of shareware (I spend maybe $60 a year) because I am unsatisfied with open source options, at least things that I can use on my mac.
I wholeheartedly agree w/r/t complex software. Further, if you look at organic artifacts, even something as simple as a knife, the tool is always changing. The knife gets duller, and you sharpen it (removing molecules), and so it is constantly changing.
I think a lot of people say "uh oh, when are cars going to start getting the BSOD?", but what they don't realize is that, as complicated mechanical entities, cars DO give the equivalent of a BSOD. The word "crash" has a real-life meaning, too, after all:) Tires blow, mufflers fall off, things are changing all the time.
At least software is not subject to the laws of entropy (or is it?!)....
wholeheartedly agree. can't stand it myself, though plenty of people enjoy it I guess. maybe to them, having played fantasy games so long, the rogue/dwarf/wizard whatever is familiar turf like PBJ on wonderbread.
I think you're being a little harsh. Regarding the realtors, I think you have a couple good points, but this does not negate the forces of self-interest, described in Freakonomics, that would be likely to motivate a realtor, e.g.
The study found that an agent keeps her own house on the market an average ten extra days, waiting for a better offer, and sells it for over 3 percent more than your house--or $10,000 on the sale of a $300,000 house. That's $10,000 going into her pocket that does not go into yours, a nifty profit produced by the abuse of information and a keen understanding of incentives. The problem is that the agent only stands to personally gain an additional $150 by selling your house for $10,000 more, which isn't much reward for a lot of extra work.
So there is a clear reason why real-estate agents would hurry to sell someone else's house for a little less (because the commission is about the same). It's not a conspiracy, it's a simple force of the market. If I got into being a real-estate agent without ever contacting another real-estate agent, these kinds of tactics would be clear to me if I ever actually thought about how to maximize my profits.
Regarding drug dealers, you say the following: The good professor takes perfectly valid data and draws wild and crazy conclusions from them, when simple explanations will do. Are there really crack dealers out there risking their lives for pennies? Perhaps there are... but given his sloppy methods in the rest of the book, I find myself taking the entire shebang with a bushel of salt.
First of all, I don't think he drew any conclusions at all regarding the wages of drug dealers. Instead, he presented the actual data from the drug lord's financial notebook. A little division and rounding, e.g.,
Here are the monthly totals for the wages that J. T. paid his gang members:
Combined wages paid to all three officers: $2,100
Combined wages paid to all foot soldiers: $7,400
Total monthly gang wages (excluding leader): $9,500
So J. T. paid his employees $9,500, a combined monthly salary that was only $1,000 more than his own official salary. J. T.'s hourly wage was $66. His three officers, meanwhile, each took home $700 a month, which works out to about $7 an hour. And the foot soldiers earned just $3.30 an hour, less than the minimum wage. So the answer to the original question--if drug dealers make so much money, why are they still living with their mothers?--is that, except for the top cats, they don't make much money.
The top 120 men in the Black Disciples gang represented just 2.2 percent of the full-fledged gang membership but took home well more than half the money.
Please let me know what kind of crazy conclusions he is drawing from those numbers, or what the sloppiness you see in that method is.
Also, if you see any "blatantly wrong answers", let me know. I have the entire book with me, searchable. I have no strong reason for defending it except that I enjoyed the book and think about the conclusions a lot. If there are gaping holes in that, sure, I'd like to know, but so far you haven't really presented a strong case for that.
It is in nasty RTF format, but readable; you will certainly want to buy the book to get full effect. That said, nothing like a little instant gratification...
If you are vegan, you may be interested in the non-cheesy parts of the following URL at the same random domain:
Read freakonomics. Really. In addition to being a generally a good book, it does talk about the wages that drug dealers make.
All but the highest rungs of drug "employees" (i won't say dealers, since dealers are usually not in the high echelons) do not make much money, and in fact do live at home with their parents and barely survive on their wages.
If you are truly interested in questions like "what kind of money do drug dealers make", I do highly suggest reading this book. I'm not affiliated with them or anything, I just think it was interesting and really insightful regarding a lot of these social taboo subjects.
I don't think this discussion is really about remixes, which could potentially fall into fair use. Instead, it is about mix tapes, which are compilations of various tracks put together by a single person.
E.g., if I make a CD for a friend and put some Madonna on there, some Radiohead, a Metallica track, I am putting whole copyrighted works (entire songs) on a compilation.
As far as I know this is clearly illegal, though many many people (including myself) think they are ethical, and thus continue to break the law. Usually no legal action is taken, but in this case the RIAA has taken some very serious action against this hip-hop dude.
There is nothing either ethical or legal about mixtapes
Well, there may be no legal grounds for them, but we are in totally different moral worlds if you think that mix tapes are unethical. There are plenty of people in both camps -- I'm not saying you're wrong, just that it's not absolutely clear that mixtapes are unethical.
A lot of artists (and consumers) think mixtapes are great from all standpoints.
Any time you pick a particular age as a global standard, it's not necessarily fair, particularly if that age is as high as 15. When I was 17 I dated a 14-year old, who was 5'9 and believe me, as fully developed as at least my 17 year old peers. On the other hand, there are 15 or 16 year olds who are less developed than some 13 year olds.
Puberty age, as well, is decreasing globally, so that people are maturing earlier and earlier.
When I was 14, I was interested in people about my own age. Would it be so unethical for me, at that age, to want to look at pictures or movies of people my own age? What about when I was 12? 10? 8? It's unclear territory, I think. While the idea of pornography including people who are pre-sexual (aka children) is appalling to me, and while I personally wouldn't even be interested in young teens, not everybody shares my mentality.
I don't offer any solutions. I certainly think that pornography involving the coercion of pre-sexual humans should be eliminated, legislated, with high penalties. (On the other hand, when I was 8 I would have happily done porn if given the chance). Anyway, I just think you are a little too straight-laced on this one, thinking there is an absolute gold standard which should be globally legislated. The truth is much more slippery and I like the idea of individuals -- from children to senior citizens -- being able to make up their own minds about the matter, not have their opinions and desires be legislated and governed globally.
You and about 500 other luddites in the world, who want all their appliances to do one function (and to do it well, of course). Everyone else, I think, recognizes the need for a little convergence around here.
And since this phone realizes the beautiful convergence of WiFi tablet, phone, music player, and camera, it is getting a ton of attention. I think this is fully deserved, considering what a wonderful job Apple has done with the design; this is the first solution I've ever seen that looks like it's going to work and work well for a lot of people.
(I am one of those people for whom the design is near-perfect [within the technology constraints of the present -- I'd love a 600GB hard drive in there])
There comes a point when there are too many functions to do to justify one function per device.. in these days of printing, scanning, picture taking, emailing, voice mailing, calling, web browsing, navigating streets, portable music, disc ripping, disc burning, etc. etc., we simply need devices that can do more than one of these functions (the PC is a great example of "getting it right")
I don't really think these physical property analogies are valid. If I rent a car and don't return it, the car rental place has just lost a car, and lost money. Same with the Blockbuster physical media.
But what if I rent the blockbuster DVD, rip it, and return the DVD? Nobody loses anything. This is why data/IP is a totally different game from physical property.
Yeah, there should totally be "consolidate" type option, similar to what you do when you mix down a pro tools session, just to get a document that is portable, which you can open but don't have much editing power on.
Regarding large file sizes with images: both ms-word and powerpoint files are documents meant for editing as well as viewing -- so they have to store more information about the data contained inside than, say, pdf.
Let's take the simple case -- you drag in a 640x640 image, resize it to 30x30, then decide it's too small, and resize it to 100x100. Those are significantly different sizes. If word immediately resized the source image so it only had 30x30 pixels (and a smaller file size), when you tried to expand it to 100x100 you wouldn't have any picture quality.
So, sure, you want it to remember what the original image was when you're working in the document. But just because you save the file, quit, and re-open the file, does that mean you're done with resizing? Since word and ppt documents are always work-in-progress type formats, they have to store the original images and objects.
What I'm saying is not just relevant, it's crucial to Nyquist's theorem an digital sampling.
There is, of course, no way to tell what the original waveform was with 15k and 3 samples per crest. But, it is forced into a sine wave because of bandlimiting. It is the harmonics of that 15k fundamental which would make it a square wave, a triangle wave, etc. Those harmonics are filtered. Everything above 22.05kHz is removed. The square wave is turned into a sine wave. The triangle wave is turned into a sine wave.
And that's what we want. Because if you followed the rules of bandlimiting when you encoded the signal, only a 15kHz sine wave would have been encoded (you could have fed it a square wave, but all the harmonics that made it square would have been chopped off). At 15kHz only the sine-wave fundamental frequency would make it in to the encoded signal, and only a 15kHz sine wave will come out.
Does this make more sense to you? I think what you're not understanding is that a wave shape other than sine wave simply means that there is more than one frequency to that signal.
That would be a pretty funny comment if I knew that you were only kidding around...??
Really, if you don't understand how to turn a square wave into a sine wave (by bandlimiting), you should read up on it. You will then understand why 18k in a 22kHz bandlimited encoding system has to be a sine wave.
You could start with Sine Waves. But the really important quote to take away from there is:
Any non-sinusoidal waveforms, such as square waves or even the irregular sound waves made by human speech, are actually a collection of sinusoidal waves of different periods and frequencies blended together. The technique of transforming a complex waveform into its sinusoidal components is called Fourier analysis.
This page kind of talks about how square waves are made up of sine waves.
I generally avoid seedy places, too. I was kind of assuming my hypothetical $1 coke movie theatre wasn't a complete shithole. In fact, my whole little fantasy revolved around a nice, charming movie theatre that didn't jack up prices (and since I know I can get a 32oz fountain coke from a nice, non-seedy local store for 50 cents, I assume a movie theatre could be nice and have $1 cokes)
The point is more don't artifically inflate prices... even if (gasp) you won't maximize your immediate profit by doing so. You'll certainly maximize your immediate karma by having reasonable prices:)
I think you might not understand what that actually means. A square wave is a waveform which contains the fundamental and all odd integer harmonics. So if we have a square wave at 10kHz (F=10k), that waveform contains 10kHz (1 * F), 30kHz(3*f), 60kHz(5*f), etc. Each of its component frequencies is, of course a sine tone. Read up on Fourier Transform if you're not following.
So let's take that 10khz square wave again, which has as its frequencies 10kHz, 30kHz, 60kHz, etc. When we perform a band-limit on that square wave, only allowing frequencies below 22kHz, we are left only with the 10kHz sine wave.
I hope that you can extrapolate this to the other waveforms -- sawtooth, triangle, etc., which are comprised of the fundamental and various integer harmonics. The point is that by the time a frequency is high enough to only have a few sample points, such that you could not tell what the original waveform was, the post-decoding bandlimiting will force it to be a sine tone (which is what it inevitably should be, if you followed all the encoding constraints a la Nyquist's theorem).
Right, but what's the first harmonic of a 16kHz tone? 32kHz. That is why the bandlimiting step after decoding is so important, and that's why Nyquist states that in order to reproduce signals up to the Nyquist frequency you must bandlimit!
So what's the difference between a 16kHz sine wave and a 16kHz square or triangle wave that's band-limited to 22kHz?
(Hopefully none! So aliasing isn't that big of a deal, right?)
Slashdot Mirror
Two things:
1. If you can type on your iPhone as fast as you can on your laptop, then you are a slow typist, so your opinion on speed really doesn't matter. You'll probably never get 60 WPM on your iPhone, and definitely not the 100+ WPM that a "power-typists" have.
2. Predictive text is incredibly useful, but only up to a point. Unfortunately, it sucks if you aren't typing normal english words. For example, try typing this message on your iPhone: "Dear Svrisathana, the composer's name is Shostakovich, and it cannot be spelled Schostakovitch or Shostackovitsch. Will we meet at AX-2470Y or AY-2470X?"
(To say nothing of foreign languages. Try typing this on your phone: "Ni3 hao3, wo3 shi4 fei1chang2 ben4 de5 lao3wai4")
Predictive text is nearly useless for the bulk of programming -- sure, variable and function names, auto-suggest, etc. are great, but since the general keywords in languages are short, and punctuation/non-letter keys are very important, touch-typists have a huge advantage.
Yes, but think of it this way. That 170M is the pool of potential army members -- they are not all going to be fit for the job, and some will be better than others. Prune the tree. Take the best 1.7 million from those and you've got the top 1% killers, the elite. And that's a lot of fucking badass killers.
Let's break this question into 2 parts.
Commercial software without any OSS counterpart.
So far as I know, Reason by Propellerhead has no counterpart.
I am certain that Wenlin, the chinese language learning software, has no OSS counterpart, and unfortunately I never had the money to purchase it when I was studying Chinese...
Commercial software without any GOOD OSS counterpart.
Such as Pro Tools (come on, Ardour -- not really useable for professional work). Sibelius & Finale have no useable counterpart.
That's just a small list from my background (music arranging and multitracking), but it is a field where OSS is really lacking. You simply can't do professional level work in Linux at this moment. But even in my experience, sure there are counterparts for software like iPhoto or iTunes, but none of it is really as well done. I tend to support a decent amount of shareware (I spend maybe $60 a year) because I am unsatisfied with open source options, at least things that I can use on my mac.
I wholeheartedly agree w/r/t complex software. Further, if you look at organic artifacts, even something as simple as a knife, the tool is always changing. The knife gets duller, and you sharpen it (removing molecules), and so it is constantly changing.
I think a lot of people say "uh oh, when are cars going to start getting the BSOD?", but what they don't realize is that, as complicated mechanical entities, cars DO give the equivalent of a BSOD. The word "crash" has a real-life meaning, too, after all :) Tires blow, mufflers fall off, things are changing all the time.
At least software is not subject to the laws of entropy (or is it?!)....
wholeheartedly agree. can't stand it myself, though plenty of people enjoy it I guess. maybe to them, having played fantasy games so long, the rogue/dwarf/wizard whatever is familiar turf like PBJ on wonderbread.
I hate being refuted as much as the next guy, but, nice job.
When it comes to the RIAA, I'm more of a chainsaw kind of guy. But, yeah, other than that I wholeheartedly agree...
I think you're being a little harsh. Regarding the realtors, I think you have a couple good points, but this does not negate the forces of self-interest, described in Freakonomics, that would be likely to motivate a realtor, e.g.
So there is a clear reason why real-estate agents would hurry to sell someone else's house for a little less (because the commission is about the same). It's not a conspiracy, it's a simple force of the market. If I got into being a real-estate agent without ever contacting another real-estate agent, these kinds of tactics would be clear to me if I ever actually thought about how to maximize my profits.
Regarding drug dealers, you say the following:
The good professor takes perfectly valid data and draws wild and crazy conclusions from them, when simple explanations will do. Are there really crack dealers out there risking their lives for pennies? Perhaps there are... but given his sloppy methods in the rest of the book, I find myself taking the entire shebang with a bushel of salt.
First of all, I don't think he drew any conclusions at all regarding the wages of drug dealers. Instead, he presented the actual data from the drug lord's financial notebook. A little division and rounding, e.g.,
Please let me know what kind of crazy conclusions he is drawing from those numbers, or what the sloppiness you see in that method is.Also, if you see any "blatantly wrong answers", let me know. I have the entire book with me, searchable. I have no strong reason for defending it except that I enjoyed the book and think about the conclusions a lot. If there are gaping holes in that, sure, I'd like to know, but so far you haven't really presented a strong case for that.
Lucky day. I found a copy of the relevant chapter at the following URL, which I am not affiliated with:
Freakonomics Ch. 3
It is in nasty RTF format, but readable; you will certainly want to buy the book to get full effect. That said, nothing like a little instant gratification...
If you are vegan, you may be interested in the non-cheesy parts of the following URL at the same random domain:
barcelona gorditasum, joe haydn was well dead for a good half-century before brahms wrote those variations..
Read freakonomics. Really. In addition to being a generally a good book, it does talk about the wages that drug dealers make.
All but the highest rungs of drug "employees" (i won't say dealers, since dealers are usually not in the high echelons) do not make much money, and in fact do live at home with their parents and barely survive on their wages.
If you are truly interested in questions like "what kind of money do drug dealers make", I do highly suggest reading this book. I'm not affiliated with them or anything, I just think it was interesting and really insightful regarding a lot of these social taboo subjects.
I don't think this discussion is really about remixes, which could potentially fall into fair use. Instead, it is about mix tapes, which are compilations of various tracks put together by a single person.
E.g., if I make a CD for a friend and put some Madonna on there, some Radiohead, a Metallica track, I am putting whole copyrighted works (entire songs) on a compilation.
As far as I know this is clearly illegal, though many many people (including myself) think they are ethical, and thus continue to break the law. Usually no legal action is taken, but in this case the RIAA has taken some very serious action against this hip-hop dude.
The labels get MOST of the money from album sales. The artists get a really small percentage, actually.
Here is an interesting article on digital vs. physical music purchases and the artist's cut:
There is nothing either ethical or legal about mixtapes
Well, there may be no legal grounds for them, but we are in totally different moral worlds if you think that mix tapes are unethical. There are plenty of people in both camps -- I'm not saying you're wrong, just that it's not absolutely clear that mixtapes are unethical.
A lot of artists (and consumers) think mixtapes are great from all standpoints.
Any time you pick a particular age as a global standard, it's not necessarily fair, particularly if that age is as high as 15. When I was 17 I dated a 14-year old, who was 5'9 and believe me, as fully developed as at least my 17 year old peers. On the other hand, there are 15 or 16 year olds who are less developed than some 13 year olds.
Puberty age, as well, is decreasing globally, so that people are maturing earlier and earlier.
When I was 14, I was interested in people about my own age. Would it be so unethical for me, at that age, to want to look at pictures or movies of people my own age? What about when I was 12? 10? 8? It's unclear territory, I think. While the idea of pornography including people who are pre-sexual (aka children) is appalling to me, and while I personally wouldn't even be interested in young teens, not everybody shares my mentality.
I don't offer any solutions. I certainly think that pornography involving the coercion of pre-sexual humans should be eliminated, legislated, with high penalties. (On the other hand, when I was 8 I would have happily done porn if given the chance). Anyway, I just think you are a little too straight-laced on this one, thinking there is an absolute gold standard which should be globally legislated. The truth is much more slippery and I like the idea of individuals -- from children to senior citizens -- being able to make up their own minds about the matter, not have their opinions and desires be legislated and governed globally.
You and about 500 other luddites in the world, who want all their appliances to do one function (and to do it well, of course). Everyone else, I think, recognizes the need for a little convergence around here.
And since this phone realizes the beautiful convergence of WiFi tablet, phone, music player, and camera, it is getting a ton of attention. I think this is fully deserved, considering what a wonderful job Apple has done with the design; this is the first solution I've ever seen that looks like it's going to work and work well for a lot of people.
(I am one of those people for whom the design is near-perfect [within the technology constraints of the present -- I'd love a 600GB hard drive in there])
There comes a point when there are too many functions to do to justify one function per device.. in these days of printing, scanning, picture taking, emailing, voice mailing, calling, web browsing, navigating streets, portable music, disc ripping, disc burning, etc. etc., we simply need devices that can do more than one of these functions (the PC is a great example of "getting it right")
also called expresso with an x in portuguese. i get looked at condescendingly when i call it "espresso" in a portuguese cafe...
I don't really think these physical property analogies are valid. If I rent a car and don't return it, the car rental place has just lost a car, and lost money. Same with the Blockbuster physical media.
But what if I rent the blockbuster DVD, rip it, and return the DVD? Nobody loses anything. This is why data/IP is a totally different game from physical property.
Yeah, there should totally be "consolidate" type option, similar to what you do when you mix down a pro tools session, just to get a document that is portable, which you can open but don't have much editing power on.
Regarding large file sizes with images: both ms-word and powerpoint files are documents meant for editing as well as viewing -- so they have to store more information about the data contained inside than, say, pdf.
Let's take the simple case -- you drag in a 640x640 image, resize it to 30x30, then decide it's too small, and resize it to 100x100. Those are significantly different sizes. If word immediately resized the source image so it only had 30x30 pixels (and a smaller file size), when you tried to expand it to 100x100 you wouldn't have any picture quality.
So, sure, you want it to remember what the original image was when you're working in the document. But just because you save the file, quit, and re-open the file, does that mean you're done with resizing? Since word and ppt documents are always work-in-progress type formats, they have to store the original images and objects.
What I'm saying is not just relevant, it's crucial to Nyquist's theorem an digital sampling.
There is, of course, no way to tell what the original waveform was with 15k and 3 samples per crest. But, it is forced into a sine wave because of bandlimiting. It is the harmonics of that 15k fundamental which would make it a square wave, a triangle wave, etc. Those harmonics are filtered. Everything above 22.05kHz is removed. The square wave is turned into a sine wave. The triangle wave is turned into a sine wave.
And that's what we want. Because if you followed the rules of bandlimiting when you encoded the signal, only a 15kHz sine wave would have been encoded (you could have fed it a square wave, but all the harmonics that made it square would have been chopped off). At 15kHz only the sine-wave fundamental frequency would make it in to the encoded signal, and only a 15kHz sine wave will come out.
Does this make more sense to you? I think what you're not understanding is that a wave shape other than sine wave simply means that there is more than one frequency to that signal.
That would be a pretty funny comment if I knew that you were only kidding around...??
Really, if you don't understand how to turn a square wave into a sine wave (by bandlimiting), you should read up on it. You will then understand why 18k in a 22kHz bandlimited encoding system has to be a sine wave.
You could start with Sine Waves. But the really important quote to take away from there is:
This page kind of talks about how square waves are made up of sine waves.I generally avoid seedy places, too. I was kind of assuming my hypothetical $1 coke movie theatre wasn't a complete shithole. In fact, my whole little fantasy revolved around a nice, charming movie theatre that didn't jack up prices (and since I know I can get a 32oz fountain coke from a nice, non-seedy local store for 50 cents, I assume a movie theatre could be nice and have $1 cokes)
:)
The point is more don't artifically inflate prices... even if (gasp) you won't maximize your immediate profit by doing so. You'll certainly maximize your immediate karma by having reasonable prices
I think you might not understand what that actually means. A square wave is a waveform which contains the fundamental and all odd integer harmonics. So if we have a square wave at 10kHz (F=10k), that waveform contains 10kHz (1 * F), 30kHz(3*f), 60kHz(5*f), etc. Each of its component frequencies is, of course a sine tone. Read up on Fourier Transform if you're not following.
So let's take that 10khz square wave again, which has as its frequencies 10kHz, 30kHz, 60kHz, etc. When we perform a band-limit on that square wave, only allowing frequencies below 22kHz, we are left only with the 10kHz sine wave.
I hope that you can extrapolate this to the other waveforms -- sawtooth, triangle, etc., which are comprised of the fundamental and various integer harmonics. The point is that by the time a frequency is high enough to only have a few sample points, such that you could not tell what the original waveform was, the post-decoding bandlimiting will force it to be a sine tone (which is what it inevitably should be, if you followed all the encoding constraints a la Nyquist's theorem).
Right, but what's the first harmonic of a 16kHz tone? 32kHz. That is why the bandlimiting step after decoding is so important, and that's why Nyquist states that in order to reproduce signals up to the Nyquist frequency you must bandlimit!
So what's the difference between a 16kHz sine wave and a 16kHz square or triangle wave that's band-limited to 22kHz?
(Hopefully none! So aliasing isn't that big of a deal, right?)