On Tuesday, January 25, 2011 4:21:08 AM UTC-5, Rune Allnor wrote: ...> You have presented the view that studying an analogy is > equivalent to studying the 'real thing', whatever that might be.As I saw it, Chris observed that analogy can aid understanding. That's not quite the same assertion.> In other words, you claim that precision in language and semantics > does not matter.No /that/ seems to me to be a non sequitor. ... This, from an old (2003) post of mine illustrates the utility of analogy: My experience in related fields helps me to understand DSP concepts and to find fortunate analogies for explaining them. In keeping with tradition, I'll illustrate that with a story. A traffic-light progression system is one in which cars moving at the right speed always face a green light when they reach a controlled intersection, whichever way they travel along the road. (If they go too fast, the light will not yet be green; if too slow, it will have turned red again. Catching another red after restarting is almost inevitable.) My uncle Nathan had, in the 20s, been the only electrical engineer in the Traffic Division of the NYC Police Department (Later, the Traffic Department), responsible for the timing of every traffic light in the city. When I was a kid, his responsibility was only one borough, and he had assistants. He designed the progression system for the Grand Concourse, one of the first in the country. From time to time, it needed to be modified (retimed side streets, speed limit changes, relocated signals). He liked to work on this at our house, spreading his "schematics" out on our big dining-room table (we usually ate in the dinette half of the kitchen). They consisted of vertical station lines representing the signal lights, and bands with negative (northbound) and positive (southbound) slopes. I learned how the system worked by watching him and asking questions. I, my uncle, and his colleagues at large took it for granted that the analysis was complex and esoteric. When a friend from Boston (another EE) visited, we had occasion to drive down the road. He commented that red lights always turned green as we approached. I replied that it was designed to do that, and he opined that it must be lousy for cars going the other way. I said that it was the same for them and he asked how it was possible. Lacking graphics but sharing a common fund of knowledge, I suggested he think of the lights as a standing wave, with traveling waves going north and south. He was briefly silent, then made it clear that he completely understood by deducing that cars must in consequence move in bunches. Traffic lights are binary, so the traveling waves are square. (Later, when remote control of timing became possible, I helped my uncle retime the lights for rush hour so that the bunches were longer in the heavy direction at the expense of shorter ones the other way. A good analogy can be a powerful aid to understanding, creating new possibilities.) Jerry
Re: Appendix A: Types of Fourier Transforms
Started by ●January 25, 2011
Reply by ●January 26, 20112011-01-26
On Jan 25, 10:50�pm, Jerry Avins <j...@ieee.org> wrote:> On Tuesday, January 25, 2011 4:21:08 AM UTC-5, Rune Allnor wrote: > > � ... > > > You have presented the view that studying an analogy is > > equivalent to studying the 'real thing', whatever that might be. > > As I saw it, Chris observed that analogy can aid understanding. That's not quite the �same assertion.I'll see if I can dig out the exact statements next weekend, but Chris coems across as stating that it does not matter if one sticks with the analogy.> > In other words, you claim that precision in language and semantics > > does not matter. > > No /that/ seems to me to be a non sequitor. > > � ... > > This, from an old (2003) post of mine illustrates the utility of analogy:Analogies are dangerous at the best of times. As arsenic, which apparently once upon a time was used to heal certain VDs, they are toxic in too large doses. 'Too large', of course, being disturbingly small. And of course, as arsenic the use of analogies is ridiculous when simple, precise and accurate alternatives exist. As is the case with the FT. Rune
Reply by ●January 26, 20112011-01-26
Rune Allnor wrote:> On Jan 25, 10:50 pm, Jerry Avins<j...@ieee.org> wrote: >> On Tuesday, January 25, 2011 4:21:08 AM UTC-5, Rune Allnor wrote: >> >> ... >> >>> You have presented the view that studying an analogy is >>> equivalent to studying the 'real thing', whatever that might be. >> >> As I saw it, Chris observed that analogy can aid understanding. That's not quite the same assertion. > > I'll see if I can dig out the exact statements next weekend, > but Chris coems across as stating that it does not matter > if one sticks with the analogy. > >>> In other words, you claim that precision in language and semantics >>> does not matter. >> >> No /that/ seems to me to be a non sequitor. >> >> ... >> >> This, from an old (2003) post of mine illustrates the utility of analogy: > > Analogies are dangerous at the best of times. As arsenic, which > apparently once upon a time was used to heal certain VDs, they > are toxic in too large doses. 'Too large', of course, being > disturbingly small.I come late to discussion. BUT Is oxygen a poison or vital to life? It blinds some infants. And what about water? Vital or killer? Analogies have limits.> > And of course, as arsenic the use of analogies is ridiculous when > simple, precise and accurate alternatives exist. As is the case > with the FT. > > Rune
Reply by ●January 26, 20112011-01-26
On Jan 26, 5:23�am, Rune Allnor <all...@tele.ntnu.no> wrote:> On Jan 25, 10:50�pm, Jerry Avins <j...@ieee.org> wrote: > > > On Tuesday, January 25, 2011 4:21:08 AM UTC-5, Rune Allnor wrote: > > > � ... > > > > You have presented the view that studying an analogy is > > > equivalent to studying the 'real thing', whatever that might be. > > > As I saw it, Chris observed that analogy can aid understanding. That's not quite the �same assertion. > > I'll see if I can dig out the exact statements next weekend, > but Chris coems across as stating that it does not matter > if one sticks with the analogy.Found it: http://groups.google.com/group/comp.dsp/msg/991e51c116509d70 "So while I agree that the DFT may be presented as an abstract mapping that has no 'assumptions' outside its own scope, I also suggest that when discussing signal processing as opposed to abstract math, those assumptions must be included." Plain wrong. The maths exists and works independently of any 'real world'. 2+2 = 4 no matter if we are talking about dollars in economy, sheep in agriculture, voltages in EE, or abstractly. "Both [views], I think, are valid and lead to similar conclusions." Of course not. Either the maths works independently, in its own right, or it depends on the quantity being discussed as listed above. One or the other. "However, what I do disagree with is the idea that the 'abstract' approach is somehow devoid of context." Here we go: "The maths depends on context." Somebody who actually means this, is deserves the label 'mediocre'. The last sentence should come as no surprise, then: "I agree with everybody but I do what I find easiest." Rune
Reply by ●January 26, 20112011-01-26
On Jan 26, 7:15�am, Rune Allnor <all...@tele.ntnu.no> wrote:> On Jan 26, 5:23�am, Rune Allnor <all...@tele.ntnu.no> wrote: > > > On Jan 25, 10:50�pm, Jerry Avins <j...@ieee.org> wrote: > > > > On Tuesday, January 25, 2011 4:21:08 AM UTC-5, Rune Allnor wrote: > > > > � ... > > > > > You have presented the view that studying an analogy is > > > > equivalent to studying the 'real thing', whatever that might be. > > > > As I saw it, Chris observed that analogy can aid understanding. That's not quite the �same assertion. > > > I'll see if I can dig out the exact statements next weekend, > > but Chris coems across as stating that it does not matter > > if one sticks with the analogy. > > Found it: > > http://groups.google.com/group/comp.dsp/msg/991e51c116509d70> "However, what I do disagree with is the idea that the 'abstract' > approach is somehow devoid of context." > > Here we go: "The maths depends on context."Just to be clear: The latter sentence is my rephrasing of Chris' view. Chris writes nowhere in his post the sentence "The maths depends on context." Rune
Reply by ●January 26, 20112011-01-26
On 26/01/2011 07:15, Rune Allnor wrote:> > Plain wrong. The maths exists and works independently > of any 'real world'. 2+2 = 4 no matter if we are talking about > dollars in economy, sheep in agriculture, voltages in EE, > or abstractly. >There are three people interviewing for a job in the financial department of a company - a mathematician, an engineer, and an accountant. All three are equally qualified, so the interviewer decides to ask them all one more question. First, he asks the mathematician "What is 2 plus 2?" The mathematician answers "4, assuming standard operations on the set of integers, but it's quite hard to prove". The engineer answers "4, but we should really consider the margin of error in the original values". The accountant gets up from his chair, walks across the room and closes the door, then walks up to the interviewer and says quietly "how much would you like 2 plus 2 to be?"
Reply by ●January 26, 20112011-01-26
Rune Allnor wrote:> > > > Here we go: "The maths depends on context." > > Just to be clear: The latter sentence is my rephrasing of Chris' > view. Chris writes nowhere in his post the sentence "The maths > depends on context."Perhaps you should make your own arguments rather than spending your time distorting what others say. The math doesn't depend on context. But analysis depends on context The "math" works when you are dealing with a linear, band-limited, perfectly precise reality. But reality is never linear or bandlimited or measured with absolute precision.
Reply by ●January 26, 20112011-01-26
On Jan 25, 10:48�pm, Richard Owlett <rowl...@pcnetinc.com> asked:> > And what about water? > Vital or killer? >Visit http://www.dhmo.org/ for a complete update on this vital question
Reply by ●January 26, 20112011-01-26
On Jan 26, 2:08�pm, jim <"sjedgingN0Sp"@m@mwt,net> wrote:> Rune Allnor wrote: > > > > Here we go: "The maths depends on context." > > > Just to be clear: The latter sentence is my rephrasing of Chris' > > view. Chris writes nowhere in his post the sentence "The maths > > depends on context." > > Perhaps you should make your own arguments > rather than spending your time distorting > what others say. > > The math doesn't depend on context.My point exactly.> But analysis depends on contextSure. The point is that almost everybody seem to take the view that one needs to incorporate the analysis into the DFT before one can do the computations. I never got any answer to this question: http://groups.google.com/group/comp.dsp/msg/c89333b4378d72b0> The "math" works when you are dealing with > a linear, band-limited, perfectly precise reality.Nope. The maths works regardless of any reality.> But reality is never linear or bandlimited > or measured with absolute precision.That's a question to consider during data analyis, not while crunching numbers. Rune
Reply by ●January 26, 20112011-01-26
On Jan 26, 7:34�am, Rune Allnor <all...@tele.ntnu.no> wrote:> Sure. The point is that almost everybody seem > to take the view that one needs to incorporate > the analysis into the DFT before one can do the > computations. I never got any answer to this > question: > > http://groups.google.com/group/comp.dsp/msg/c89333b4378d72b0 >From the question above: ___________________________________ That would imply that the validity of the answer from the DFT is not dependant on the values outside the length N, which is clearly not the case. ________________________________ OK Rune I'll take a crack at putting a reasonable interpretation to this quote above. Interpretation 1: The DFT can give different answers depending upon what numbers are adjacent to the input data but not actually entered into the DFT equation. Rune's conclusion: steveu is a complete moron - now Rune can berate said individual for being a moron. my conclusion: I do not think steveu is a complete moron so lets look for a reasonable alternative to what is meant by the statement above. Interpetation 2: The validity of the answer from the DFT is to be looked at in the context of the whole set of data. In other words, when doing data analysis the importance (substitue importance for validity) of a single DFT should be weighed in accordance to other data around the time the DFT was taken. Rune's conclusion to interpretation 2: No, that is not what he said and steveu is a complete moron. I like puttiing people into convenient boxes, especially ones that make me a superior intellectual specimen. (ok I'll admit to making this up - but drawing conclusions about others on usenet is easy and fun) my conclusion to interpretation 2: This may not be what he meant, but given that I don't think he is a complete moron , this is more likely what he meant and it seems like a reasonable interpretation of what he meant. This is a more likely interpretation than interpretation 1 because steveu is probably not really the moron that would be required for interpretation 1.
