Anyway, they aren't the problem, it's the dioxins that are
Then why bring them up? Or are you just trying to capitalize on the visceral emotional reaction that people have been conditioned to have when they hear the words "Agent Orange". Fnord.
RoundUp is similar in usage though
So what? That doesn't mean that it's harmful in the same way.
Uh, just where exactly do you think the uncertainty principle came from? Come to think of it, do you even know what the uncertainty principle is?
Irrelevant to an explanation of "everything"? Is the multi-universe theory irrelevant?
No, the multi-universe theory isn't irrelevant. But then, it has been developed in order to explain observable effects in our universe. Otherwise there would be no point in having the theory.
Do you actually believe that this "methodology" is capable of producing an all-encompassing theory of everything?
Sure. Why not? The basic algorithm is:
Observe something no previously predicted or explained
Develop a hypothesis to explain the observations
Make predictions based on the hypothesis
Develop tests to determine whether the hypothesis is correct or incorrect
If the hypothesis is incorrect revise it based on the new observations
Rinse and repeat
Please tell me how this algorithm, given enough time, will not succeed in explaining "everything"?
Forgive me if that sounds a little arrogant.
You are forgiven.
Also, what is it that is being explained? Reality, or our perception of reality?
What is reality but what you perceive? If something is completely imperceptible (i.e. makes no observable change in the universe whatsoever) then whether it exists or not is irrelevant - it makes no difference to my life or yours. If it can be perceived, it can be observed. If it can be observed it is amenable to study via the scientific method.
There's another advantage that they don't mention. Linux plays nicely with the BSDs, Solaris, OS X, and most other operating systems, so it really is easier to have a diverse environment - not just diverse distributions, but diverse operating systems and architectures.
True. That's a function of having open standards at the interface level. Standards that Windows is just as welcome to make use of - they just choose not to.
The idea of homogeneous interface specifications and heterogeneous implementations is a good one, and does provide a level of diversity that can limit the impact of viruses preying on implementation-level vulnerabilities. However, we can't let that slightly higher level of security breed complacence. It is entirely possible that interface-level vulnerabilities could bring every compliant implementation down (which is a good argument for more careful design and auditing of interface standards).
It would perhaps be more correct to say that "There's lots of stuff that science hasn't yet explained". The term science doesn't refer to some fixed body of knowledge. It refers to a methodology for finding and refining explanations.
Most S-V costings don't account for the overhead and infrastructure cost, while every Shuttle one does.
I'd be willing to debate both of those assertions. However, I do agree that it depends a lot on what you include in your launch cost roll-up (and what you define as "overhead" and "infrastructure cost")
The reality is that the marginal cost for a Shuttle flight is around 150 million a flight, but the overhead kills it when spread across so few flights.
The reality is that the shuttle cannot support a higher flight rate, so the marginal cost is somewhat meaningless (and is dominated by the fixed costs anyway).
The passengers on an X-Prize vehicle are no more going somewhere than are the riders of a roller coaster.
Cute analogy. But you are conveniently missing the point. The X-Prize passengers will be going into space, a realm that has, until now, been restricted to hand-picked astronauts, self-made multi-millionaires, and congressmen on junkets. So what if it's only sub-orbital for now. That at least puts them on a par with the early Mercury flights. The Wright Flyer flew only a few hundred feet to begin with. That doesn't detract from the fact that it flew.
Like every aviation prize before it, the X-prize is encouraging vehicles designed specifically to win the prize.
And the prize is specifically designed to encourage vehicles that support fast-turnaround with minimal infrastructure. Those two features are essentially what the launch vehicle community is referring to when they talk about "airline-like" operations (and relative to the way launch vehicles are currently operated they do represent something much more like the way an airline operates). Ok, so you won't be using an X-Prize competitor like an actual modern airliner. But as you say "It took the airlines and manufacturers decades to achieve those levels." They did it by trying lots of different stuff, discarding what failed, and keeping what worked. The beauty of the X-Prize is that we're finally getting away from NASA's stale "one true way" of doing manned launch, and experimenting with a variety of approaches. All of these approaches must, as a result of the competition rules, give at least some consideration to reusability and operability. Some will work. Some will fail. We'll learn from them all, and probably learn a lot more than we would from the endless paper studies that characterize NASA's attempts at manned launch. The current crop of X-Prize contenders may not be the equivalent of a space-going DC-3, but they sow the seeds from which such a craft can eventually emerge.
The Shuttle has 60,000lbs of cargo capacity and up to 9 passengers/crew
The shuttle's 60,000 lb cargo capacity is wasteful and useless. It costs more per pound (even accounting for inflation) to launch on the shuttle than it did to launch on the Saturn V. It'd be fine if the shuttle provided an economical way to launch bulk cargo, but it doesn't. Better to stick with unmanned expendables for that kind of stuff - at least for the time being. As for the 9 passengers/crew, they cost so much per person to launch that only a small elite are permitted to fly. The 4 passengers on an X-Prize vehicle may only be going suborbital (for now), but at least they're going.
The key to reducing costs isn't reducing vehicle costs as many believe, but in flying the hell out of the vehicle and spreading the costs across many vehicles and flights.
True. But that's part of the point of the X-Prize. The shuttle design simply cannot support a flight rate sufficient to make its costs reasonable. Plus it requires a standing army of several thousand just to operate it. The shuttle is not capable of operating in an airline mode. The X-Prize is encouraging designs that are capable of rapid turn-around (and thus high flight rate), and require minimal infrastructure. The X-Prize designs will (hopefully) be capable of airline-like operations.
Consider the long step between the Wright Flyer and the Ford Tri-Motor or the DC-3. That's how far the X-prize vehicles are from useful and cheap space transports.
The first flight of the Wright Flyer involved a mere 12 seconds of flying time (the third and longest flight of the day attained a whopping 59 seconds). Only 10 years later the airplane was a major player in the Great War. Ok, the world had to wait another ~20 years for the DC-3. But commercial aviation was already well-established before the DC-3 came along. Useful and cheap are relative terms. The X-Prize vehicles may be closer to both of them than you think.
That's why old 5V TTL logic eats a lot more power than modern 1.5V - 1.0V transitors.
No doubt. But for a fixed voltage (and the voltages seem to be staying fairly fixed these days since there hasn't been a fundamental shift like the one from TTL to CMOS), greater frequency means greater power consumption. It'd be interesting to see the balance between drop in power consumption from reduced capacitance and increase in power consumption due to higher clock frequencies - wonder if there's a net drop or increase in power consumed...
As the processor power doubles, the power consumption also rises.
Correct me if I'm wrong, but: Malone, the author, is exaggerating by implying that the size of a transistor is remainaining constant while the number of transistors doubles. As I understand it, the smaller the processor, the less power it requires. Is this right?
You are (mostly) correct about size versus power. But Malone didn't say processor size, he said processor power, which for most people is a function of clock speed. And he is correct that greater clock speed leads to greater power consumption.
I agree that these "music licenses" are not like the GPL. My point was that they are both voluntarily entered agreements in which one agrees to abide by certain rules in exchange for receiving certain value.
Regardless of the terminology used in the Apple store (which I doubt is any different than you would see in an online software store) the terms of the purchase are clear, and freely entered into. If you do not agree with the terms, don't spend the money. If you purchase music on a CD you are bound only by copyright law, and thus fair use is legal. If you purchase music from the iTunes store you are further bound by the purchase agreement, and must abide by it. If the RIAA started making people sign purchase agreements for buying CDs that would be equally valid. They won't do that though, because CD sales would drop radically. The problem is not Apple's terms, it's that people buy the music anyway. If the terms really suck, then don't buy. If enough people think the terms suck then Apple will make no money, and will have to modify the terms (or get out of the online music business).
Amen to that. Circumventing the DRM on downloaded music is like modifying GPL software, and then repudiating your agreement to the terms of the GPL and saying you can do what you like with any derived products.
People like to say that the GPL means "free as in speech". And just like preserving free speech means allowing speech we disagree with, a strong GPL requires strong respect for licensing terms - even if we disagree with those terms.
I never said that research should be classified. I was merely pointing out that for some folks there is "room for argument" regarding the public availability of taxpayer-funded research.
With regard to the "why" of classified programs, the answer is typically less to do with the fact that "the people" would be against the program, and more to do with the practice of "security through obscurity". The government wants to hide its true combat capabilities so that potential adversaries will not know what to expect. By a similar token, they want to hide their true intelligence gathering capabilities so that potential adversaries will not know whether or not they have been observed. That is the rationale. I'll make no comment regarding the wisdom and truth of that rationale.
If that 1.5 - 2.0 Indian programmers can produce the same quality and quantity as the American programmer for less money, that's better, both for the company and for the consumer (and for the Indian programmers that gain employment).
something tells me that the banks are not about to reprice the debt that the middle class has...
Maybe not. But if the price of goods drops (e.g. if they're manufactured or developed overseas) then the middle class will have more money left over to service their debt. Or I suppose they could just stick with their current scramble to accumulate material wealth, and buy more useless crap instead of paying off their debt.
If the research is funded in whole or in part by the taxpayers, then ALL research results must be published and made freely available to ALL taxpayers. I can see no room for argument there.
Kind of like the "race to bottom" caused by FOSS? Once your competitors start giving away their software for free, you're forced to do it to compete on price (and then attempt to subsist on "service and support contracts"). Right?
- Americans have to pay for college, we don't get it for "free" like other countries provide.
And instead people in those other countries pay higher taxes. "There ain't no such thing as a free lunch". State universities are not that expensive, and there is plenty of financial aid available. Worst case, you can join the military and get them to pay for it.
- We get two weeks of paid vacation per year, unlike other countries.
Cry me a river. I'm sure you could negotiate a few extra weeks of paid vacation if you were really worth it. Or maybe you could take a slight paycut in exchange for a few extra weeks of non-productivity. Again, TANSTAAFL.
- Medical care? Same thing.....
Trust me, you do not want to live in a country with socialized medical care. Been there, done that. Not fun at all. Medical care costs a crapload in the US because people want a much higher standard of care here - more drugs, more tech, no waiting lists.
The only reason that the US dollar has any value overseas is that it can eventually be redeemed in the US. Someone has to eventually spend it here (although not necessarily the someone that you initially gave the dollars to). Otherwise it's just a piece of paper. The situation is the same with international banks - sure they convert currencies, but they do it in accordance with an exchange rate that is rooted in what a given currency will buy in its country of origin.
Actually, the tagline "the first viable Microsoft Windows alternative" originates with Sun, not with Walmart. Check out Sun's JDS webpage - the very first line includes the "first viable Microsoft Windows alternative" phrase.
These so-called improvements gloss over the continuing problems that plague spreadsheet users:
Spreadsheet models encourage the use of "spaghetti" logic, where cells point to cells that point to cells, and can grow into random networks of calculation logic;
Yes, yes. And programming languages with only gotos are inherently evil. But with a discipline on the part of the user it is possible to build maintainable systems.
They permit lots of easy off-by-one errors;
Very true, and I have seen it happen any number of times. It's always a good idea to build in validation checks that provide some feedback on how "sensible" the numbers coming out are.
They generally are difficult to verify/audit;
Well, yes and no. The "trace dependents" and "trace precedents" functionality in Excel can be quite handy for that kind of thing. I've also seen a regression testing tool (developed in VBA) that checks one spreadsheet against another to ensure that the page-to-page links have remained the same (i.e. the interface is consistent) and flags any changes. Detailed validation of the numbers produced by each sheet still had to be done by hand, but it was setp in the right direction.
They do not provide good tools for managing data either in terms of consolidation or searching for specific detail;
No argument there.
Perhaps most importantly, despite their convenience, spreadsheets are not a robust repository for information.
That depends to a certain extent on how you are using them.
While spreadsheets get their most heavy use in the financial world, they have gained a lot of popularity in the engineering world, particularly for doing the computerized equivalent of "back-of-the-envelope" calculations. While I was initially pretty skeptical of this concept (being a diehard Matlab fan), I have to admit that for quick exploration of various design options using simplfied models things like Excel are king. Most of the value comes from the same things that are cited as problems for the financial world: that its easy to tweak your assumptions until you get the results you want, and do it with immediate feedback. Bad for the financial world perhaps, but in the early requirements/design phase it becomes an incredibly valuable tool for rapid trade-space exploration. I guess the difference between engineering and finance is that in engineering the resulting numbers are validated in later design phases that use more rigorous tools - that helps to filter bad assumptions before they produce disasters. This kind of spreadsheet-driven conceptual design is very popular in the aerospace industry: JPL, NASA-Goddard, NASA-JSC, The Aerospace Corporation, ESA, NGST (ne TRW), Boeing, Ball Aerospace, Spectrum Astro, and AFRL, to name just a few, all use some variant of it in one way or another (often in teams that use inter-linked workbooks).
...HW designers use VHDL because their graphical tools are too low level...
Good point. However, I'd argue that the graphical tools really do make sense at the low level that you're talking about, because at that point you are manipulating physical artifacts with a real geometry. But software isn't physical artifacts. At the functional level, which is where software operates, VHDL and the like are the weapon of choice.
If that involves drawing lots of highly detailed pictures within pictures at the "design" level, then fine, but it wouldn't make anything less complex or less bug-ridden.
In fact, it's likely to just make things more confusing. There's a reason that mathematicians don't do geometric proofs so much anymore - symbolic manipulation is more clear, more general, and more compact. It's the same reason that hardware designers use things like VHDL now.
Many people seem to think that a "graphical language" makes things easier for lay-people to understand. And that's true at the very highest levels of abstraction. But when you get down to the details a graphical language must have the same expressiveness as an equivalent symbolic language. That means that it will almost inevitably have the same level of complexity as the symbolic language, and be equally impenetrable to lay-people. One only has to look at the newest versions of UML to see this effect in action.
Bottom line: graphics are great at a high level of abstraction, and as documentation to aid understanding of a symblic expression, but for implementing complex systems symbolic languages are much better.
Slashdot Mirror
Then why bring them up? Or are you just trying to capitalize on the visceral emotional reaction that people have been conditioned to have when they hear the words "Agent Orange". Fnord.
RoundUp is similar in usage though
So what? That doesn't mean that it's harmful in the same way.
That's the first time I've ever seen Bic Runga and the Red Hot Chilli Peppers lumped into the same genre... :-)
Uh, just where exactly do you think the uncertainty principle came from? Come to think of it, do you even know what the uncertainty principle is?
Irrelevant to an explanation of "everything"? Is the multi-universe theory irrelevant?
No, the multi-universe theory isn't irrelevant. But then, it has been developed in order to explain observable effects in our universe. Otherwise there would be no point in having the theory.
Sure. Why not? The basic algorithm is:
- Observe something no previously predicted or explained
- Develop a hypothesis to explain the observations
- Make predictions based on the hypothesis
- Develop tests to determine whether the hypothesis is correct or incorrect
- If the hypothesis is incorrect revise it based on the new observations
- Rinse and repeat
Please tell me how this algorithm, given enough time, will not succeed in explaining "everything"?Forgive me if that sounds a little arrogant.
You are forgiven.
Also, what is it that is being explained? Reality, or our perception of reality?
What is reality but what you perceive? If something is completely imperceptible (i.e. makes no observable change in the universe whatsoever) then whether it exists or not is irrelevant - it makes no difference to my life or yours. If it can be perceived, it can be observed. If it can be observed it is amenable to study via the scientific method.
True. That's a function of having open standards at the interface level. Standards that Windows is just as welcome to make use of - they just choose not to.
The idea of homogeneous interface specifications and heterogeneous implementations is a good one, and does provide a level of diversity that can limit the impact of viruses preying on implementation-level vulnerabilities. However, we can't let that slightly higher level of security breed complacence. It is entirely possible that interface-level vulnerabilities could bring every compliant implementation down (which is a good argument for more careful design and auditing of interface standards).
It would perhaps be more correct to say that "There's lots of stuff that science hasn't yet explained". The term science doesn't refer to some fixed body of knowledge. It refers to a methodology for finding and refining explanations.
I'd be willing to debate both of those assertions. However, I do agree that it depends a lot on what you include in your launch cost roll-up (and what you define as "overhead" and "infrastructure cost")
The reality is that the marginal cost for a Shuttle flight is around 150 million a flight, but the overhead kills it when spread across so few flights.
The reality is that the shuttle cannot support a higher flight rate, so the marginal cost is somewhat meaningless (and is dominated by the fixed costs anyway).
The passengers on an X-Prize vehicle are no more going somewhere than are the riders of a roller coaster.
Cute analogy. But you are conveniently missing the point. The X-Prize passengers will be going into space, a realm that has, until now, been restricted to hand-picked astronauts, self-made multi-millionaires, and congressmen on junkets. So what if it's only sub-orbital for now. That at least puts them on a par with the early Mercury flights. The Wright Flyer flew only a few hundred feet to begin with. That doesn't detract from the fact that it flew.
Like every aviation prize before it, the X-prize is encouraging vehicles designed specifically to win the prize.
And the prize is specifically designed to encourage vehicles that support fast-turnaround with minimal infrastructure. Those two features are essentially what the launch vehicle community is referring to when they talk about "airline-like" operations (and relative to the way launch vehicles are currently operated they do represent something much more like the way an airline operates). Ok, so you won't be using an X-Prize competitor like an actual modern airliner. But as you say "It took the airlines and manufacturers decades to achieve those levels." They did it by trying lots of different stuff, discarding what failed, and keeping what worked. The beauty of the X-Prize is that we're finally getting away from NASA's stale "one true way" of doing manned launch, and experimenting with a variety of approaches. All of these approaches must, as a result of the competition rules, give at least some consideration to reusability and operability. Some will work. Some will fail. We'll learn from them all, and probably learn a lot more than we would from the endless paper studies that characterize NASA's attempts at manned launch. The current crop of X-Prize contenders may not be the equivalent of a space-going DC-3, but they sow the seeds from which such a craft can eventually emerge.
The shuttle's 60,000 lb cargo capacity is wasteful and useless. It costs more per pound (even accounting for inflation) to launch on the shuttle than it did to launch on the Saturn V. It'd be fine if the shuttle provided an economical way to launch bulk cargo, but it doesn't. Better to stick with unmanned expendables for that kind of stuff - at least for the time being. As for the 9 passengers/crew, they cost so much per person to launch that only a small elite are permitted to fly. The 4 passengers on an X-Prize vehicle may only be going suborbital (for now), but at least they're going.
The key to reducing costs isn't reducing vehicle costs as many believe, but in flying the hell out of the vehicle and spreading the costs across many vehicles and flights.
True. But that's part of the point of the X-Prize. The shuttle design simply cannot support a flight rate sufficient to make its costs reasonable. Plus it requires a standing army of several thousand just to operate it. The shuttle is not capable of operating in an airline mode. The X-Prize is encouraging designs that are capable of rapid turn-around (and thus high flight rate), and require minimal infrastructure. The X-Prize designs will (hopefully) be capable of airline-like operations.
Consider the long step between the Wright Flyer and the Ford Tri-Motor or the DC-3. That's how far the X-prize vehicles are from useful and cheap space transports.
The first flight of the Wright Flyer involved a mere 12 seconds of flying time (the third and longest flight of the day attained a whopping 59 seconds). Only 10 years later the airplane was a major player in the Great War. Ok, the world had to wait another ~20 years for the DC-3. But commercial aviation was already well-established before the DC-3 came along. Useful and cheap are relative terms. The X-Prize vehicles may be closer to both of them than you think.
Well it was originally written for a young adult audience...
I stand (or sit) corrected. I obviously haven't been paying enough attention.
No doubt. But for a fixed voltage (and the voltages seem to be staying fairly fixed these days since there hasn't been a fundamental shift like the one from TTL to CMOS), greater frequency means greater power consumption. It'd be interesting to see the balance between drop in power consumption from reduced capacitance and increase in power consumption due to higher clock frequencies - wonder if there's a net drop or increase in power consumed...
You are (mostly) correct about size versus power. But Malone didn't say processor size, he said processor power, which for most people is a function of clock speed. And he is correct that greater clock speed leads to greater power consumption.
Regardless of the terminology used in the Apple store (which I doubt is any different than you would see in an online software store) the terms of the purchase are clear, and freely entered into. If you do not agree with the terms, don't spend the money. If you purchase music on a CD you are bound only by copyright law, and thus fair use is legal. If you purchase music from the iTunes store you are further bound by the purchase agreement, and must abide by it. If the RIAA started making people sign purchase agreements for buying CDs that would be equally valid. They won't do that though, because CD sales would drop radically. The problem is not Apple's terms, it's that people buy the music anyway. If the terms really suck, then don't buy. If enough people think the terms suck then Apple will make no money, and will have to modify the terms (or get out of the online music business).
People like to say that the GPL means "free as in speech". And just like preserving free speech means allowing speech we disagree with, a strong GPL requires strong respect for licensing terms - even if we disagree with those terms.
With regard to the "why" of classified programs, the answer is typically less to do with the fact that "the people" would be against the program, and more to do with the practice of "security through obscurity". The government wants to hide its true combat capabilities so that potential adversaries will not know what to expect. By a similar token, they want to hide their true intelligence gathering capabilities so that potential adversaries will not know whether or not they have been observed. That is the rationale. I'll make no comment regarding the wisdom and truth of that rationale.
something tells me that the banks are not about to reprice the debt that the middle class has...
Maybe not. But if the price of goods drops (e.g. if they're manufactured or developed overseas) then the middle class will have more money left over to service their debt. Or I suppose they could just stick with their current scramble to accumulate material wealth, and buy more useless crap instead of paying off their debt.
Unless the research is classified...
Kind of like the "race to bottom" caused by FOSS? Once your competitors start giving away their software for free, you're forced to do it to compete on price (and then attempt to subsist on "service and support contracts"). Right?
And instead people in those other countries pay higher taxes. "There ain't no such thing as a free lunch". State universities are not that expensive, and there is plenty of financial aid available. Worst case, you can join the military and get them to pay for it.
- We get two weeks of paid vacation per year, unlike other countries.
Cry me a river. I'm sure you could negotiate a few extra weeks of paid vacation if you were really worth it. Or maybe you could take a slight paycut in exchange for a few extra weeks of non-productivity. Again, TANSTAAFL.
- Medical care? Same thing.....
Trust me, you do not want to live in a country with socialized medical care. Been there, done that. Not fun at all. Medical care costs a crapload in the US because people want a much higher standard of care here - more drugs, more tech, no waiting lists.
The only reason that the US dollar has any value overseas is that it can eventually be redeemed in the US. Someone has to eventually spend it here (although not necessarily the someone that you initially gave the dollars to). Otherwise it's just a piece of paper. The situation is the same with international banks - sure they convert currencies, but they do it in accordance with an exchange rate that is rooted in what a given currency will buy in its country of origin.
Actually, the tagline "the first viable Microsoft Windows alternative" originates with Sun, not with Walmart. Check out Sun's JDS webpage - the very first line includes the "first viable Microsoft Windows alternative" phrase.
Yes, yes. And programming languages with only gotos are inherently evil. But with a discipline on the part of the user it is possible to build maintainable systems.
Very true, and I have seen it happen any number of times. It's always a good idea to build in validation checks that provide some feedback on how "sensible" the numbers coming out are.
Well, yes and no. The "trace dependents" and "trace precedents" functionality in Excel can be quite handy for that kind of thing. I've also seen a regression testing tool (developed in VBA) that checks one spreadsheet against another to ensure that the page-to-page links have remained the same (i.e. the interface is consistent) and flags any changes. Detailed validation of the numbers produced by each sheet still had to be done by hand, but it was setp in the right direction.
No argument there.
That depends to a certain extent on how you are using them.
While spreadsheets get their most heavy use in the financial world, they have gained a lot of popularity in the engineering world, particularly for doing the computerized equivalent of "back-of-the-envelope" calculations. While I was initially pretty skeptical of this concept (being a diehard Matlab fan), I have to admit that for quick exploration of various design options using simplfied models things like Excel are king. Most of the value comes from the same things that are cited as problems for the financial world: that its easy to tweak your assumptions until you get the results you want, and do it with immediate feedback. Bad for the financial world perhaps, but in the early requirements/design phase it becomes an incredibly valuable tool for rapid trade-space exploration. I guess the difference between engineering and finance is that in engineering the resulting numbers are validated in later design phases that use more rigorous tools - that helps to filter bad assumptions before they produce disasters. This kind of spreadsheet-driven conceptual design is very popular in the aerospace industry: JPL, NASA-Goddard, NASA-JSC, The Aerospace Corporation, ESA, NGST (ne TRW), Boeing, Ball Aerospace, Spectrum Astro, and AFRL, to name just a few, all use some variant of it in one way or another (often in teams that use inter-linked workbooks).
Good point. However, I'd argue that the graphical tools really do make sense at the low level that you're talking about, because at that point you are manipulating physical artifacts with a real geometry. But software isn't physical artifacts. At the functional level, which is where software operates, VHDL and the like are the weapon of choice.
In fact, it's likely to just make things more confusing. There's a reason that mathematicians don't do geometric proofs so much anymore - symbolic manipulation is more clear, more general, and more compact. It's the same reason that hardware designers use things like VHDL now.
Many people seem to think that a "graphical language" makes things easier for lay-people to understand. And that's true at the very highest levels of abstraction. But when you get down to the details a graphical language must have the same expressiveness as an equivalent symbolic language. That means that it will almost inevitably have the same level of complexity as the symbolic language, and be equally impenetrable to lay-people. One only has to look at the newest versions of UML to see this effect in action.
Bottom line: graphics are great at a high level of abstraction, and as documentation to aid understanding of a symblic expression, but for implementing complex systems symbolic languages are much better.
Ha! You've obviously never been involved in the government acquisitions process. More fools, and a lot more money.