I wouldn't want to try to teach "programming" to any child who hadn't had Algebra I [and preferably Algebra II].
Eh, I'd been programming in BASIC for three years, and Logo and Pascal for about a year when I took Algebra I. I don't see the dependency for really basic things.
It's about knowing what is going on inside computers. [...] Definitely stay away from languages like BASIC.
I don't get endorsing C and telling people to stay away from BASIC when its "about knowing what is going on inside computers". Unstructured, old-school BASIC is, in many ways, a lot closer conceptually to what goes on inside computers than any structured programming language, C certainly included.
Instead of teaching them how to write a dummy program in a particular language, it is by far better idea to lay the foundation work by teaching them how to design and formulate a solution to a particular problem in a logical, concise, and efficient practice.
You can teach all that along with any programming language, probably more easily than you can without one; that's basically the approach taken by How to Design Programs (which uses a series of Scheme variants, which I think is a good way, though the general outline would be usable with standard Scheme or even a completely different language.) HtDP may be a little steep for the age group (but perhaps not if they are gifted), but you could probably follow HtDP as a "course outline" and make things friendlier, and the Dr. Scheme environment that supports its is pretty nice for educational use.
can assure you 11 year olds, let alone 14, will look at you funny when you tell them to move a turtle (that's what I remember it being called) around the screen drawing senseless shapes.
Sure, moving a turtle around drawing "senseless shapes" is probably not going to be very interesting to 11-14 year olds, especially those on the older end of that range, but then even the versions of Logo I used at 11 could be used to a lot more than that, and modern, agent-based versions of Logo are designed to do quite a lot more, easily. Take a look at NetLogo or StarLogo TNG.
That's exactly what my intro CS course did. We worked with a functional language, an object-oriented language, a low-level assembly language, a logic programming language, and also theoretical constructs like finite automata.
There are several languages that can do functional, OO, and structured imperative programming naturally using subsets of the same language, without being forced (Scheme, for instance; Oz does all that plus logic programming, though I personally think that Scheme's Lispy syntax would be easier for beginners.) I'm not sure throwing a bunch of different languages at students in a first course is a good idea, both because there is cognitive overhead for each language and because it reinforces the idea that choice of programming paradigm is directed by choice of language.
Functional, OO, or (structured) imperative methods are all valid; I think How to Design Programs is a good basis for a first course not because of paradigm, but because of its focus on problem analysis, design for reuse, and testing, which is largely paradigm independent.
If colleges did their job and taught students business skills such as economics, accounting, project planning as well as professional programming skills such as object oriented programming, secure coding, MVC, three tier system design and SQL, and not calculus IV obscure algorithms, zxy-tree discrete unified field algorithm garbage we would not have this problem.
Teaching "business skills" (outside of a business program) and "professional programming skills" (outside of a notional "vocational programming" curriculum) are not colleges' jobs.
and not calculus IV obscure algorithms, zxy-tree discrete unified field algorithm garbage we would not have this problem.
Actually, all those things are part of colleges' job, not the preceding, at least in a Computer Science curriculum (well, probably not the things you actually name, but the actual mathematical and algorithm-related classes that lie behind your made-up names.)
HR is convinced computer science is required for any I.T. job when in fact they do not teach practical skills outside of academia.
Clearly, if HR doesn't know what degree or other qualifications are actually relevant to the jobs they are recruiting for, there is an education problem: and it lies completely with the people doing HR.
Every other profession out there is trained at a university on how to do their job.
Not even close, except, perhaps, for the professions which have actual professional degrees (Law, Medicine, etc.) and "professions" which have vocational degrees.
HR is trained from a B.A. with a minor in H.R.
Most people I've known in HR don't have a minor in HR.
Many of them don't even have a B.A.
Chemists are trained with a science degree in chemistry
Which, incidentally, teaches them a lot of theory, and not most of the vocational things that working chemists do, which if they learn at university at all they do through (a) an on-campus job, or (b) optional internships. [At least, assuming that chemistry is anything like, say, biochemistry in that regard; that's the field where I've known more people going to school and working.)
Its logical to assume HR thinks computer science teaches students how to be practical in business.
The academic major that, at least notionally, teaches students how to be practical in business is called "Business", not "Computer Science". If graduates in the former discipline don't get business, that is a problem with their education. If graduates in the latter do not, that is a problem with expectations.
The math co-processor wasn't made obsolete. It became so vital to system performance that Intel and friends started including it in on the CPU proper. These days, they call it an FPU.
A more cynical view is that as they became more popular, Intel started losing marketshare to alternative math coprocessor vendors, which they ended by putting the co-processor onto the CPU, which made it much harder for alternatives to compete.
Anyone have a rough idea of the performance comparison between the two and what bottlenecks people are likely to hit on either side? I know the GIL in Python is sometimes blamed for a lot of problems...
Ruby 1.9.1 trades green threads for native threads with a GIL, so probably has similar issues to Python (the plan is to eliminate the GIL eventually.) JRuby, I think, uses Java threads (which, on most platforms, IIRC, are native threads) without a GIL.
Well JRuby is an implementation of Ruby in Java which explains the speed but it is still quite slow as it has to 'interpret' everything into java.
JRuby can do either JIT or ahead-of-time compilation to JVM bytecode, it does not 'interpret' (or even 'compile') everything to Java.
If you are a fan of Java and like Ruby, I'd suggest Groovy. It's blazingly fast and even puts these to shame. Plus works well Rails, Struts and most MVC frameworks.
Well, Rails doesn't run in Groovy, though Groovy has a Rails-inspired framework ("Grails") of its own.
Does Egyptian military really rely on the use of GPS for their operations? The GPS satellites are controlled by the US.
Something like a quarter of the funding of the Egyptian military comes from the US government, so relying on the US for GPS isn't all that big of a deal.
However, the theory of those was (correctly or incorrectly) that they are a part of the teetering leveraged house of cards that must be kept partially standing or the resulting collapse will devastate nearly everyone
I would call that a rationalization rather than a "theory". It doesn't even begin to hold water. The form in which it was usually advanced was that, if the banks didn't have capital, everyone else would fail because without capital the banks couldn't make loans.
Of course, if the problem is "capital needs to be available to make loans", rather than giving money to banks who had demonstrating they couldn't handle it well and rewarding failure, the government could just go out and use the same money and make the loans itself to prevent the rest of the economy from collapsing (as, it turns out, it appears likely to need to do anyway as the bank bailout doesn't seem to have done any good.)
Also, most of the "bailouts" have been loans. If the entities survive, the loans will be paid back (with interest). If the entities fail in the end, the shareholders of those companies will also have lost every penny of their investment (to the extent that their investments are worth anything now).
So what? Many of the institutions took bailouts and immediately declared dividends. So shareholders whose assets were looking like they might end up being worthless (and still have that look) just extracted the cash from the bailout, and face essentially zero downside risk. If the terms had required stockholder approval which included a waiver of the shield of individual stockholders from liability for the debt incurred by taking bailout loans, there might be some substance to that, but as it is, its been a bonanza for shareholders.
Also, a LOT of people on Wall Street are losing their jobs (esp. the quants - the "wizards" I'm assuming you're referring to) in the consolidation and downsizing. They are not getting a free lunch here (just a discounted one).
The workers aren't getting a free lunch. OTOH, the capital holders (who extracted all the benefit on the upside of the economy) are getting a free lunch at the hands of the government (sure, its not as good a living as they've extracted from the market in the past, but its still a free lunch; the only thing they even risk giving up in exchange is something they wouldn't have without the handout, and might not even have with the handout.)
Presidential econ policies take years to come about.
Perhaps, though the rest of your post applies this rather inconsistently, in a way which shows a lot more partisanship than economic analysis.
Clinton rode on the success of Reagan/Bush Sr's policies.
I suppose you blame the severe 1990-1991 recession on Carter, rather than Reagan/Bush policies.
and Bush staved off a recession with his 02 tax cuts
So Bush's policies get credit for immediately "staving off" a recession (even though Bush's Presidency saw the shortest period between two succeeding recessions in half a century), despite your claim that "Presidential econ policies take years to come about"?
Our gas prices are now $1.56 so I don't care what else he does just being "different and the same" has been enough to half gas prices.
Gas prices have gone down because the economic meltdown has caused a drop in demand for moving things (including people) around.
Obama has nothing to do with it, and (while it is, itself, a nice thing for those of us still employed, for now) its a sign of how bad things are, not a positive sign.
If they can stay down for the next year or so, he's got it made.
The only way gas prices are going to stay down for the next year or so is if unemployment keeps climbing, in which case there are going to be fewer and fewer people who are happy about the gas prices.
When Obama announced that he was going to start the largest public works program since the Interstate system, I thought he might be talking about an interstate high speed rail network.
Local/regional mass transit is more likely a focus (and one of the things that has been discussed as something they are looking at making money available tos state and local governments for), since there a more existing, well-developed proposals and it would take less time to get off the ground and people working on building rather than study and planning. I'm not saying that such a network is a bad idea from an infrastructure and long-term stimulus perspective, but from a short-term stimulus perspective its far from ideal.
Printing money is a way for a Gov to tax anyone holding its currency.
Even the indication that you might do that, though, is a way to pretty instantly dry up the ability to borrow money -- either the government, or anyone else trying to borrow money denominated in your currency. Monetization of debt is a pretty poor policy choice in any case, and you can usually acheive any of the desirable effects by simply borrowing in the open market, as its the people that hold dollars now but aren't investing them in either your goods or capital markets because the short-term situation is bad for that that are going to be most likely to put money into government-issued debt, provided that things haven't gotten so bad that they perceive a serious risk of the debt being repudiated.
In which case, you're in such deep trouble already that there's probably nothing you can do but hope for a miracle, and you can try monetization without any particular risk of killing your ability to borrow (since you've already lost that), but it probably won't help.
Why not give to every US taxpayer (read, payer of taxes, not non-payers) over the age of 18, about $25K-$100K or whatever.
Well, for one thing, everyone in the US, almost without exception pays taxes (federal taxes, even), its just a matter of which taxes.
Second, that would have pretty much the same problem as giving it to corporations without controls, the only difference is that it would probably be a little more effective as stimulus because the propensity for each dollar to be spent in the domestic economy would be a little higher.
(Those people who are saying it would have no effect but price inflation are either assuming you are printing the money, rather than expending it from some combination of revenues and borrowing, or they simply have no idea what they are talking about.)
How about serious money back to the taxpayers of the US, and let us choose where to spend those dollars in our economy.
Neither US consumers nor US banks and other large corporations are likely to prioritize spending in the ways that have the most short-term stimulative effect to high velocity of money in the domestic economy, or the most long-term stimulative effect due to making transfers more efficient (e.g., by improving infrastructure.)
If the government wants to spend for stimulative purposes, it ought to assure that the use of money is carefully tailored to that effect, which means it probably ought to, largely, carefully target the spending. OTOH, bailouts, either of individuals or corporations, might conceivably have some stabilizing effects, and probably shouldn't be ruled out as tools for stabilization rather than stimulus per se.
Or the completely accurate but much less trite version:
Correlation implies either causation or mutual causation by a third factor.
That's not completely accurate. The completely accurate form is:
Degree of correlation implies a certain probability of some causal link (either direct or through a shared cause.)
Its quite possible for corresponding values from two completely unrelated sequences to show some degree of correlation, after all. If I have two sequences whose corresponding (e.g., by time) values lok like this:
S1: 1 1 2 3 4 3 2 1 1 S2: 2 2 3 4 5 4 3 2 2
I certainly might suspect that there is a tight correlation between S1 and S2, but each of them could just be random integers chosen from the range 1 to 6, inclusive. Using statistics, I can say how unlikely that coincidence is, but that doesn't mean that I can simply state as a fact that there is a causal link because there is a correlation.
The Earth's Magnetic field has reversed hundreds of times in geological history.
Tens of thousands, actually.
There is NO evidence whatsoever that it has ever caused an extinction.
True, that.
There's a potential for knocking out satellites or computers, but it's certainly not going to sterilize the earth.
It could be much worse than knocking out satellites or computers (reduced protection from radiation for the period of the weak field could plausibly cause a substantial increase in, e.g., cancer rates, and could have other environmental consequences which could have substantial social effects) but it is, as you note, not likely, even in combination with other events, to be anything like an end to life on Earth (or even "civilization as we know it").
Perhaps the most terrifying prediction is the reversal of the Earth's magnetic field (combined with untimely solar activity), a periodic event which is currently 1/4 million years overdue.
Geomagnetic field reversals are perhaps "periodic" in the sense that they happen repeatedly over time, but they aren't particularly regular; they have been known to be erratic since the 1960s. The last reversal was ~780Kya, so the contention here seems based on the assumption of a regular ~500Ky pattern. There is no reasonable basis for this assumption, as the past history of reversals has been nowhere near a regular pattern with a 500,000 year cycle.
Well, viewed generously, not until fairly late in the 1929-1933 recession, in the immediate wake of the 1929 crash he didn't do much, in 1930 he favored fairly moderate federal stimulus while asking state and local government to provide more stimulus, later he tried more significant stimulus (though still focussed on direct aid to capital with some public works), but the recession that started in 1929 didn't turn around until after Roosevelt began truly massive stimulus combined with more aggressive bank regulation to restore confidence in institutions. (Though, at least arguably, the more significant stimulus of the Hoover attempted toward the end, did play a big role, since the trough wasn't far into Roosevelt's efforts.)
One might expect that people now would prefer not to have a four year recession that even another four subsequent years of strong top-line growth will still leave overall conditions so bad that it'll be called a "depression".
When will politicians learn that increased govt spending and employment do not stimulate production and advance the economy, but in fact have the opposite effect?
They will probably never learn that because nothing so simple is true. Whether government spending has a stimulative effect or not depends on whether the spending is in an area where the funds will have a higher velocity in the domestic economy than wherever the funds were taken from (whether taxed, borrowed, or transferred from other government spending.)
If wealthy Americans have a low marginal propensity to spend, and are investing largely overseas, then taxing them and spending on labor intensive projects in the domestic economy that have their own economic utility and where most of the workforce will have a high marginal propensity to spend on goods that are supplied domestically (at least in the immediate sense, e.g., at retail) is likely to have a stimulative effect. Even more stimulative, if economic conditions have driven down direct foreign investment in US private capital markets, is borrowing money from abroad for the same purpose.
OTOH, taxing people with a higher marginal propensity to spend and transferring funds to Americans with a lower marginal proposenity to spend (as, in effect, the massive bank bailouts with little control on use have done) is exactly the opposite of stimulative.
The fact is that the US is 80% urban and suburban, so getting decent services to those folks (in both broadband and public transport) shouldn't be a problem.
Broadband, perhaps. For public transport, though, US cities and, particularly, suburbs are deliberately laid out in a way which is very good for individual transport via cars and very bad for public transit. This is not an accident, its a deliberate choice. You can't just overlay public transit on top of that and expect it to be efficient, you've got to do a lot of work (on the order of several decades and, I would guess, trillions of dollars) in transit-oriented redevelopment of communities before public transit will be anything like efficient.
Slashdot Mirror
I think Logo (in forms like NetLogo and StarLogo) is the new Logo.
Eh, I'd been programming in BASIC for three years, and Logo and Pascal for about a year when I took Algebra I. I don't see the dependency for really basic things.
I don't get endorsing C and telling people to stay away from BASIC when its "about knowing what is going on inside computers". Unstructured, old-school BASIC is, in many ways, a lot closer conceptually to what goes on inside computers than any structured programming language, C certainly included.
You can teach all that along with any programming language, probably more easily than you can without one; that's basically the approach taken by How to Design Programs (which uses a series of Scheme variants, which I think is a good way, though the general outline would be usable with standard Scheme or even a completely different language.) HtDP may be a little steep for the age group (but perhaps not if they are gifted), but you could probably follow HtDP as a "course outline" and make things friendlier, and the Dr. Scheme environment that supports its is pretty nice for educational use.
Sure, moving a turtle around drawing "senseless shapes" is probably not going to be very interesting to 11-14 year olds, especially those on the older end of that range, but then even the versions of Logo I used at 11 could be used to a lot more than that, and modern, agent-based versions of Logo are designed to do quite a lot more, easily. Take a look at NetLogo or StarLogo TNG.
There are several languages that can do functional, OO, and structured imperative programming naturally using subsets of the same language, without being forced (Scheme, for instance; Oz does all that plus logic programming, though I personally think that Scheme's Lispy syntax would be easier for beginners.) I'm not sure throwing a bunch of different languages at students in a first course is a good idea, both because there is cognitive overhead for each language and because it reinforces the idea that choice of programming paradigm is directed by choice of language.
Functional, OO, or (structured) imperative methods are all valid; I think How to Design Programs is a good basis for a first course not because of paradigm, but because of its focus on problem analysis, design for reuse, and testing, which is largely paradigm independent.
Teaching "business skills" (outside of a business program) and "professional programming skills" (outside of a notional "vocational programming" curriculum) are not colleges' jobs.
Actually, all those things are part of colleges' job, not the preceding, at least in a Computer Science curriculum (well, probably not the things you actually name, but the actual mathematical and algorithm-related classes that lie behind your made-up names.)
Clearly, if HR doesn't know what degree or other qualifications are actually relevant to the jobs they are recruiting for, there is an education problem: and it lies completely with the people doing HR.
Not even close, except, perhaps, for the professions which have actual professional degrees (Law, Medicine, etc.) and "professions" which have vocational degrees.
Most people I've known in HR don't have a minor in HR.
Many of them don't even have a B.A.
Which, incidentally, teaches them a lot of theory, and not most of the vocational things that working chemists do, which if they learn at university at all they do through (a) an on-campus job, or (b) optional internships. [At least, assuming that chemistry is anything like, say, biochemistry in that regard; that's the field where I've known more people going to school and working.)
The academic major that, at least notionally, teaches students how to be practical in business is called "Business", not "Computer Science". If graduates in the former discipline don't get business, that is a problem with their education. If graduates in the latter do not, that is a problem with expectations.
A more cynical view is that as they became more popular, Intel started losing marketshare to alternative math coprocessor vendors, which they ended by putting the co-processor onto the CPU, which made it much harder for alternatives to compete.
The US government spends lots of money making sure any such efforts would be unsuccessful.
Ruby 1.9.1 trades green threads for native threads with a GIL, so probably has similar issues to Python (the plan is to eliminate the GIL eventually.) JRuby, I think, uses Java threads (which, on most platforms, IIRC, are native threads) without a GIL.
JRuby can do either JIT or ahead-of-time compilation to JVM bytecode, it does not 'interpret' (or even 'compile') everything to Java.
Well, Rails doesn't run in Groovy, though Groovy has a Rails-inspired framework ("Grails") of its own.
Something like a quarter of the funding of the Egyptian military comes from the US government, so relying on the US for GPS isn't all that big of a deal.
As of about two years ago, my next door neighbor was still using the rotary phone that she had originally rented from AT&T.
For all I know, she still is.
I would call that a rationalization rather than a "theory". It doesn't even begin to hold water. The form in which it was usually advanced was that, if the banks didn't have capital, everyone else would fail because without capital the banks couldn't make loans.
Of course, if the problem is "capital needs to be available to make loans", rather than giving money to banks who had demonstrating they couldn't handle it well and rewarding failure, the government could just go out and use the same money and make the loans itself to prevent the rest of the economy from collapsing (as, it turns out, it appears likely to need to do anyway as the bank bailout doesn't seem to have done any good.)
So what? Many of the institutions took bailouts and immediately declared dividends. So shareholders whose assets were looking like they might end up being worthless (and still have that look) just extracted the cash from the bailout, and face essentially zero downside risk. If the terms had required stockholder approval which included a waiver of the shield of individual stockholders from liability for the debt incurred by taking bailout loans, there might be some substance to that, but as it is, its been a bonanza for shareholders.
The workers aren't getting a free lunch. OTOH, the capital holders (who extracted all the benefit on the upside of the economy) are getting a free lunch at the hands of the government (sure, its not as good a living as they've extracted from the market in the past, but its still a free lunch; the only thing they even risk giving up in exchange is something they wouldn't have without the handout, and might not even have with the handout.)
Perhaps, though the rest of your post applies this rather inconsistently, in a way which shows a lot more partisanship than economic analysis.
I suppose you blame the severe 1990-1991 recession on Carter, rather than Reagan/Bush policies.
So Bush's policies get credit for immediately "staving off" a recession (even though Bush's Presidency saw the shortest period between two succeeding recessions in half a century), despite your claim that "Presidential econ policies take years to come about"?
Gas prices have gone down because the economic meltdown has caused a drop in demand for moving things (including people) around.
Obama has nothing to do with it, and (while it is, itself, a nice thing for those of us still employed, for now) its a sign of how bad things are, not a positive sign.
The only way gas prices are going to stay down for the next year or so is if unemployment keeps climbing, in which case there are going to be fewer and fewer people who are happy about the gas prices.
Local/regional mass transit is more likely a focus (and one of the things that has been discussed as something they are looking at making money available tos state and local governments for), since there a more existing, well-developed proposals and it would take less time to get off the ground and people working on building rather than study and planning. I'm not saying that such a network is a bad idea from an infrastructure and long-term stimulus perspective, but from a short-term stimulus perspective its far from ideal.
Even the indication that you might do that, though, is a way to pretty instantly dry up the ability to borrow money -- either the government, or anyone else trying to borrow money denominated in your currency. Monetization of debt is a pretty poor policy choice in any case, and you can usually acheive any of the desirable effects by simply borrowing in the open market, as its the people that hold dollars now but aren't investing them in either your goods or capital markets because the short-term situation is bad for that that are going to be most likely to put money into government-issued debt, provided that things haven't gotten so bad that they perceive a serious risk of the debt being repudiated.
In which case, you're in such deep trouble already that there's probably nothing you can do but hope for a miracle, and you can try monetization without any particular risk of killing your ability to borrow (since you've already lost that), but it probably won't help.
Well, for one thing, everyone in the US, almost without exception pays taxes (federal taxes, even), its just a matter of which taxes.
Second, that would have pretty much the same problem as giving it to corporations without controls, the only difference is that it would probably be a little more effective as stimulus because the propensity for each dollar to be spent in the domestic economy would be a little higher.
(Those people who are saying it would have no effect but price inflation are either assuming you are printing the money, rather than expending it from some combination of revenues and borrowing, or they simply have no idea what they are talking about.)
Neither US consumers nor US banks and other large corporations are likely to prioritize spending in the ways that have the most short-term stimulative effect to high velocity of money in the domestic economy, or the most long-term stimulative effect due to making transfers more efficient (e.g., by improving infrastructure.)
If the government wants to spend for stimulative purposes, it ought to assure that the use of money is carefully tailored to that effect, which means it probably ought to, largely, carefully target the spending. OTOH, bailouts, either of individuals or corporations, might conceivably have some stabilizing effects, and probably shouldn't be ruled out as tools for stabilization rather than stimulus per se.
That's not completely accurate. The completely accurate form is:
Degree of correlation implies a certain probability of some causal link (either direct or through a shared cause.)
Its quite possible for corresponding values from two completely unrelated sequences to show some degree of correlation, after all. If I have two sequences whose corresponding (e.g., by time) values lok like this:
S1: 1 1 2 3 4 3 2 1 1
S2: 2 2 3 4 5 4 3 2 2
I certainly might suspect that there is a tight correlation between S1 and S2, but each of them could just be random integers chosen from the range 1 to 6, inclusive. Using statistics, I can say how unlikely that coincidence is, but that doesn't mean that I can simply state as a fact that there is a causal link because there is a correlation.
Tens of thousands, actually.
True, that.
It could be much worse than knocking out satellites or computers (reduced protection from radiation for the period of the weak field could plausibly cause a substantial increase in, e.g., cancer rates, and could have other environmental consequences which could have substantial social effects) but it is, as you note, not likely, even in combination with other events, to be anything like an end to life on Earth (or even "civilization as we know it").
Geomagnetic field reversals are perhaps "periodic" in the sense that they happen repeatedly over time, but they aren't particularly regular; they have been known to be erratic since the 1960s. The last reversal was ~780Kya, so the contention here seems based on the assumption of a regular ~500Ky pattern. There is no reasonable basis for this assumption, as the past history of reversals has been nowhere near a regular pattern with a 500,000 year cycle.
Well, viewed generously, not until fairly late in the 1929-1933 recession, in the immediate wake of the 1929 crash he didn't do much, in 1930 he favored fairly moderate federal stimulus while asking state and local government to provide more stimulus, later he tried more significant stimulus (though still focussed on direct aid to capital with some public works), but the recession that started in 1929 didn't turn around until after Roosevelt began truly massive stimulus combined with more aggressive bank regulation to restore confidence in institutions. (Though, at least arguably, the more significant stimulus of the Hoover attempted toward the end, did play a big role, since the trough wasn't far into Roosevelt's efforts.)
One might expect that people now would prefer not to have a four year recession that even another four subsequent years of strong top-line growth will still leave overall conditions so bad that it'll be called a "depression".
They will probably never learn that because nothing so simple is true. Whether government spending has a stimulative effect or not depends on whether the spending is in an area where the funds will have a higher velocity in the domestic economy than wherever the funds were taken from (whether taxed, borrowed, or transferred from other government spending.)
If wealthy Americans have a low marginal propensity to spend, and are investing largely overseas, then taxing them and spending on labor intensive projects in the domestic economy that have their own economic utility and where most of the workforce will have a high marginal propensity to spend on goods that are supplied domestically (at least in the immediate sense, e.g., at retail) is likely to have a stimulative effect. Even more stimulative, if economic conditions have driven down direct foreign investment in US private capital markets, is borrowing money from abroad for the same purpose.
OTOH, taxing people with a higher marginal propensity to spend and transferring funds to Americans with a lower marginal proposenity to spend (as, in effect, the massive bank bailouts with little control on use have done) is exactly the opposite of stimulative.
Broadband, perhaps. For public transport, though, US cities and, particularly, suburbs are deliberately laid out in a way which is very good for individual transport via cars and very bad for public transit. This is not an accident, its a deliberate choice. You can't just overlay public transit on top of that and expect it to be efficient, you've got to do a lot of work (on the order of several decades and, I would guess, trillions of dollars) in transit-oriented redevelopment of communities before public transit will be anything like efficient.