***If ITER is successful, a commercial reactor could be built by 2040. Funny, I seem to remember fusion researchers from Livermore in the 70s say that commercial power was 20 years away... ***
That's funny. because what I remember "them" telling us in the 1970s is that commercial fusion was 50-80 years away. Sounds to me like they might be pretty much on schedule. (Of course, they might be off by a couple of centuries also).
Maybe we were listening to different people.
BTW, I can't find any record via Google of Lawrence-Livermore people predicting commercial fusion by 1990, but that certainly doesn't prove that there weren't some of them projecting that. I really don't believe that was a majority view in 1970.
***Right. First he's playing with fusion, the next thing you know, he's building terrorist bombs! Let's ban Deuterium right quick "for the children".***
It's not just Deuterium y'know. Normal water contains about 1.5 parts per ten thousand Deuterium. Our security demands that we contain all supplies of water on the planet. I am informed that President Bush will, with heavy heart, be sending the 101st Airborne to bring the largest supply of uncontrolled Deuterium on the planet -- the Pacific Ocean -- to heel. My sources inform me that the US will lead a massive coalition of allies including Transylvania, Haute Volta, Monaco, and Pitcairns Island in this undertaking. The President's advisors anticipate no difficulties and project that the effort will pay for itself. Intelligence community advisors whose names are being withheld by request inform us that marine mammals are preparing garlands of flowers to drape over our brave liberting troops as they march into the surf at San Onofre.
===
BTW. you're correct about your comments on fear of inappropriate things. Anyone smart enough to build a Fansworth-Hirsch Fusor or similar device is probably less of a menace to themselves and humanity in general than a softball team that has stopped off for a couple of beers after game. The kid might, I suppose, on a really bad day, manage to burn the house down -- but kids (and adults) do that all the time with low tech devices like cigarettes or candles.
What I'd do is small projects with five man teams -- two coders, two testers, and a coordinator who maintains the problem list, does the builds, tracks the configuration, tries to persuade folks to meet the schedule, etc. Ideally, you'd do five projects so everyone got a shot at coordinating.
I'd also throw in at least one major specification change for each team (without changing the schedule) and at least one plausible programming document on a necessary interface that contains important errors or omissions. In fact, you might get someone with a nasty sense of humor who is not a student in this class to conjure up one major (but surmountable) suprise for every team.
The class could be a lot of fun, and you just might persuade some people who shouldn't be in CS to begin with to consider whether they have made an appropriate career choice.
***That's a funny way to spell Windows 2000. Or were you thinking of Windows NT 3.51?***
No, I didn't misspell anything. I tried W2K on several different machines and found it to be very slow, and that applications crashed even more often than on Windows 9 -- for no good reason. I was also deeply suspicious of the large number of processes that were running and the fact that I couldn't easily find out at that time what some of them were. Since I was sysadmining at that time and there was no way short of amputation to keep my users from acquiring viruses, I wasn't all that wild about not being able to easily find strange new tasks just by looking at the task list which is generally pretty short in Windows 9. And I found that installing software on W2K was all too often an adventure. Install it, then figure out where it installed and why it doesn't work. (That happens in Windows 9 sometimes, but nowhere near as often). I solved both problems by backing the machines up to Windows 98, and never had any reason to think that was not a good idea.
When Windows XP came along, I took one look at the number of services that were running and decided that no sane person would deploy that thing on a PC connected to a network. Boy did I turn out to be right about that. I actually can, just barely, tolerate Windows XP, but my P166 with Windows 9 performs better with the same software than much faster hardware running XP. And while Window 95 does crash every few weeks it actually does fewer wierd, wonderous, and quite inexplicable things than the XP machines in the house.
I kind of think that the "reliability" problems that NT fixed, while real enough, aren't, for the most part, the problems that make Windows such a nightmare to support.
***manufacturers have no choice but to accede, adding hundreds of dollars to the cost of each PC.***
I yield to no man in my bewilderment as to why the hell the 90% of computer savvy consumers who would be better off with Windows 98, Apple, or Linux insist on buying Microsoft's ever more complex, bloated, and questionably usable software. And I don't have the slightest idea why any sane person would contemplate buying a computer with Windows Vista. Maybe there are folks out there whose life includes too little aggravation.
But I don't see how Microsoft's ongoing blunder with Vista is going to bump PC prices by "hundreds of dollars". OEM pricing of Windows XP Home is something of a secret, but it's clearly under $100. Probably significantly under $100. See http://blogs.zdnet.com/Bott/?p=118 . If you actually read the first article linked in the post and get past the absurd pricing of boxed copies of Windows (gee, I wonder why Microsoft has a piracy problem.) you find that Vista Home Premium is going to cost OEMs 10% more than XP Home. I make that something like $8-$10 a copy.
Maybe I'm missing something, but it looks to me like your Vista machine might cost a few bucks extra, but no more than a couple of cups of upscale coffee.
Should Microsoft get whacked for their occasionally abominable conduct. Won't bother me a bit if they are. But it won't likely happen until a Democratic president appoints a Democratic Attorney General. That's barring the unexpected appearance of the reincarnation of Teddy Roosevelt. Don't expect the Democrats to do anything much before 2008. They are, if they have any sense at all, going to be focusing on making sure that there are new headlines every week spotlighting some idiocy or other of the Bush administration during the five year reign of stupidity following 9/11. Fortunately for them, there is a near infinite supply of tragedy, lunacy, corruption, and even a few honest mistakes for them to draw on. No need to demonize Microsoft when Dick Cheney and George W Bush are stuck there in the headlights and there is enough ammunition handy to blast them and anyone who ever appeared in a photo with them into orbit around Alpha Centauri.
I had something to protect*** Bob Dylan. "My back Pages."
ANY form of security is a pain in the ass. Given that hardly anything in this industry works quite right, it's a safe bet that anything new or or complicated has a high probability of being a bundle of grief. I wouldn't go near biometrics unless you have some really stringent and unusual requirements... but that's just me. Most IT people are less pain-adverse and seem to enjoy beating themselves up. (What could possibly go wrong with biometrics? How about the stupid hardware only recognizes your boss on alternate Tuedays? Or some days, it doesn't recognize YOU? The possibilities for embarassing/annoying problems are just about boundless. -- maybe you better keep manual login around as an alternative.)
So, the first question is -- Do you really have anything to protect? And is protecting it going to be cost effective?
If the answer is yes, I suppose you might be able to do something not too painful with USB flash memory. Put the user's ID data and software that enables whatever the hell it is that you want to do on the flash drive, Employee wants to use the machine. He or she plugs the flash drive in. Have the employees lock their flash drives up when they leave the building so you don't have hundreds of them lost. You can probably make that work -- even under Windows. But it probably will not be especially easy nor especially cheap and will have hidden costs. It'd probably have fewer of the latter under Linux.
I suppose you might also be able to automatically log Windows on to one general user on boot and bring up a shell that had it's own primitive (but fast) login logic to give users access to whatever the heck it is that they need access to.
BTW, if you can back off to Windows 9 and can use an old version of Netware for the server, multi user will be pretty much broken. Nothing fancy will work, but your users may be able to login in a not too painful amount of time and you'll be able to use Netware access controls to determine which files they can access and alter.
***When... will we finally get our goddamn flying cars?***
My guess is not too long after hell freezes over. Sounds to me like you are expecting folks many of whom can't navigate all that well in two dimensions, and will drive for months with a CHECK ENGINE light on to handle aircraft safely. Is that realistic?
Maybe someday there will be a technology that will allow people to take to the air without being a menace to themselves, other flyers and anybody on the ground for miles around. But I don't think anything remotely resembling a Chevrolet with wings is that technology.
***But when will it become truly affordable for the masses? That's what most of us want to know. Wake me when it's time to disconnect from the petroleum/nuclear fired grid.***
It's affordable today for domestic water heating.... if you live in a climate where water isn't frozen solid much of the year. That's a non-trivial expense for folks in industrialized countries. And it's affordable in some remote locations where the cost of running in a power line is extortionate. Every reduction in cost increases the number of niche situations where solar is practical and maybe even preferable to the alternatives.
Other than that. I'll believe that it is affordable when real plants come on line and have been delivering cheap power to the grid for a decade. I'm old enough to remember promises of nuclear power too cheap to meter. I'm a supporter of nuclear power BTW, because it is relatively non-polluting when compared to the true pollution effects of fossil fuels. But it's actually a bit more expensive than hydro or the least expensive fossil fuel plants.
As for taking yourself off the grid. Unless you have a waterfall in your back yard, (or maybe a fast flowing river that you can drop a turbine into without having the local authorites going ballistic) I don't think you are likely to be able to do that without a lot of expense and compromises in living style for three or four decades.
***Now I'm using IE7 as my main browser (quiet!) and don't anticipate any problems with it, either.***
Reminds me of a progress review many decades ago at a large defense contractor who shall remain nameless. The project in question was the development of a large phased array radar, and the topic was the steps that had been taken to alleviate the erratic performance from the very large and very expensive computer used to drive the radar. The problem had been traced to another project in the same facility that occasionally fired up its radar and emitted enough EMI to scramble the data in the first project's computer's circuits. The solution was mostly not to run the computer when the radar was being tested.
A USAF officer waved his hand and asked if just maybe there might be an interference problem when the computer was deployed in a room just behind it's own multimegawatt radar antenna.
"We don't anticipate a problem" said the guy from Nameless.
"Did you anticipate a problem with the other project here?" asked the officer.
"Well... ehrrr... no"
Just sayin. Failure to anticipate a problem doesn't mean there won't be any
BTW, I haven't seen a virus on this Win95 system ever. Doesn't prove that viruses don't exist. Or that Windows 95 is secure..
Me? I'd use an obscure 18th Century procedure called an Arrest Warrant. That has a few advantages -- like allowing an arrest instead of merely freeing the terrorist to either get a false ID or blow up something or somebody in the US instead someplace else. But that's just me.
The drawback to an arrest warrant? You actually need a bit of credible evidence against the person in order to get one.
The first thing to do is talk to the lawyers and make sure that they understand EXACTLY what has happened. If there is, or might be, a legal obligation to make disclosures, the company would have to be run by total fools not to do so. If the lawyers say disclose and the management waffles or decides not to, it's probably time to bail.
Second, all the smoke and mirrors notwithstanding, Windows 9 probably is not much more (or less) insecure than NT based Windows. They both suck as far as I can see. If anything, Windows 9 may be better for two not particularly admirable reasons. First, newer and more sophisicated attacks tend to focus on Windows 2K and XP flaws, which often don't exist in older versions. Or they use API calls and Services not available in the old software. Second, Windows 9 is a lot simpler and therefore easier to work with. The chances of detecting and/or preventing future attacks are probably better.
Third, putting any machine working with sensitive data on the Internet is not a very good idea. In fact, the most sensible practice for any operation in the current or likely future state of computer security is to segregate really important sensitive data and keep it on machines with NO network connection. That isn't always possible of course. But to the extent it can be done, it should be. I don't know what the current military/government policies are, but two decades ago when I last worked with classified data, connecting ANY machine with ANY classified data to a network that wasn't isolated from the outside world would have been a major security violation and would probably have resulted in immediate loss of individual and maybe facility security clearances. I don't see any reason to believe that wasn't a wise and prudent policy.
(Lest anyone protest that there was a provision for connecting to less or un- secure systems even back then. They would be correct. But the hurdles that had to be overcome to do it were so great, that it was hardly ever done)
***What is so hard about the concept of a program that can go out to the Internet, look at what is there and renders it for me. WITH NO WAY TO CHANGE ANYTHING ON MY COMPUTER.
Is that so much to ask for, of ANY browser?***
Apparently it is. Web site designers are absolutely certain that you need a gazillion goodies and stand ready to deliver them whether YOU (or I) want them or not. With a few exceptions -- The Google home page- renders usably in just about any browser ever written and does not depend on Ajax, Java, Flash, or black magic-- these guys are engaged in a red queens race not seen since the great tailfin and porthole battle engaged in by US car makers in the late 1950s.
But this is just plain silly and the users hate it? Of course. But the web folks seem to think that feedback is something that makes amplifiers whine, not something that is relevant to their job. Customers -- what do they know?
***If you read the regulation proposal, what this regulation change would actually do is require manifests to be transmitted to US Customs before the aircraft pushes back from the gate, rather than 15 minutes after takeoff (which is the current regulation), so that DHS can have do-not-fly list passengers removed from the flight before it takes off rather than causing a possible situation in the air.***
Let's see if I have this straight? You think that it is important to KEEP terrorists in the US and not to let the bastards go elsewhere? No way those guys are getting out of this country once the elite federal anti-terrorism troops get wind of them. They'll just have to buckle down and blow things up here.
You surely don't take the Do Not Fly List seriously? It is, so far as anyone with any sense at all can see just another item in the long list of stupid neoconservative fiascos. You can feel secure, the president of Bolivia (who is on the list three times) is not going to sneak in or out of the US on one of OUR planes... unless of course, he has a false ID which might be pretty easy to come by since he runs a country and can probably get a Bolivian passport in any name he wants. In fact, I imagine that a serious terrorist, just might choose to use a false ID, and I can't think why he or she would have difficulty coming by one.
***Really, I don't know why any US companies can do business with China. China does terrible, horrible things to their people. We're talking on par with Cuba, Iraq, and many rogue African and S. American countries.***
Dead on man. All civilized countries should boycott countries that ignore Habeaus Corpus, hold prisoners indefinitely without trial, refuse to let defendents see the evidence against them, ignore the Geneva Convention, and retain the death penalty despite overwhelming evidence that a high percentage of those executed are actually innocent.
But don't you expect a few economic consequences in the US when civilized countries like Canada and the EU cut off the petroleum we are addicted to, stop buying our aircraft, and calls in their loans?
As a true patriot, (who successfully dodged service in Vietnam like all the other true patriots) I'm with you man. It'll be tough, but when it's all over maybe we'll be the country that folks like Washington, Jefferson, and Franklin had in mind.
To the barricades..."Allons enfants de la patrie, Le jour de gloire est arrivé ! Contre nous de la tyrannie L'étendard sanglant..."
OK then. But how about you get your facts straight? What NPR is complaining about is that many of their stations are on the frequencies at the low end of the FM spectrum and generally are not ALLOWED by their license to use a lot of power. The problem is that the makers of external device to FM translators either preset the devices for XM, Sirius, MP3 players etc, to 88.1 or to the first "empty" channel they find then broadcast at high, and often illegal, levels. I'm pretty that ANY power level that interferes with an existing service on these channels is illegal even if it might be legal on some other channel. And at some power level, the signal is illegal on any channel even if no one is using that channel.
These translators are a secondary service. i.e. they got there AFTER the licensed NPR or whatever religious, Canadian or Mexican stations preoccupy the channels. Legally and morally, it is up to the translators to stay clear of the services that are already there. This is EXACTLY the sort of thing that the FCC is supposed to handle, and NPR and other broadcasters who are being interfered with are well within their rights to complain. It is exactly the same situation as if your neighbor's wireless router were defective or poorly designed and was laying down a strong interfering signal on top of your favorite TV station.
***Open Office or the like...? There are still local apps that compete with MSOffice. Free local apps in this case still might (ought?) beat web ones.***
You're correct, but that's a different issue -- why do folks pay Microsoft money for MSOffice when there are free products that do the job every bit as well? Beats me. Yes, there are a small number of Microsoft office product users that genuinely need the real thing because they need VBA macros or because of their support arrangements and such. I'd put the number at maybe 5-15% of the MSOffice users. Maybe that's too low. Could be 50% for all I know. But in any case, there are surely a bunch of folks that'd be better off with Open Office or Abiword.
But I think the subject here is why people would choose a slow, limited, web oriented tool over a faster, more capable MSOffice tool.
Basically you're right about bandwidth. For the majority of users today -- and for many, many years to come -- web users are going to get dismal performance from web based applications.
And I think it's unlikely that many people whose jobs cause them to live in Word or Excel for 8 hours a day, 5 days a week, are going to change to on-line alternatives any time soon.
But... the are other factors. Budget for example. A lot of users do about three documents a year and spreadsheets maybe twice a decade. Is it more cost effective for them to stuggle a bit with network speed once in a great while or to pay a few hundred bucks plus support costs for Word and Excel? I suspect that in a lot of cases -- especially outside the developed world -- the answer is going to be let's use the network we are already paying for and put up with a little slow. Even in the US, I think many K-6 schools for example are eventually going to think twice about plunking down bucks to put MSOffice on every PC. (Even with educational discounts, MS software isn't all that cheap when you are talking hundreds of copies).
And collaboration. The online document and spreadsheet programs already are more convenient than shipping files around -- especially given the tower of Babel document format situation that the industry has created. "Dear Agent 007, Her majesty's government greatly appreciates your risking your life to obtain plans describing the Mongolian nuclear weapons program. However, M says that he is unable to figure out their file format and that his folks have never heard of a.YAK file. He wishes to know if, once you have finished your dalliance with the nice blond lady spy, you could possibly resend the files as ASCII text with the images converted to bitmaps or, at worst,.JPG? We will be looking forward to hearing from you"
There are probably other advantages to on-line document and spreadsheet creation. Cost and Collaboration are just two that happen to come to mind.
***When, oh when, will we be able to use what we paid for for what we want, within the limits of the law, without asking permission.***
When you either:
A, Elect a bunch of socialists (which would, in my opinion, maybe not be the best of all possible moves, but would beat the hell out of the crew of stunningly incompetent right wing flakes currently running the US). Maybe if you have Nader, Stallman et al in or advising the Congress, you'll get laws that reflect your wants and needs rather than those of the World's Corporations. Until then, it's DMCA, software patents, and worse.
B, You simply walk away from ANY PRODUCT that incorporates any form of copy protection, digital rights management, etc.
BTW, the things that set Microsoft on its road to success way back when were usable products at fair prices without copy protection. Since they have abandoned that policy, what possible reason could there be for most people to continue to buy their products? As far as I can see, the 90% of desktop users who don't actually need a crash resistant kernel, the latest version of direct-X, or one of the other small number of genuine features of XP, should have stayed with Windows 9, Only if you all -- collectively -- do things like that, are you going to convince Microsoft that there is a reason to modify its behavior ,
And don't overlook the two problems with solar generated electricity over and above deploying the technology -- storage and distribution. It is well and good to generate a gazillion Megawatt hours somewhere between Mojave and Boron California, for maybe an 8 hour production day in July. But some years in December, we are only going to get power for six hours a day and some weeks only three days because it is overcast and drizzling the other four days (yes, some years it drizzles in the Mojave). How do we deal with not having the massive solar plants on line when folks on the East Coast start up the machinery of commerce at 0500 AM PST? There are plenty of possibly feasible storage technologies that could allow us to use excess power today or this week, to generate power tomorrow or next week. but, I believe, there are essentially no meaningful storage facilities currently on line.
There is also the non-trivial problem of getting electricity from where the sun shines a lot to where people need to use it. The US grid is pretty fragile and nowhere near being up to shipping massive amounts of power across the country. During the California power crisis five years ago, they couldn't even get power between the Northern and Southern parts of the state. That problem is probably solvable using existing technology. But it might take a decade or two to solve it. I don't see anyone solving it or any leadership from either political party in the US toward getting it solved.
Your argument is basically that encyclopedias (and room mates) are unreliable sources. No Sale. They may be poor sources. They may not. If they are, and they are depended on then a lousy product will result.
IMO, barring cheating and exercises designed to teach process -- what should be graded is the product not the process. I assume that as a teacher you would be at least somewhat qualified to judge product quality. How and why does the ability to judge product, qualify one to judge process? Especially in as much as not everyone agrees with you as to what an acceptable process is. Why go out of your way to do something that you possibly can't do well, when you have an alternative that you presumably can do well?
If you spend much time looking at the Wikipedia in areas that you have real expertise in, you'll probably find that much of it is quite good and is perfectly suitable as a reference. Problem is that some of it isn't and cross checking is desirable if the material is important. In an awful lot of cases, you'll find that "Real reference material" on the same topic is substantially WORSE than the Wikipedia. An example picked more or less at random. Compare the Wikipedia article on The Nanking Massacre to articles by Japanese nationalists on the one hand or Iris Chang's "Rape of Nanking" on the other. Which is the best and fairest source? Almost certainly the Wikipedia. But Chang's work when it was published in 1977 was regarded by many as a stunning piece of scholarship and the last word on the subject -- surely an acceptable primary reference by your standards.
They didn't design the PS/2 port without an Interrupt. It's (usually) on IRQ12 on ISA buses. But I think that they did design it without a hot plug notification -- because it never crossed their mind that anyone would try. PS/2 ports are almost always used for mice. I don't think I've ever seen or heard of one used for anything else, although I'm sure it must happen. Windows 9 at least doesn't even support hot-plugging a PS/2 mouse (It does support hot plugging serial mice). Linux apparently does, but since you can't depend on the device being plugged in to generate an interrupt when powered up, you would need to poll to support hotplugging.
Apparently you don't know much about PCs -- at least not older ones. The widely copied (cloned or pirated depending on your viewpoint) parallel port on the original IBM PC either was miswired or misdocumented. I can't recall which. Anyway, it didn't do interrupts. As a result, well up into the 1990s, most PCs couldn't interrupt on the parallel port. No one had all that much trouble dealing with it.
You were designing I/O devices and writing real-time I/O code before there were computer science teachers or courses? So was I. Wrote my first code in 1961 (or maybe it was 1960 -- Long time ago anyway). If your devices and code worked that's fine. But it doesn't make you right about what is nonsense and what isn't. In many cases, input intervals are predictable. If they are, you can often plan your code to be waiting for the input. The process of doing that is called "Engineering". (Not to be confused with "Software Engineering" -- an artform that bears roughly the same relationship to engineering that sculpture does to architecture). If you are able to engineer your system -- and not all systems can be engineered, polling will probably be superior and is worth considering. If you can't predict when inputs will arrive then you have to use interrupts.
As I have to keep on repeating, there are times and places where interrupts are appropriate. But your conviction that they are always superior to polling is simply wrong. To see that, take the limiting case where you must process a high data rate input stream on a computer that can barely keep up. That used to be a very common situation, and I imagine that it still is. Polling requires no context switching and can, on some hardware, be done with one instruction. An interrupt requires at least two context switches -- one to get to the processing code and one to get back. A polling loop probably will not require any register saves because it is inserted where it does not disrupt other activities. An interrupt will minimally require saving and restoring every reqister you use plus the program counter. Are seriously suggesting that you can process as fast a data stream with interrupts as the guy at the next desk can with polling? In your heart, you know you can't.
***...and as such, should not be used as a reference in any research above the grade-school level. Period. If I were teaching a college class and anyone used encyclopedias in their paper, I wouldn't give them above a C.***
No offense, but that's just silly.
The proper criterion for judging a (non-plagiarized) paper is not "Where did the information come from?". The proper criteria are things like whether the material is complete, accurate, well reasoned, and clearly presented. With rare, and obvious, exceptions, the source is no more relevant than whether any conclusions in the paper coincide with your personal predjudices and biases.
A historical note. Ethan Allen was well over six feet tall in an era when most men weren't much taller than 5 feet. He was apparently articulate, profane and was known to down a belt or two... or three. At various times he was a guerilla leader, a petitioner to the Continental Congress for statehood for Vermont, and a negotiator with Britian attempting to attach Vermont to Canada as a separate province. It appears entirely possible that Allen's capture of Ft Ticonderoga had its roots in a drinking bout at the Catamount Tavern in Bennington in 1775. "Hey man, let's go up and take Ticonderoga away from them slimey brits". "Cool with me, round up them boys that can still walk, grab your guns and let's go..."
BTW, they somewhat uncleverly walked up the wrong side of Lake Champlain. The size of the attack force was limited by the fact that they could only find one boat to ferry men to the other shore, and it wasn't very big. (Remind you of any projects you've worked on?). In the long run, it didn't matter. There were only 22 British troops garrisoned at Ticonderoga and 21 of them were asleep when Allen, Benedict Arnold and a few dozen of the Green Mountain Boys strolled in the gates.
Anyway, pitching a chair at Allen probably would have been what we call in the trade a really, really bad idea. There is an excellent chance that it would be the last chair the thrower would launch for quite some time.
***They poll I/O ports?! Have these people never heard of hardware interrupts? I knew that a lot of lore had been lost in the PC revolution, but I had no idea the situation was this bad.***
I don't think that the PS/2 Auxiliary Device Port is designed to generate an interrupt when something is hot plugged into it..
BTW -- notwithstanding what your computer science teachers taught you, polling is quite efficient if loads are predictable. Polling is usually much less resource intensive than interrupts... if the polls have a high hit rate. And it's much less subject to wierd, difficult to reproduce problems and to race conditions. Across the broad spectrum of computing, there are probably far more cases where interrupts are used when polling would work better than vice versa.
That said, 20 polls per second seems excessive for a detecting a new device on a port that is rarely used except for mice. Once every 5 seconds would seem more appropriate.
***Private investment in space is the only thing that will change it from a huge, shiny waste of tons of money to a useful endeavor.***
Good logic based on two faulty assumptions:
Space exploration can be profitable using current or currently achievable technology.
Governments can't do anything right
With current technology, there are a (pitiful) few applications where space can be profitably exploited. Communications, imaging, one or two others. There is not and has not been any problem attracting private investment in those areas. Point: When things evolve to the point that money can be made, there will probably be plenty of private money put up. Unfortunately, space overall is, and is likely to remain, a huge money pit. The fantasies of various taxa of space cadet notwithstanding, there is no sign this situation is going to change any time soon.
We can't get stuff to orbit cheaply. Until we can, we aren't going anywhere.
As to the second assumption. Governments sometimes push through projects that private investment is unwilling to gamble on. The best example is the Erie Canal which was built by the state of New York after the money men in New York ignored pleas of advocates for years. The Erie Canal and its less known sister the Champlain Canal were wildly successful. The Erie dropped shipped costs to what is now the American heartland by an order of magnitude and opened a path for settlement of the West. It was largely responsible for New York City becoming the predominant city in the US. Without the canal, serious settlement would have waited a couple of decades for the railroads. And the canal was also profitable -- wildly so. But is a way, the money guys were right. Many of the early 19th Century canal projects in the US were financial disasters.
Please do not take this is an endorsement of NASA. Transport NASA to New York in 1819 and those guys would wave flags, make speaches, and build a wildly over budget ditch that connected Albany to Schenectady via New York City. (For those not familiar with New York geography, the Hudson River connects Albany and NYC and is navigable every inch of the way. Schenectady is about a days walk West of Albany)
While I voted for Al Gore in 2000 and think that subsequent events have proved quite convincingly that he is far more competent than George W Bush, that does not mean that I think that either or those beauties was or is qualified to be president of the United States (or anything else).
Slashdot Mirror
That's funny. because what I remember "them" telling us in the 1970s is that commercial fusion was 50-80 years away. Sounds to me like they might be pretty much on schedule. (Of course, they might be off by a couple of centuries also).
Maybe we were listening to different people.
BTW, I can't find any record via Google of Lawrence-Livermore people predicting commercial fusion by 1990, but that certainly doesn't prove that there weren't some of them projecting that. I really don't believe that was a majority view in 1970.
It's not just Deuterium y'know. Normal water contains about 1.5 parts per ten thousand Deuterium. Our security demands that we contain all supplies of water on the planet. I am informed that President Bush will, with heavy heart, be sending the 101st Airborne to bring the largest supply of uncontrolled Deuterium on the planet -- the Pacific Ocean -- to heel. My sources inform me that the US will lead a massive coalition of allies including Transylvania, Haute Volta, Monaco, and Pitcairns Island in this undertaking. The President's advisors anticipate no difficulties and project that the effort will pay for itself. Intelligence community advisors whose names are being withheld by request inform us that marine mammals are preparing garlands of flowers to drape over our brave liberting troops as they march into the surf at San Onofre.
===
BTW. you're correct about your comments on fear of inappropriate things. Anyone smart enough to build a Fansworth-Hirsch Fusor or similar device is probably less of a menace to themselves and humanity in general than a softball team that has stopped off for a couple of beers after game. The kid might, I suppose, on a really bad day, manage to burn the house down -- but kids (and adults) do that all the time with low tech devices like cigarettes or candles.
I'd also throw in at least one major specification change for each team (without changing the schedule) and at least one plausible programming document on a necessary interface that contains important errors or omissions. In fact, you might get someone with a nasty sense of humor who is not a student in this class to conjure up one major (but surmountable) suprise for every team.
The class could be a lot of fun, and you just might persuade some people who shouldn't be in CS to begin with to consider whether they have made an appropriate career choice.
No, I didn't misspell anything. I tried W2K on several different machines and found it to be very slow, and that applications crashed even more often than on Windows 9 -- for no good reason. I was also deeply suspicious of the large number of processes that were running and the fact that I couldn't easily find out at that time what some of them were. Since I was sysadmining at that time and there was no way short of amputation to keep my users from acquiring viruses, I wasn't all that wild about not being able to easily find strange new tasks just by looking at the task list which is generally pretty short in Windows 9. And I found that installing software on W2K was all too often an adventure. Install it, then figure out where it installed and why it doesn't work. (That happens in Windows 9 sometimes, but nowhere near as often). I solved both problems by backing the machines up to Windows 98, and never had any reason to think that was not a good idea.
When Windows XP came along, I took one look at the number of services that were running and decided that no sane person would deploy that thing on a PC connected to a network. Boy did I turn out to be right about that. I actually can, just barely, tolerate Windows XP, but my P166 with Windows 9 performs better with the same software than much faster hardware running XP. And while Window 95 does crash every few weeks it actually does fewer wierd, wonderous, and quite inexplicable things than the XP machines in the house.
I kind of think that the "reliability" problems that NT fixed, while real enough, aren't, for the most part, the problems that make Windows such a nightmare to support.
I yield to no man in my bewilderment as to why the hell the 90% of computer savvy consumers who would be better off with Windows 98, Apple, or Linux insist on buying Microsoft's ever more complex, bloated, and questionably usable software. And I don't have the slightest idea why any sane person would contemplate buying a computer with Windows Vista. Maybe there are folks out there whose life includes too little aggravation.
But I don't see how Microsoft's ongoing blunder with Vista is going to bump PC prices by "hundreds of dollars". OEM pricing of Windows XP Home is something of a secret, but it's clearly under $100. Probably significantly under $100. See http://blogs.zdnet.com/Bott/?p=118 . If you actually read the first article linked in the post and get past the absurd pricing of boxed copies of Windows (gee, I wonder why Microsoft has a piracy problem.) you find that Vista Home Premium is going to cost OEMs 10% more than XP Home. I make that something like $8-$10 a copy.
Maybe I'm missing something, but it looks to me like your Vista machine might cost a few bucks extra, but no more than a couple of cups of upscale coffee.
Should Microsoft get whacked for their occasionally abominable conduct. Won't bother me a bit if they are. But it won't likely happen until a Democratic president appoints a Democratic Attorney General. That's barring the unexpected appearance of the reincarnation of Teddy Roosevelt. Don't expect the Democrats to do anything much before 2008. They are, if they have any sense at all, going to be focusing on making sure that there are new headlines every week spotlighting some idiocy or other of the Bush administration during the five year reign of stupidity following 9/11. Fortunately for them, there is a near infinite supply of tragedy, lunacy, corruption, and even a few honest mistakes for them to draw on. No need to demonize Microsoft when Dick Cheney and George W Bush are stuck there in the headlights and there is enough ammunition handy to blast them and anyone who ever appeared in a photo with them into orbit around Alpha Centauri.
Yes, my guard stood hard when abstract threats
Too noble to neglect
Deceived me into thinking
I had something to protect*** Bob Dylan. "My back Pages."
ANY form of security is a pain in the ass. Given that hardly anything in this industry works quite right, it's a safe bet that anything new or or complicated has a high probability of being a bundle of grief. I wouldn't go near biometrics unless you have some really stringent and unusual requirements ... but that's just me. Most IT people are less pain-adverse and seem to enjoy beating themselves up. (What could possibly go wrong with biometrics? How about the stupid hardware only recognizes your boss on alternate Tuedays? Or some days, it doesn't recognize YOU? The possibilities for embarassing/annoying problems are just about boundless. -- maybe you better keep manual login around as an alternative.)
So, the first question is -- Do you really have anything to protect? And is protecting it going to be cost effective?
If the answer is yes, I suppose you might be able to do something not too painful with USB flash memory. Put the user's ID data and software that enables whatever the hell it is that you want to do on the flash drive, Employee wants to use the machine. He or she plugs the flash drive in. Have the employees lock their flash drives up when they leave the building so you don't have hundreds of them lost. You can probably make that work -- even under Windows. But it probably will not be especially easy nor especially cheap and will have hidden costs. It'd probably have fewer of the latter under Linux.
I suppose you might also be able to automatically log Windows on to one general user on boot and bring up a shell that had it's own primitive (but fast) login logic to give users access to whatever the heck it is that they need access to.
BTW, if you can back off to Windows 9 and can use an old version of Netware for the server, multi user will be pretty much broken. Nothing fancy will work, but your users may be able to login in a not too painful amount of time and you'll be able to use Netware access controls to determine which files they can access and alter.
My guess is not too long after hell freezes over. Sounds to me like you are expecting folks many of whom can't navigate all that well in two dimensions, and will drive for months with a CHECK ENGINE light on to handle aircraft safely. Is that realistic?
Maybe someday there will be a technology that will allow people to take to the air without being a menace to themselves, other flyers and anybody on the ground for miles around. But I don't think anything remotely resembling a Chevrolet with wings is that technology.
It's affordable today for domestic water heating. ... if you live in a climate where water isn't frozen solid much of the year. That's a non-trivial expense for folks in industrialized countries. And it's affordable in some remote locations where the cost of running in a power line is extortionate. Every reduction in cost increases the number of niche situations where solar is practical and maybe even preferable to the alternatives.
Other than that. I'll believe that it is affordable when real plants come on line and have been delivering cheap power to the grid for a decade. I'm old enough to remember promises of nuclear power too cheap to meter. I'm a supporter of nuclear power BTW, because it is relatively non-polluting when compared to the true pollution effects of fossil fuels. But it's actually a bit more expensive than hydro or the least expensive fossil fuel plants.
As for taking yourself off the grid. Unless you have a waterfall in your back yard, (or maybe a fast flowing river that you can drop a turbine into without having the local authorites going ballistic) I don't think you are likely to be able to do that without a lot of expense and compromises in living style for three or four decades.
Reminds me of a progress review many decades ago at a large defense contractor who shall remain nameless. The project in question was the development of a large phased array radar, and the topic was the steps that had been taken to alleviate the erratic performance from the very large and very expensive computer used to drive the radar. The problem had been traced to another project in the same facility that occasionally fired up its radar and emitted enough EMI to scramble the data in the first project's computer's circuits. The solution was mostly not to run the computer when the radar was being tested.
A USAF officer waved his hand and asked if just maybe there might be an interference problem when the computer was deployed in a room just behind it's own multimegawatt radar antenna.
"We don't anticipate a problem" said the guy from Nameless.
"Did you anticipate a problem with the other project here?" asked the officer.
"Well ... ehrrr ... no"
Just sayin. Failure to anticipate a problem doesn't mean there won't be any
BTW, I haven't seen a virus on this Win95 system ever. Doesn't prove that viruses don't exist. Or that Windows 95 is secure..
The drawback to an arrest warrant? You actually need a bit of credible evidence against the person in order to get one.
Second, all the smoke and mirrors notwithstanding, Windows 9 probably is not much more (or less) insecure than NT based Windows. They both suck as far as I can see. If anything, Windows 9 may be better for two not particularly admirable reasons. First, newer and more sophisicated attacks tend to focus on Windows 2K and XP flaws, which often don't exist in older versions. Or they use API calls and Services not available in the old software. Second, Windows 9 is a lot simpler and therefore easier to work with. The chances of detecting and/or preventing future attacks are probably better.
Third, putting any machine working with sensitive data on the Internet is not a very good idea. In fact, the most sensible practice for any operation in the current or likely future state of computer security is to segregate really important sensitive data and keep it on machines with NO network connection. That isn't always possible of course. But to the extent it can be done, it should be. I don't know what the current military/government policies are, but two decades ago when I last worked with classified data, connecting ANY machine with ANY classified data to a network that wasn't isolated from the outside world would have been a major security violation and would probably have resulted in immediate loss of individual and maybe facility security clearances. I don't see any reason to believe that wasn't a wise and prudent policy.
(Lest anyone protest that there was a provision for connecting to less or un- secure systems even back then. They would be correct. But the hurdles that had to be overcome to do it were so great, that it was hardly ever done)
Is that so much to ask for, of ANY browser?***
Apparently it is. Web site designers are absolutely certain that you need a gazillion goodies and stand ready to deliver them whether YOU (or I) want them or not. With a few exceptions -- The Google home page- renders usably in just about any browser ever written and does not depend on Ajax, Java, Flash, or black magic-- these guys are engaged in a red queens race not seen since the great tailfin and porthole battle engaged in by US car makers in the late 1950s.
But this is just plain silly and the users hate it? Of course. But the web folks seem to think that feedback is something that makes amplifiers whine, not something that is relevant to their job. Customers -- what do they know?
Let's see if I have this straight? You think that it is important to KEEP terrorists in the US and not to let the bastards go elsewhere? No way those guys are getting out of this country once the elite federal anti-terrorism troops get wind of them. They'll just have to buckle down and blow things up here.
You surely don't take the Do Not Fly List seriously? It is, so far as anyone with any sense at all can see just another item in the long list of stupid neoconservative fiascos. You can feel secure, the president of Bolivia (who is on the list three times) is not going to sneak in or out of the US on one of OUR planes ... unless of course, he has a false ID which might be pretty easy to come by since he runs a country and can probably get a Bolivian passport in any name he wants. In fact, I imagine that a serious terrorist, just might choose to use a false ID, and I can't think why he or she would have difficulty coming by one.
Dead on man. All civilized countries should boycott countries that ignore Habeaus Corpus, hold prisoners indefinitely without trial, refuse to let defendents see the evidence against them, ignore the Geneva Convention, and retain the death penalty despite overwhelming evidence that a high percentage of those executed are actually innocent.
But don't you expect a few economic consequences in the US when civilized countries like Canada and the EU cut off the petroleum we are addicted to, stop buying our aircraft, and calls in their loans?
As a true patriot, (who successfully dodged service in Vietnam like all the other true patriots) I'm with you man. It'll be tough, but when it's all over maybe we'll be the country that folks like Washington, Jefferson, and Franklin had in mind.
To the barricades ..."Allons enfants de la patrie, Le jour de gloire est arrivé ! Contre nous de la tyrannie L'étendard sanglant ..."
These translators are a secondary service. i.e. they got there AFTER the licensed NPR or whatever religious, Canadian or Mexican stations preoccupy the channels. Legally and morally, it is up to the translators to stay clear of the services that are already there. This is EXACTLY the sort of thing that the FCC is supposed to handle, and NPR and other broadcasters who are being interfered with are well within their rights to complain. It is exactly the same situation as if your neighbor's wireless router were defective or poorly designed and was laying down a strong interfering signal on top of your favorite TV station.
You're correct, but that's a different issue -- why do folks pay Microsoft money for MSOffice when there are free products that do the job every bit as well? Beats me. Yes, there are a small number of Microsoft office product users that genuinely need the real thing because they need VBA macros or because of their support arrangements and such. I'd put the number at maybe 5-15% of the MSOffice users. Maybe that's too low. Could be 50% for all I know. But in any case, there are surely a bunch of folks that'd be better off with Open Office or Abiword.
But I think the subject here is why people would choose a slow, limited, web oriented tool over a faster, more capable MSOffice tool.
And I think it's unlikely that many people whose jobs cause them to live in Word or Excel for 8 hours a day, 5 days a week, are going to change to on-line alternatives any time soon.
But ... the are other factors. Budget for example. A lot of users do about three documents a year and spreadsheets maybe twice a decade. Is it more cost effective for them to stuggle a bit with network speed once in a great while or to pay a few hundred bucks plus support costs for Word and Excel? I suspect that in a lot of cases -- especially outside the developed world -- the answer is going to be let's use the network we are already paying for and put up with a little slow. Even in the US, I think many K-6 schools for example are eventually going to think twice about plunking down bucks to put MSOffice on every PC. (Even with educational discounts, MS software isn't all that cheap when you are talking hundreds of copies).
And collaboration. The online document and spreadsheet programs already are more convenient than shipping files around -- especially given the tower of Babel document format situation that the industry has created. "Dear Agent 007, Her majesty's government greatly appreciates your risking your life to obtain plans describing the Mongolian nuclear weapons program. However, M says that he is unable to figure out their file format and that his folks have never heard of a .YAK file. He wishes to know if, once you have finished your dalliance with the nice blond lady spy, you could possibly resend the files as ASCII text with the images converted to bitmaps or, at worst, .JPG? We will be looking forward to hearing from you"
There are probably other advantages to on-line document and spreadsheet creation. Cost and Collaboration are just two that happen to come to mind.
When you either:
A, Elect a bunch of socialists (which would, in my opinion, maybe not be the best of all possible moves, but would beat the hell out of the crew of stunningly incompetent right wing flakes currently running the US). Maybe if you have Nader, Stallman et al in or advising the Congress, you'll get laws that reflect your wants and needs rather than those of the World's Corporations. Until then, it's DMCA, software patents, and worse.
B, You simply walk away from ANY PRODUCT that incorporates any form of copy protection, digital rights management, etc.
BTW, the things that set Microsoft on its road to success way back when were usable products at fair prices without copy protection. Since they have abandoned that policy, what possible reason could there be for most people to continue to buy their products? As far as I can see, the 90% of desktop users who don't actually need a crash resistant kernel, the latest version of direct-X, or one of the other small number of genuine features of XP, should have stayed with Windows 9, Only if you all -- collectively -- do things like that, are you going to convince Microsoft that there is a reason to modify its behavior ,
There is also the non-trivial problem of getting electricity from where the sun shines a lot to where people need to use it. The US grid is pretty fragile and nowhere near being up to shipping massive amounts of power across the country. During the California power crisis five years ago, they couldn't even get power between the Northern and Southern parts of the state. That problem is probably solvable using existing technology. But it might take a decade or two to solve it. I don't see anyone solving it or any leadership from either political party in the US toward getting it solved.
IMO, barring cheating and exercises designed to teach process -- what should be graded is the product not the process. I assume that as a teacher you would be at least somewhat qualified to judge product quality. How and why does the ability to judge product, qualify one to judge process? Especially in as much as not everyone agrees with you as to what an acceptable process is. Why go out of your way to do something that you possibly can't do well, when you have an alternative that you presumably can do well?
If you spend much time looking at the Wikipedia in areas that you have real expertise in, you'll probably find that much of it is quite good and is perfectly suitable as a reference. Problem is that some of it isn't and cross checking is desirable if the material is important. In an awful lot of cases, you'll find that "Real reference material" on the same topic is substantially WORSE than the Wikipedia. An example picked more or less at random. Compare the Wikipedia article on The Nanking Massacre to articles by Japanese nationalists on the one hand or Iris Chang's "Rape of Nanking" on the other. Which is the best and fairest source? Almost certainly the Wikipedia. But Chang's work when it was published in 1977 was regarded by many as a stunning piece of scholarship and the last word on the subject -- surely an acceptable primary reference by your standards.
Apparently you don't know much about PCs -- at least not older ones. The widely copied (cloned or pirated depending on your viewpoint) parallel port on the original IBM PC either was miswired or misdocumented. I can't recall which. Anyway, it didn't do interrupts. As a result, well up into the 1990s, most PCs couldn't interrupt on the parallel port. No one had all that much trouble dealing with it.
You were designing I/O devices and writing real-time I/O code before there were computer science teachers or courses? So was I. Wrote my first code in 1961 (or maybe it was 1960 -- Long time ago anyway). If your devices and code worked that's fine. But it doesn't make you right about what is nonsense and what isn't. In many cases, input intervals are predictable. If they are, you can often plan your code to be waiting for the input. The process of doing that is called "Engineering". (Not to be confused with "Software Engineering" -- an artform that bears roughly the same relationship to engineering that sculpture does to architecture). If you are able to engineer your system -- and not all systems can be engineered, polling will probably be superior and is worth considering. If you can't predict when inputs will arrive then you have to use interrupts.
As I have to keep on repeating, there are times and places where interrupts are appropriate. But your conviction that they are always superior to polling is simply wrong. To see that, take the limiting case where you must process a high data rate input stream on a computer that can barely keep up. That used to be a very common situation, and I imagine that it still is. Polling requires no context switching and can, on some hardware, be done with one instruction. An interrupt requires at least two context switches -- one to get to the processing code and one to get back. A polling loop probably will not require any register saves because it is inserted where it does not disrupt other activities. An interrupt will minimally require saving and restoring every reqister you use plus the program counter. Are seriously suggesting that you can process as fast a data stream with interrupts as the guy at the next desk can with polling? In your heart, you know you can't.
No offense, but that's just silly.
The proper criterion for judging a (non-plagiarized) paper is not "Where did the information come from?". The proper criteria are things like whether the material is complete, accurate, well reasoned, and clearly presented. With rare, and obvious, exceptions, the source is no more relevant than whether any conclusions in the paper coincide with your personal predjudices and biases.
BTW, they somewhat uncleverly walked up the wrong side of Lake Champlain. The size of the attack force was limited by the fact that they could only find one boat to ferry men to the other shore, and it wasn't very big. (Remind you of any projects you've worked on?). In the long run, it didn't matter. There were only 22 British troops garrisoned at Ticonderoga and 21 of them were asleep when Allen, Benedict Arnold and a few dozen of the Green Mountain Boys strolled in the gates.
Anyway, pitching a chair at Allen probably would have been what we call in the trade a really, really bad idea. There is an excellent chance that it would be the last chair the thrower would launch for quite some time.
I don't think that the PS/2 Auxiliary Device Port is designed to generate an interrupt when something is hot plugged into it..
BTW -- notwithstanding what your computer science teachers taught you, polling is quite efficient if loads are predictable. Polling is usually much less resource intensive than interrupts ... if the polls have a high hit rate. And it's much less subject to wierd, difficult to reproduce problems and to race conditions. Across the broad spectrum of computing, there are probably far more cases where interrupts are used when polling would work better than vice versa.
That said, 20 polls per second seems excessive for a detecting a new device on a port that is rarely used except for mice. Once every 5 seconds would seem more appropriate.
Good logic based on two faulty assumptions:
With current technology, there are a (pitiful) few applications where space can be profitably exploited. Communications, imaging, one or two others. There is not and has not been any problem attracting private investment in those areas. Point: When things evolve to the point that money can be made, there will probably be plenty of private money put up. Unfortunately, space overall is, and is likely to remain, a huge money pit. The fantasies of various taxa of space cadet notwithstanding, there is no sign this situation is going to change any time soon.
We can't get stuff to orbit cheaply. Until we can, we aren't going anywhere.
As to the second assumption. Governments sometimes push through projects that private investment is unwilling to gamble on. The best example is the Erie Canal which was built by the state of New York after the money men in New York ignored pleas of advocates for years. The Erie Canal and its less known sister the Champlain Canal were wildly successful. The Erie dropped shipped costs to what is now the American heartland by an order of magnitude and opened a path for settlement of the West. It was largely responsible for New York City becoming the predominant city in the US. Without the canal, serious settlement would have waited a couple of decades for the railroads. And the canal was also profitable -- wildly so. But is a way, the money guys were right. Many of the early 19th Century canal projects in the US were financial disasters.
Please do not take this is an endorsement of NASA. Transport NASA to New York in 1819 and those guys would wave flags, make speaches, and build a wildly over budget ditch that connected Albany to Schenectady via New York City. (For those not familiar with New York geography, the Hudson River connects Albany and NYC and is navigable every inch of the way. Schenectady is about a days walk West of Albany)
While I voted for Al Gore in 2000 and think that subsequent events have proved quite convincingly that he is far more competent than George W Bush, that does not mean that I think that either or those beauties was or is qualified to be president of the United States (or anything else).