Sure, that's a really good argument. We'll just ignore that it works equally well against any and all forms of security whatsoever.
The lowly secretary becoming a whistleblower sounds neat and all, but I suspect it's more common that he or she get their hands on some insider information and calls his or her stock broker.
Let's see, what percentage of those numbers represent people who had unlisted numbers, but were too clueless to realize they weren't getting telemarketing calls and signed up anyway? Yeah, I bet charities are really salivating at the chance to go through that list to find both the numbers on it they didn't have before.
WTF are you talking about? We've got plenty of standard rockets capable of lifting nukes. There really not particularly heavy compared to a lot of the stuff we loft. And if we need to hit it early, I'll start with the biggest chemical rocket available. Even if I have to do some development to make one even bigger, I imagine I'll be launching about when you're through "dusting off schematics". Your (theoretically) greater thrust isn't going to compete with my earlier launch date by a long, long shot.
And the Orion isn't nuclear powered in the way say, a nuclear submarine is. It's nuclear powered in the way a continous series of nuclear bombs being detonated in the atmosphere all the way up are. Obviously one would be a complete eco-freak to see any problem with that.
I don't buy it. I don't think there are any lawyers out there who think they would be making more money if only there were more laws. A fair number of laws probaly reduce the work pool for lawyers by removing ambiguities.
In your take on my analogy, you suggest asking a city planner or traffic engineer about a bridge. Both of these are people with training and experience in the question at hand, so if the question were about laws instead of bridges, they would be lawyers too. "Lawyer" encompasses a wide range of specialties just like "Engineer" does. For some reason, most people judge all lawers by one fairly small subset: "Unscrupulous Plaintifs Attorney". I would guess that former plaintifs attorneys are not particularly likely to be in Congress (as opposed to say, prosecutors).
Anyway, laws effect all segments of society. What profession would you like Congress people to come from? Would people from that background really have enough less conflict of interest to balance out not having any training in the job you're electing them to do?
In any case, I don't vote based on someones previous proffession, nor do I expect many others do. I would guess people who are interested in law are more likely to become lawyers, and theose same people are more likely to run for office, so naturally the candidate pool is biased toward lawers. I can see that you might disagree with my opinion that that is a good thing, but I don't see that it's a particularly bad thing, and it certainly doesn't look like a conspiracy.
"Any email address supplied to this database can be instantly turned into a name, address, and telephone number"
Make up your mind, are you paranoid or not? I mean, you sound a lot more paranoid than me, but I would have assumed any government agency interested in matching an email to an address wouldn't have any big problem even before this. And are you really going to claim to be a self respecting paranoid while admitting you don't understand how to use different email adresses in different contexts?
Well, I signed up for the Colorado list, which has roughly the same exceptions, and I'd say the 80% is conservative. Before signing up, I got a couple unwanted calls a day, now it's maybe one a week. Big improvement.
One of my credit card companies was willing to understand that even though we have an existing relationship, I don't want calls about exciting aditional opportunities, the other I cancelled. Other than that I don't appear to have any existing business relationships, at least with anyone intersted in using that to get around the law with a new sales pitch.
The political campaigns I don't really mind, in my experience it's one call right before election day, not even advocating a particular candidate. Just reminding me to vote and offering directions to my polling station. The call is actually from whichever party thinks they'll be helped by greater turnout in my precinct.
I've gotten a few 'pollsters' who's poll is obviously designed to determine if you warant a visit from a door-to-door salesman. That's kind of lame. But the vast majority of the remaining calls are charities. I explain that while I realize they don't have to use the do-not-call-list, the fact that they have not done so voluntarily has caused me to add them to my do-not-donate list. Not that I imagine the jerk in the call center cares.
Anyway, the total volume of unwanted calls is vastly reduced since I signed up for the list. Sufficiently that telemarketers have completely dropped off my things that annoy me list, after years of being near the top.
A disproportionate number of Congress people are lawyers? I'm horrified! Do the voters actually think it makes sense that the people making laws should have training and experience in the law? What crazy thing are we going to do next, have our bridges designed by people with degrees in structural engineering?
Damn, now I'm going to have to go find a copy. I thought this was the greatest book ever at age 12 or something, but now my recolections are a bit fuzzy, and have probably been confused by subsequently seeing the movie. That could certainly explain thinking it was a comet causing the green sky. Was the Soviet weapon established fact, or speculation by the heroes?
While I could be wrong, I still think I remember the ambiguous ending correctly:
The good guys are in a compound they've managed to fence the Trifids out of, but it's clear they'll get in soon. Our hero manages to repair and Trifid-proof a truck, which they load up with all the gas and supplies they can. She: "Must we keep running forever?" He: "We'll just have to pray that one day we can find a way to beat these things." Fires up the engine and floors it toward the wall of advancing Trifids. The End.
This scene is so firmly embedded in my psyche I'd really hate to find out I made it up. It's my first memory of a story that didn't wrap it all up with a happy good-guys-win scenario. I suppose it's possible my young brain just rebelled at whatever deux-ex-machina really happened and constructed this in it's place. I've certainly seen psych studies showing that peoples distant memories are fabulously unreliable, to the point of firmly believing in events that never were. You never think that applies to your own memories though.
Damn, you beat me to identifying it. FWIW, the trifids were murderous, intelligent plants that evolved on earth. They weren't a big deal to control though, because humans were smarter and faster, and were a common zoo exhibit (but somehow nobody ran across the salt-water thing). The meteor passed by earth, releasing some weird radiation that blinded almost everyone on earth (except the small number who for various reasons never saw the sky during a certain 24 hour period). Versus blind people, the trifids suddenly had the upper hand. All this is from my memories of the book, but I think the movie was pretty faithful, except that in the book, salt water doesn't come into it. The books science is pretty reasonable (once you buy mobile, inteligent, carniverous plants and blindness-inducing comet radiation of course) The book leaves it open whether the few sighted survivors will manage to beat back the Trifids, or if humanity will just be wiped out (which actually seems the most likely outcome). Obviously an ending that wasn't going to fly with Hollywood. The book was high quality scary sci-fi (for a kid anyway). The movie is camp though and through.
"would it not be simple enough to observe that the mean distance between 2 bodies is changing, however microscopically, and conclude that the orbiting body will collide or escape, even if we can't say when?"
No, because you don't know what else might happen before they collide or escape. A comet could crash in and disturb the orbit. Or the sun could go nova. etc.
As to the constant rate of the moon moving away, it's not. See this page: http://seds.lpl.arizona.edu/nineplanets/nin eplanet s/luna.html
Basically, tidal effects transfer energy from the earths rotation to the moons orbital speed (thus increasing its orbital distance). So the rate has been steadily decreasing for the last 5 billion years.
In the far distant future (ignoring the sun going nova possibility) the earth will have been slowed to the point where it's rotaional speed is the same as the moons orbital period and this effect will stop, so the moon won't move any farther away. The moon will hang there in geosyncronous orbit, always over the same point on the earth. Weird.
Theoretically, an orbiting body not affected by some mass other than its primary is not going to escape, and it's not going to "spiral in" unless it is under drag. In a sufficiently pure theoretical abstraction, ALL orbits are perfect.
Non-theoretically, there is drag on the planets, but it is so incredibly miniscule we can safely ignore it. (think about how many inches the moon has to move away from us to make any difference at all). So we're down to being affected by other masses. The planets (particularly the gas giants) mess with each others orbits sufficiently that the existence of the outermost planets was detected by noticing the deviations before they were observed. Add in minor disturbances from smaller masses (comets, etc.) passing through, and it becomes very difficult to exactly calculate where everything will be a very long way in the future. Note I say "exactly". None of these disturbances makes much of a big difference. It's a pretty ridiculously small chance that anything will escape short of the sun going nova (at which point probably everything will escape that doesn't get incinerated). THe chances of a collision (between planets) before the sun goes away look ridiculously small even in comparison to the ridiculously small chances of an escape. Relatively speaking, the planets are small, and the space they're moving in is just insanely vast.
"It's not as if mars or any other planet moves opposite the stars on any given night"
Of course it does. For several nights and days in a row in fact, but slowly. You would need the (somewhat) precise mesurements to tell in the course of one night. Over several days a sketchbook (or good memory) will suffice. After all, pretty much every ancient culture noticed it without telescopes or sextants.
Please understand that I was responding to the assertion that C++ is "bad". C++ is pretty near to the high end of the difficulty/power curve. It's not right for everything. The dirt-simple scripting language I spoke of grew out of our early days when everything was in C++, so you wound up needing C++ do, for example, web-authoring. Bad.
"Exactly. You have isolated the 'dangerous scary stuff' to the surgeons, and left the grunt work to the lower skill workers"
Well put. Exactly. In my experience, a few very high-end and a handful of fairly low-end developers get much more done than a whole bunch of mid-range developers.
"You have, in a sense written your own mini programming language interpreter. The only difference between your interpreter and Java/Python/... is that the latter are general purpose, and your is custom tailored."
More than in a sense. That's exactly what we've done. But our language is much further toward the bottom of the dificulty/power curve than just about anything, certainly Java.
I guess I just don't have much use for the middle of that difficulty/power curve. In my experience, if the problem you're trying to solve is too dificult for low end, then a programmer who can understand the problem and how to solve it can understand C++.
I don't agree that smart pointer represents a tradeoff. Those who want it can have it with an hours work, those who don't don't have to use it. Both types of coder are supported. (Most people will be both types at different times) Java would force everyone to use it, in order to protect those who need it but can't figure out that they need it. My smart pointer template is part of the headers I include as a matter of course, so for me, it is part of the language just as much as the STL templates. The debugger also makes no distinction as far as I can tell. The thing I think really rocks about C++ is that well written templates let you build whatever you want cleanly into the language. In Java, garbage collection is the baseline, so the small fraction of users who care about the performance penalty are just SOL.
"Please understand me, I respect the C++ tool, and know it is indeed mighty in the right hands. *BUT* I do think it is a big mistake to consider it the de-facto end-all OO language, and the only 'real' productivity tool for 'real' programmers"
I don't consider it the end-all, or only worthwhile language. I do consider it the current high-point for power and flexibility. Which comes with the price of having to know what to do with that power and flexibility.
"CPUs are simply too fast, RAM is too cheap" For a very large number of applications, those statements are just false. Computer power is one of those things where the demand expands to fill the supply. And not just because of inneficient code. When CPUs get faster, you can do things you wouldn't have considered before. The software I write is available in both web-based and desktop flavors. There are numerous things you can do with the desktop version that you can't on the web, and the only reason is that on the desktop, it's your CPU. If you want to pin it at 100% for five minutes, go ahead. On the web, it's a server used by lots of other people. That server is a lot nicer than your desktop box (and certainly puts the lie to the hardware-is-cheap theory), but I can't let you pin its CPU for more than a second or two. As CPU speed increases, the things you can do on that server will be loosened up. But the desktop will stay ahead as we add features previosly rejected because nobody wants to pin their CPU for a month. In that server environment, maximally efficient code is essential.
Java tries to hit a point somwhere in the middle of the power/diffiulty curve. As, I've said, I don't have much use for that point in any case, but in my opinion, Java misses it by trading away a lot of power and flexibility, and still winding up being fairly difficult.
"Wait till you've been working on larger projects with a moderate sized team of people in the industry for a few years"
Been there, doing that. I'm with you on the high level scripting language; that shouold be most of your development, reusing modular peices of efficient code wirtten in? You lost me with the "plain C". Well designed and written C++ is more reusable and maintainable than well designed and written C by a long shot. If it's quite well written, it's just as fast, and in a few cases, faster. Reliability is a matter of code quality and testing; I don't think language enters into it.
"YES, a well-trained, experienced C++ programmer can work wonders, BUT that's like 1% or less of the programming population!"
And I'd argue 1% of the programming population gets 99% of the programming work done. Where I work, a number of mostly-comfortable-with-HTML folks use a dirt-simple scripting language which for anything significant calls code written by a couple of well-trained, experienced (are those different?) C++ programmers. As a team, we do indeed work wonders.
"C++ made a lot of design tradeoffs (e.g. it does not automatically handle allocated space well without extreme programmer care) connected to computer technology EONs ago"
If C++ has a weakness, I'd say it's that it didn't make ANY design tradeoffs. If there are five ways to do a thing, you can do it 5 ways in C++ (or maybe 6). I would consider handling allocated space automatically, which is to say, inflexibly, the design tradeoff. How exactly is this (or any C++ feature) realted to computer technology at any particular point, much less "EONs ago"? I don't understand what you're refering to there. In any case, memory de-allocation is the thing everyone points too as the thing sure to bite you in C++ (particularly if they're a Java fan). In my experience, it never has. Not once. I can probaly count on one hand the times I've used "new" or "malloc", and I was careful those times. When I started with C++, my boss/mentor pointed out a class he had written: "SmartPointer". Calls new in its constructor, reference counts, calls delete in the last destructor, acts like a pointer otherwise. About half a page of template code. Why it's not in the STL instead of the half-assed auto_pointer I don't know. (OK, I do know: SmartPointer is marginally less efficient than auto_pointer since it has to do an extra allocation; auto_pointer is just as efficient as new/delete but IMHO is pointless because you still have to think; it's in those rabid-for-efficiency cases I've just gone ahead and used new/dellete)
C++ can be written to be as fast (or faster than) plain old C, so I don't know why you'd need plain C.
Java beats C++ on platform independence without recompile (which I don't have a need to care about, YMMV) and otherwise blows goats. OK, that's a bit harsh, sorry. But it is typically slower, and it "protects" you from doing things in ways it doesn't want you to, even if it's right for the situation. C++ "protects" me from nothing. That's OK, I can handle it. Sure, I know programmers who can't, but generally they can't handle the problem they're trying to solve in any language, so at least C++ protects me from having to work with them (for very long at least).
Gotta disagree. C++ is the current high water mark. If you told me more about why you don't think so I could tell you more about why I disagree. But I suspect why you disagree with me is that you haven't seen/used well done C++ much. I'll freely grant that badly written code in C++ is worse than badly written code in many other languages. But I won't stand for badly written code in any language, so the point is moot. Comparing only well written code, C++ delivers the goods over C. Not surprising, since it was desigend solely to address real-world problems real programers were having writing in C.
Perl? Hey, it kicks ass for a few things, but that's easy. In the C vs. C++ we're talking general-pupose programming languages with no built-in performance penalty, so Perl isn't even in the running (or trying to be, it's busy over there in the do-stupefyingly-complicated-operations-on-text-fil es-with-stupidly-terse-syntax competition).
I'll go along with most widely available. printf is great when it's all you have. Heck, I use it myself (in alert() form) to debug javascript). But if you've got say, the debugger in Visual C++? It will give you all the information printf will, but you don't have to decide what information you want before you get to that point in the code. And it can't very well give you too much information, since it doesn't give you any you don't ask for. Want to see the complete call stack for multiple threads when a particular global variable changes value? good luck w/ printf.
"someone who can't debug with print statements isn't a Real Programmer" Sure, fine. Someone who doesn't use the best tool available to them isn't a Real Programmer either.
"Another problem with a debugger is if your app is split into DLL's: how do you debug into a DLL if that is where it is choking?"
Uh, use Visual C++? All my apps are split into DLLs. When I'm stepping through something in the debugger, I neither notice nor generally care that I've crossed a DLL boundary. (Except when I'm tracking a memory leak, in which case I notice on purpose...) Most of the time I'm setting a breakpoint deep in the guts of some dll and letting the rest just run until it gets there. Seriously, what's suposed to be the problem with debugging into dlls?
Not sure what you mean by running a DLL "stand-alone". Most dlls don't stand alone. the debugger will tell you everyting passed in to a dll function, at which point it shouldn't matter what called it; ocasionally I write a stupid exe to call one function I think is a problem, but just to speed the debug cycle.
Bull. You've been programming extensively for 9 years in bad C++. Any language sucks if misused. You should evaluate it based on writing good code.
You can write almost the exact same code as you would in C, and call it C++. So it's not setting you back any. Some argue that C++ (and OOP in general) don't gain you anything. To put it bluntly, they are wrong. They generally then continue with some example that demostrates that they don't understand the point of OOP. This is partly caused by the fact that everyone you ask will give you a different definition of OOP. I woun't bother defining it, I'll just tell you what the point is: "Object Oriented Programming allows old code to call new code" (If anyone can tell me who I'm quoting, I'd apreciate it, I've forgotten) In traditional straight-procedural programming, code re-use consists largely of new code calling old code, which is a fine thing. But it's difficult to have old code call new code.
When I started my current job, I had to write (in C++) some code to do various things with geographic areas. There were two ways geographic areas could be defined (radius around a point or list of zip codes). So I went for one of your useless abstractions and made a base class "GeographicArea" that exposed functions to get the info I would need (e.g. GetArea() or ContainsPoint()) and two derived classes for the radii and the zips. I then wrote various functions that used the base class (e.g. WriteDemographicReport()).
All in all, that probably took me about the same amount of time as if I had skipped the abstraction. Actually I think it took me less, because I've found that well chosen abstractions encourage me to write clean code free of special cases, so most of the stuff I had to write I wrote once instead of twice. But for the sake of argument lets say it took me the same time.
Here's where well written C++, and OOP in general, really shines: It's now 5 years later. There are now about a dozen derived classes of GeographicArea representing different ways areas on the ground can be defined. I've not even looked at WriteDemoGraphicReport() in the intervening years, yet it works for all the new derived classes. That old code calls GetArea() and gets routed to the new code without me having to think about it.
That's what OOP is about: code re-use at a level that is not possible (without stupefying hoop-jumping) in a non-oop language.
If you want to write a progam to solve one problem, and never expand it or modify it after next week, forget OOP. But note that if your manager tells you "We just need this done by next week, we'll never use it again", He's almost certainly wrong.
In my experience there is no such thing as "quick and dirty". There is "quick because it's clean". There is "clean but takes a while, because that's the way it is". And there is "Dirty because I tried to do it quick, and hence it's going to take even longer". The last one has this tendency to hang around being sucky forever until someone throws the whole thing out and does it the clean way. In many cases, it would be very hard to do things the clean way without C++.
Practical C++ programming is understanding that most of the C++ you write will be plain old C. And it is understanding when the extra features of C++ are apropriate and how to use them well, because doing the right thing in that small fraction of the code is what is going to let you, the C++ programmer, kick the C programmers metaphorical asses a couple years down the line.
"With Windows, you... um... good question. You COULD write drivers, but I don't know how you would get started. I'm sure there are a few books on the subject, and maybe some web pages"
There are quite a few boks on the subject, and certainly some web pages. In particular MSDN comes to mind: all the docs and apis you need in one place. You don't know how you would get started, because you've never tried.
If the abiltity tinker and experiment with your OS is a driving factor, go with Linux, absolutely. (I'd also go with linux based on various other things being your driving motivation) But if you're interested in extending MS lock-in, I mean, uh, writing Windows specific software, you'll find that MS is as helpful and friendly as you could ever hope for. They understand that as mammoth as they are, it's the many third-party developers writing Windows specific code that really make the monopoly thing work. If you want to (or have to) write such code, they really do make life pretty nice for you.
In any case, I've got to agree "Ham radio operators use Linux more than the general population" is definitely a no-brainer. Why is this newsworthy? I mean it's obviously Ham-radio booster month here on Slashdot, but why exactly? Did one of the editors just get into ham radio? I don't dipute that ham radio is cool enough, or rather nerdy enough. Based on the Hams I've known it's got a bigger nerd factor than basically anything, but pace yourselves! This at-least-one-Ham-story-every-day just can't stay interesting. As evidenced by this article. Or even this post; I'm just rambling at this point, clearly. Insomnia will do that. Sigh. Why am I still awake?
That link goes to a clearly pro-gun site, and in any case, he comes up with roughly equal numbers for deaths of children by drowning and firearm accidents. Not 5 times. So you were "breezy" with more than just sources.
Not that I really expect accuracy in slashdot posts, but I was feeling argumentative, and "Informative" mods on falsehoods drive me up the wall. I actually thought the "pool-industrial complex" was kind of funny.
"The broader point was that we don't ban everything just because it is unsafe."
Absolutely, and we should not. But we should weigh the relative pros and cons.
Note that I am not disputing (at least in this thread) your right to own a gun. I do think owning a gun for purposes of self-defense is in most cases stupid. The only way it's going to be worth anything for that purpose is if you keep it loaded and accessible. In that situation, for most people, the chances of needing to use it to defend yourself, and doing so successfully, just aren't anywhere near the chances of it being used to kill someone (maybe you) in some other situation. (accidental or intentional)
If you're working the night shift at a liquor store in a high crime area, maybe the pros outweigh the cons. The gun will probably scare off someone without one after all. But guns are really easy to come by, so most assailants will probably have one, and then your chances are probably 50-50 at best since he'll have the drop on you. So if I were working retail night shift in a high crime area, I'd probably consider getting a gun, but I'd probably decide against it. (On second thought, I'd quit).
But for most of us, in everyday life? Our society is civil enough that having a gun around ready to use is a bad decision. So I think it would be a good thing if there were a lot less cheap handguns floating about. They're useless for "maintaining the security of a free state" in any case, they're only good for killing people, and they mostly kill the wrong people.
"Have you ever seen the map that showed the displacement of who voted for who in the 2000 election? The only place that gore won anything was in the big cities"
So? He got more votes than Bush. A majority of the voters voted for Gore. In any race but president, there wouldn't have been any arguing about the mess in Florida, Gore would have just won.
The electoral college gives less populous states slightly more influence than they deserve based on their population. This is stupid firstly because I can't think of any issue in recent memory that was Big States vs. Little states. It's stupid secondly because it's not nearly enough. If there was an issue that was Big State vs. Little State, and it was a big enough deal to swing the election, the Big States would just roll over the little ones with no trouble at all. The relative populations of the states have just diverged way beyond what the founders imagined. The top 11 states can elect the president all by themselves.
But while that's stupid, it's not my real problem with the electoral college. My real problem is that if a candidate gets 100% of the votes in my state, he gets 8 electoral votes. If he gets 50.001%, he gets 8 electoral votes. It seems to me that in the later case, he should get 4. Nobody has ever been able to offer me one good reason why he gets all 8.
More people voted for Gore than for Bush. Under our election rules, Bush won anyway. I say that's stupid.
"5 times more children die of backyard-pool drowning then gun accidents"
An Informative for one probably made up statistic with no source? I'll show you informative.
Best I could find was for all ages, total accidental deaths in 2000:
Drowning and submersion while in or falling into swimming-pool: 567
Firearms discharge: 776
I suppose it's probable children are a larger percent of the drownings than the gun accidents, but your 5 times number is seeming improbable. On the other hand, I don't see why child vs. adult matters. And of course, we could add in the non-accidental deaths by assault with a firearm: 10,801, which pretty much dwarfs the other numbers. Yes, I know, some of those people might well have been killed by other means if their assailant didn't have a gun. But total deaths from assault was only 16,765 (i.e. 64% of people killed by assault were shot.) I think it's safe to say some fraction of those would have survived if their assailant didn't have a gun. Probably an even more significant fraction of the 270 killed by "Legal intervention involving firearm discharge" would have survived if the deceased hadn't had a gun, because the police wouldn't have felt the need to shoot him.
More debateable is what fraction of the 16,586 killed by "intentional self-harm by firearm" would have found another way to do themselves in. Probably quite a few, but firearm was more than half the intentional self harm deaths.
In short, I think that cutting the number of firearms in the country by (for example) half would clearly save more lives than cutting the number of swimming-pools by half.
And lest I forget, source: http://www.nsc.org/lrs/statinfo/odds.htm
Slashdot Mirror
Sure, that's a really good argument. We'll just ignore that it works equally well against any and all forms of security whatsoever.
The lowly secretary becoming a whistleblower sounds neat and all, but I suspect it's more common that he or she get their hands on some insider information and calls his or her stock broker.
Let's see, what percentage of those numbers represent people who had unlisted numbers, but were too clueless to realize they weren't getting telemarketing calls and signed up anyway? Yeah, I bet charities are really salivating at the chance to go through that list to find both the numbers on it they didn't have before.
So if more computer programs were written by non-programmers, we wouldn't need so many programmers to debug them?
WTF are you talking about? We've got plenty of standard rockets capable of lifting nukes. There really not particularly heavy compared to a lot of the stuff we loft. And if we need to hit it early, I'll start with the biggest chemical rocket available. Even if I have to do some development to make one even bigger, I imagine I'll be launching about when you're through "dusting off schematics". Your (theoretically) greater thrust isn't going to compete with my earlier launch date by a long, long shot.
And the Orion isn't nuclear powered in the way say, a nuclear submarine is. It's nuclear powered in the way a continous series of nuclear bombs being detonated in the atmosphere all the way up are. Obviously one would be a complete eco-freak to see any problem with that.
I don't buy it. I don't think there are any lawyers out there who think they would be making more money if only there were more laws. A fair number of laws probaly reduce the work pool for lawyers by removing ambiguities.
In your take on my analogy, you suggest asking a city planner or traffic engineer about a bridge. Both of these are people with training and experience in the question at hand, so if the question were about laws instead of bridges, they would be lawyers too. "Lawyer" encompasses a wide range of specialties just like "Engineer" does. For some reason, most people judge all lawers by one fairly small subset: "Unscrupulous Plaintifs Attorney". I would guess that former plaintifs attorneys are not particularly likely to be in Congress (as opposed to say, prosecutors).
Anyway, laws effect all segments of society. What profession would you like Congress people to come from? Would people from that background really have enough less conflict of interest to balance out not having any training in the job you're electing them to do?
In any case, I don't vote based on someones previous proffession, nor do I expect many others do. I would guess people who are interested in law are more likely to become lawyers, and theose same people are more likely to run for office, so naturally the candidate pool is biased toward lawers. I can see that you might disagree with my opinion that that is a good thing, but I don't see that it's a particularly bad thing, and it certainly doesn't look like a conspiracy.
"Any email address supplied to this database can be instantly turned into a name, address, and telephone number"
Make up your mind, are you paranoid or not? I mean, you sound a lot more paranoid than me, but I would have assumed any government agency interested in matching an email to an address wouldn't have any big problem even before this. And are you really going to claim to be a self respecting paranoid while admitting you don't understand how to use different email adresses in different contexts?
"what are the odds that charities, pollsters and political campaigns just got a new source for their call list."
Pretty low, since their current call list, (aka "The Phone Book") already included all those numbers.
Well, I signed up for the Colorado list, which has roughly the same exceptions, and I'd say the 80% is conservative. Before signing up, I got a couple unwanted calls a day, now it's maybe one a week. Big improvement.
One of my credit card companies was willing to understand that even though we have an existing relationship, I don't want calls about exciting aditional opportunities, the other I cancelled. Other than that I don't appear to have any existing business relationships, at least with anyone intersted in using that to get around the law with a new sales pitch.
The political campaigns I don't really mind, in my experience it's one call right before election day, not even advocating a particular candidate. Just reminding me to vote and offering directions to my polling station. The call is actually from whichever party thinks they'll be helped by greater turnout in my precinct.
I've gotten a few 'pollsters' who's poll is obviously designed to determine if you warant a visit from a door-to-door salesman. That's kind of lame. But the vast majority of the remaining calls are charities. I explain that while I realize they don't have to use the do-not-call-list, the fact that they have not done so voluntarily has caused me to add them to my do-not-donate list. Not that I imagine the jerk in the call center cares.
Anyway, the total volume of unwanted calls is vastly reduced since I signed up for the list. Sufficiently that telemarketers have completely dropped off my things that annoy me list, after years of being near the top.
A disproportionate number of Congress people are lawyers? I'm horrified! Do the voters actually think it makes sense that the people making laws should have training and experience in the law? What crazy thing are we going to do next, have our bridges designed by people with degrees in structural engineering?
Damn, now I'm going to have to go find a copy. I thought this was the greatest book ever at age 12 or something, but now my recolections are a bit fuzzy, and have probably been confused by subsequently seeing the movie. That could certainly explain thinking it was a comet causing the green sky. Was the Soviet weapon established fact, or speculation by the heroes?
While I could be wrong, I still think I remember the ambiguous ending correctly:
The good guys are in a compound they've managed to fence the Trifids out of, but it's clear they'll get in soon. Our hero manages to repair and Trifid-proof a truck, which they load up with all the gas and supplies they can. She: "Must we keep running forever?" He: "We'll just have to pray that one day we can find a way to beat these things." Fires up the engine and floors it toward the wall of advancing Trifids. The End.
This scene is so firmly embedded in my psyche I'd really hate to find out I made it up. It's my first memory of a story that didn't wrap it all up with a happy good-guys-win scenario. I suppose it's possible my young brain just rebelled at whatever deux-ex-machina really happened and constructed this in it's place. I've certainly seen psych studies showing that peoples distant memories are fabulously unreliable, to the point of firmly believing in events that never were. You never think that applies to your own memories though.
Damn, you beat me to identifying it. FWIW, the trifids were murderous, intelligent plants that evolved on earth. They weren't a big deal to control though, because humans were smarter and faster, and were a common zoo exhibit (but somehow nobody ran across the salt-water thing). The meteor passed by earth, releasing some weird radiation that blinded almost everyone on earth (except the small number who for various reasons never saw the sky during a certain 24 hour period). Versus blind people, the trifids suddenly had the upper hand. All this is from my memories of the book, but I think the movie was pretty faithful, except that in the book, salt water doesn't come into it. The books science is pretty reasonable (once you buy mobile, inteligent, carniverous plants and blindness-inducing comet radiation of course) The book leaves it open whether the few sighted survivors will manage to beat back the Trifids, or if humanity will just be wiped out (which actually seems the most likely outcome). Obviously an ending that wasn't going to fly with Hollywood.
The book was high quality scary sci-fi (for a kid anyway). The movie is camp though and through.
"would it not be simple enough to observe that the mean distance between 2 bodies is changing, however microscopically, and conclude that the orbiting body will collide or escape, even if we can't say when?"
n eplanet s/luna.html
No, because you don't know what else might happen before they collide or escape. A comet could crash in and disturb the orbit. Or the sun could go nova. etc.
As to the constant rate of the moon moving away, it's not. See this page:
http://seds.lpl.arizona.edu/nineplanets/ni
Basically, tidal effects transfer energy from the earths rotation to the moons orbital speed (thus increasing its orbital distance). So the rate has been steadily decreasing for the last 5 billion years.
In the far distant future (ignoring the sun going nova possibility) the earth will have been slowed to the point where it's rotaional speed is the same as the moons orbital period and this effect will stop, so the moon won't move any farther away. The moon will hang there in geosyncronous orbit, always over the same point on the earth. Weird.
Theoretically, an orbiting body not affected by some mass other than its primary is not going to escape, and it's not going to "spiral in" unless it is under drag. In a sufficiently pure theoretical abstraction, ALL orbits are perfect.
Non-theoretically, there is drag on the planets, but it is so incredibly miniscule we can safely ignore it. (think about how many inches the moon has to move away from us to make any difference at all). So we're down to being affected by other masses. The planets (particularly the gas giants) mess with each others orbits sufficiently that the existence of the outermost planets was detected by noticing the deviations before they were observed. Add in minor disturbances from smaller masses (comets, etc.) passing through, and it becomes very difficult to exactly calculate where everything will be a very long way in the future. Note I say "exactly". None of these disturbances makes much of a big difference. It's a pretty ridiculously small chance that anything will escape short of the sun going nova (at which point probably everything will escape that doesn't get incinerated). THe chances of a collision (between planets) before the sun goes away look ridiculously small even in comparison to the ridiculously small chances of an escape. Relatively speaking, the planets are small, and the space they're moving in is just insanely vast.
"It's not as if mars or any other planet moves opposite the stars on any given night"
Of course it does. For several nights and days in a row in fact, but slowly. You would need the (somewhat) precise mesurements to tell in the course of one night. Over several days a sketchbook (or good memory) will suffice. After all, pretty much every ancient culture noticed it without telescopes or sextants.
Please understand that I was responding to the assertion that C++ is "bad". C++ is pretty near to the high end of the difficulty/power curve. It's not right for everything. The dirt-simple scripting language I spoke of grew out of our early days when everything was in C++, so you wound up needing C++ do, for example, web-authoring. Bad.
"Exactly. You have isolated the 'dangerous scary stuff' to the surgeons, and left the grunt work to the lower skill workers"
Well put. Exactly. In my experience, a few very high-end and a handful of fairly low-end developers get much more done than a whole bunch of mid-range developers.
"You have, in a sense written your own mini programming language interpreter. The only difference between your interpreter and Java/Python/... is that the latter are general purpose, and your is custom tailored."
More than in a sense. That's exactly what we've done. But our language is much further toward the bottom of the dificulty/power curve than just about anything, certainly Java.
I guess I just don't have much use for the middle of that difficulty/power curve. In my experience, if the problem you're trying to solve is too dificult for low end, then a programmer who can understand the problem and how to solve it can understand C++.
I don't agree that smart pointer represents a tradeoff. Those who want it can have it with an hours work, those who don't don't have to use it. Both types of coder are supported. (Most people will be both types at different times) Java would force everyone to use it, in order to protect those who need it but can't figure out that they need it. My smart pointer template is part of the headers I include as a matter of course, so for me, it is part of the language just as much as the STL templates. The debugger also makes no distinction as far as I can tell. The thing I think really rocks about C++ is that well written templates let you build whatever you want cleanly into the language. In Java, garbage collection is the baseline, so the small fraction of users who care about the performance penalty are just SOL.
"Please understand me, I respect the C++ tool, and know it is indeed mighty in the right hands. *BUT* I do think it is a big mistake to consider it the de-facto end-all OO language, and the only 'real' productivity tool for 'real' programmers"
I don't consider it the end-all, or only worthwhile language. I do consider it the current high-point for power and flexibility. Which comes with the price of having to know what to do with that power and flexibility.
"CPUs are simply too fast, RAM is too cheap"
For a very large number of applications, those statements are just false. Computer power is one of those things where the demand expands to fill the supply. And not just because of inneficient code. When CPUs get faster, you can do things you wouldn't have considered before. The software I write is available in both web-based and desktop flavors. There are numerous things you can do with the desktop version that you can't on the web, and the only reason is that on the desktop, it's your CPU. If you want to pin it at 100% for five minutes, go ahead. On the web, it's a server used by lots of other people. That server is a lot nicer than your desktop box (and certainly puts the lie to the hardware-is-cheap theory), but I can't let you pin its CPU for more than a second or two. As CPU speed increases, the things you can do on that server will be loosened up. But the desktop will stay ahead as we add features previosly rejected because nobody wants to pin their CPU for a month.
In that server environment, maximally efficient code is essential.
Java tries to hit a point somwhere in the middle of the power/diffiulty curve. As, I've said, I don't have much use for that point in any case, but in my opinion, Java misses it by trading away a lot of power and flexibility, and still winding up being fairly difficult.
Obviously, your mileage may vary.
"Wait till you've been working on larger projects with a moderate sized team of people in the industry for a few years"
Been there, doing that. I'm with you on the high level scripting language; that shouold be most of your development, reusing modular peices of efficient code wirtten in? You lost me with the "plain C". Well designed and written C++ is more reusable and maintainable than well designed and written C by a long shot. If it's quite well written, it's just as fast, and in a few cases, faster. Reliability is a matter of code quality and testing; I don't think language enters into it.
"YES, a well-trained, experienced C++ programmer can work wonders, BUT that's like 1% or less of the programming population!"
And I'd argue 1% of the programming population gets 99% of the programming work done. Where I work, a number of mostly-comfortable-with-HTML folks use a dirt-simple scripting language which for anything significant calls code written by a couple of well-trained, experienced (are those different?) C++ programmers. As a team, we do indeed work wonders.
"C++ made a lot of design tradeoffs (e.g. it does not automatically handle allocated space well without extreme programmer care) connected to computer technology EONs ago"
If C++ has a weakness, I'd say it's that it didn't make ANY design tradeoffs. If there are five ways to do a thing, you can do it 5 ways in C++ (or maybe 6).
I would consider handling allocated space automatically, which is to say, inflexibly, the design tradeoff.
How exactly is this (or any C++ feature) realted to computer technology at any particular point, much less "EONs ago"? I don't understand what you're refering to there.
In any case, memory de-allocation is the thing everyone points too as the thing sure to bite you in C++ (particularly if they're a Java fan). In my experience, it never has. Not once. I can probaly count on one hand the times I've used "new" or "malloc", and I was careful those times. When I started with C++, my boss/mentor pointed out a class he had written: "SmartPointer". Calls new in its constructor, reference counts, calls delete in the last destructor, acts like a pointer otherwise. About half a page of template code. Why it's not in the STL instead of the half-assed auto_pointer I don't know. (OK, I do know: SmartPointer is marginally less efficient than auto_pointer since it has to do an extra allocation; auto_pointer is just as efficient as new/delete but IMHO is pointless because you still have to think; it's in those rabid-for-efficiency cases I've just gone ahead and used new/dellete)
C++ can be written to be as fast (or faster than) plain old C, so I don't know why you'd need plain C.
Java beats C++ on platform independence without recompile (which I don't have a need to care about, YMMV) and otherwise blows goats. OK, that's a bit harsh, sorry. But it is typically slower, and it "protects" you from doing things in ways it doesn't want you to, even if it's right for the situation. C++ "protects" me from nothing. That's OK, I can handle it. Sure, I know programmers who can't, but generally they can't handle the problem they're trying to solve in any language, so at least C++ protects me from having to work with them (for very long at least).
Gotta disagree. C++ is the current high water mark. If you told me more about why you don't think so I could tell you more about why I disagree. But I suspect why you disagree with me is that you haven't seen/used well done C++ much. I'll freely grant that badly written code in C++ is worse than badly written code in many other languages. But I won't stand for badly written code in any language, so the point is moot. Comparing only well written code, C++ delivers the goods over C. Not surprising, since it was desigend solely to address real-world problems real programers were having writing in C.
Perl? Hey, it kicks ass for a few things, but that's easy. In the C vs. C++ we're talking general-pupose programming languages with no built-in performance penalty, so Perl isn't even in the running (or trying to be, it's busy over there in the do-stupefyingly-complicated-operations-on-text-fi
I'll go along with most widely available. printf is great when it's all you have. Heck, I use it myself (in alert() form) to debug javascript). But if you've got say, the debugger in Visual C++?
It will give you all the information printf will, but you don't have to decide what information you want before you get to that point in the code. And it can't very well give you too much information, since it doesn't give you any you don't ask for. Want to see the complete call stack for multiple threads when a particular global variable changes value? good luck w/ printf.
"someone who can't debug with print statements isn't a Real Programmer"
Sure, fine. Someone who doesn't use the best tool available to them isn't a Real Programmer either.
"Another problem with a debugger is if your app is split into DLL's: how do you debug into a DLL if that is where it is choking?"
Uh, use Visual C++? All my apps are split into DLLs. When I'm stepping through something in the debugger, I neither notice nor generally care that I've crossed a DLL boundary. (Except when I'm tracking a memory leak, in which case I notice on purpose...) Most of the time I'm setting a breakpoint deep in the guts of some dll and letting the rest just run until it gets there. Seriously, what's suposed to be the problem with debugging into dlls?
Not sure what you mean by running a DLL "stand-alone". Most dlls don't stand alone. the debugger will tell you everyting passed in to a dll function, at which point it shouldn't matter what called it; ocasionally I write a stupid exe to call one function I think is a problem, but just to speed the debug cycle.
Bull. You've been programming extensively for 9 years in bad C++. Any language sucks if misused. You should evaluate it based on writing good code.
You can write almost the exact same code as you would in C, and call it C++. So it's not setting you back any. Some argue that C++ (and OOP in general) don't gain you anything. To put it bluntly, they are wrong. They generally then continue with some example that demostrates that they don't understand the point of OOP. This is partly caused by the fact that everyone you ask will give you a different definition of OOP. I woun't bother defining it, I'll just tell you what the point is:
"Object Oriented Programming allows old code to call new code" (If anyone can tell me who I'm quoting, I'd apreciate it, I've forgotten)
In traditional straight-procedural programming, code re-use consists largely of new code calling old code, which is a fine thing. But it's difficult to have old code call new code.
When I started my current job, I had to write (in C++) some code to do various things with geographic areas. There were two ways geographic areas could be defined (radius around a point or list of zip codes). So I went for one of your useless abstractions and made a base class "GeographicArea" that exposed functions to get the info I would need (e.g. GetArea() or ContainsPoint()) and two derived classes for the radii and the zips. I then wrote various functions that used the base class (e.g. WriteDemographicReport()).
All in all, that probably took me about the same amount of time as if I had skipped the abstraction. Actually I think it took me less, because I've found that well chosen abstractions encourage me to write clean code free of special cases, so most of the stuff I had to write I wrote once instead of twice. But for the sake of argument lets say it took me the same time.
Here's where well written C++, and OOP in general, really shines:
It's now 5 years later. There are now about a dozen derived classes of GeographicArea representing different ways areas on the ground can be defined. I've not even looked at WriteDemoGraphicReport() in the intervening years, yet it works for all the new derived classes. That old code calls GetArea() and gets routed to the new code without me having to think about it.
That's what OOP is about: code re-use at a level that is not possible (without stupefying hoop-jumping) in a non-oop language.
If you want to write a progam to solve one problem, and never expand it or modify it after next week, forget OOP. But note that if your manager tells you "We just need this done by next week, we'll never use it again", He's almost certainly wrong.
In my experience there is no such thing as "quick and dirty". There is "quick because it's clean". There is "clean but takes a while, because that's the way it is". And there is "Dirty because I tried to do it quick, and hence it's going to take even longer". The last one has this tendency to hang around being sucky forever until someone throws the whole thing out and does it the clean way. In many cases, it would be very hard to do things the clean way without C++.
Practical C++ programming is understanding that most of the C++ you write will be plain old C. And it is understanding when the extra features of C++ are apropriate and how to use them well, because doing the right thing in that small fraction of the code is what is going to let you, the C++ programmer, kick the C programmers metaphorical asses a couple years down the line.
"With Windows, you... um... good question. You COULD write drivers, but I don't know how you would get started. I'm sure there are a few books on the subject, and maybe some web pages"
There are quite a few boks on the subject, and certainly some web pages. In particular MSDN comes to mind: all the docs and apis you need in one place. You don't know how you would get started, because you've never tried.
If the abiltity tinker and experiment with your OS is a driving factor, go with Linux, absolutely. (I'd also go with linux based on various other things being your driving motivation) But if you're interested in extending MS lock-in, I mean, uh, writing Windows specific software, you'll find that MS is as helpful and friendly as you could ever hope for. They understand that as mammoth as they are, it's the many third-party developers writing Windows specific code that really make the monopoly thing work. If you want to (or have to) write such code, they really do make life pretty nice for you.
In any case, I've got to agree "Ham radio operators use Linux more than the general population" is definitely a no-brainer. Why is this newsworthy? I mean it's obviously Ham-radio booster month here on Slashdot, but why exactly? Did one of the editors just get into ham radio? I don't dipute that ham radio is cool enough, or rather nerdy enough. Based on the Hams I've known it's got a bigger nerd factor than basically anything, but pace yourselves! This at-least-one-Ham-story-every-day just can't stay interesting. As evidenced by this article. Or even this post; I'm just rambling at this point, clearly. Insomnia will do that. Sigh. Why am I still awake?
That link goes to a clearly pro-gun site, and in any case, he comes up with roughly equal numbers for deaths of children by drowning and firearm accidents. Not 5 times. So you were "breezy" with more than just sources.
Not that I really expect accuracy in slashdot posts, but I was feeling argumentative, and "Informative" mods on falsehoods drive me up the wall. I actually thought the "pool-industrial complex" was kind of funny.
"The broader point was that we don't ban everything just because it is unsafe."
Absolutely, and we should not. But we should weigh the relative pros and cons.
Note that I am not disputing (at least in this thread) your right to own a gun. I do think owning a gun for purposes of self-defense is in most cases stupid. The only way it's going to be worth anything for that purpose is if you keep it loaded and accessible. In that situation, for most people, the chances of needing to use it to defend yourself, and doing so successfully, just aren't anywhere near the chances of it being used to kill someone (maybe you) in some other situation. (accidental or intentional)
If you're working the night shift at a liquor store in a high crime area, maybe the pros outweigh the cons. The gun will probably scare off someone without one after all. But guns are really easy to come by, so most assailants will probably have one, and then your chances are probably 50-50 at best since he'll have the drop on you. So if I were working retail night shift in a high crime area, I'd probably consider getting a gun, but I'd probably decide against it. (On second thought, I'd quit).
But for most of us, in everyday life? Our society is civil enough that having a gun around ready to use is a bad decision. So I think it would be a good thing if there were a lot less cheap handguns floating about. They're useless for "maintaining the security of a free state" in any case, they're only good for killing people, and they mostly kill the wrong people.
"Have you ever seen the map that showed the displacement of who voted for who in the 2000 election? The only place that gore won anything was in the big cities"
So? He got more votes than Bush. A majority of the voters voted for Gore. In any race but president, there wouldn't have been any arguing about the mess in Florida, Gore would have just won.
The electoral college gives less populous states slightly more influence than they deserve based on their population. This is stupid firstly because I can't think of any issue in recent memory that was Big States vs. Little states. It's stupid secondly because it's not nearly enough. If there was an issue that was Big State vs. Little State, and it was a big enough deal to swing the election, the Big States would just roll over the little ones with no trouble at all. The relative populations of the states have just diverged way beyond what the founders imagined. The top 11 states can elect the president all by themselves.
But while that's stupid, it's not my real problem with the electoral college. My real problem is that if a candidate gets 100% of the votes in my state, he gets 8 electoral votes. If he gets 50.001%, he gets 8 electoral votes. It seems to me that in the later case, he should get 4. Nobody has ever been able to offer me one good reason why he gets all 8.
More people voted for Gore than for Bush. Under our election rules, Bush won anyway. I say that's stupid.
"5 times more children die of backyard-pool drowning then gun accidents"
An Informative for one probably made up statistic with no source? I'll show you informative.
Best I could find was for all ages, total accidental deaths in 2000:
Drowning and submersion while in or falling into swimming-pool: 567
Firearms discharge: 776
I suppose it's probable children are a larger percent of the drownings than the gun accidents, but your 5 times number is seeming improbable. On the other hand, I don't see why child vs. adult matters. And of course, we could add in the non-accidental deaths by assault with a firearm: 10,801, which pretty much dwarfs the other numbers. Yes, I know, some of those people might well have been killed by other means if their assailant didn't have a gun. But total deaths from assault was only 16,765 (i.e. 64% of people killed by assault were shot.) I think it's safe to say some fraction of those would have survived if their assailant didn't have a gun. Probably an even more significant fraction of the 270 killed by "Legal intervention involving firearm discharge" would have survived if the deceased hadn't had a gun, because the police wouldn't have felt the need to shoot him.
More debateable is what fraction of the 16,586
killed by "intentional self-harm by firearm" would have found another way to do themselves in. Probably quite a few, but firearm was more than half the intentional self harm deaths.
In short, I think that cutting the number of firearms in the country by (for example) half would clearly save more lives than cutting the number of swimming-pools by half.
And lest I forget, source:
http://www.nsc.org/lrs/statinfo/odds.htm