Thanks, I'm just a dumb scientific simulation and real-time/embedded processor guy.
But just thinking about how it might be done as a general case, it looks like it would be very prone to circular logic or unintentional recursion. Clearly you have to have a list of the memory in use somewhere, but then you have to clean up that list, too, and it's not clear where it would end. I can think of explicit structures that partition it so when you leave a process it kills everything, but not an obvious general way.
I know I am profoundly ignorant and as thus should be modded into oblivion, but I don't do any programming where the memory allocation is variable. Is there a good reason are people really relying on hidden mechanisms to manage it? I would think it would be a lot more robust to keep track of allocation and deallocation explicitly, add when you need, and delete when you don't need, and not count on some generic mechanism. I know it happens, I see the memory leaks, but it would seem eminently avoidable.
Jimmy Wales in an SP and in dire need of auditing while holding on to tuna cans and a voltmeter. Or has too many thetans, or needs a spaceflight in a DC-8. Or something.
Of course. I have a Trimline phone at my house, and have since it was legal to own a phone in about 1984 or so. Mine is touch-tone, not rotary. And at work we still have model 2500s all over the place.
Say what you want about them, but when they were ATT property they certainly highly motivated to build them to last. The only thing left after a nuclear war will be Twinkies, cockroaches, and Model 500 phones.
That seems petty, but you are absolutely right. Whenever this crap shows up on a site, I just move on. I don't want to be manipulated into watching a damn commercial.
Of course it requires attribution; it's used to create original data. You shouldn't use BLASTN or CLUSTALW without citing their authors, why would it be any different for something like this? As has been mentioned numerous times already, W|A is not merely a search engine. It's a set of algorithms for manipulating the data that you specify.
And how pathetic/dishonest a scientist or professional would you have to be not to want to attribute it? Sources of information should be cited and experimental results should be verifiable.
So, for example, I get the series expansion for some function I type in. I have to credit Wolfram Alpha, apparently. Same situation, but I write it out on a pad of paper with a pencil. I have to credit Boise-Cascade paper company, and Faber-Castell for making the pencil? Suppose I take the square root of 5 with a slide rule - I attribute it to Aristo corporation? I don't see any significant difference.
How many non-isomorphic labelled trees are there with 4 vertices"
Fail.
I've tried a few other relevant, factual questions and it just falls flat over, not even able to try and answer them.
I'm sure it does do a great job of making information computable, the problem is it's unable to gather the information in the first place.
Ironically, Google, that doesn't claim to make information computable manage to provide answers for all these questions within it's first page, often as the first hit. Sure it may not be presented in a standardised format, but data that needs to be parsed is certainly more computable than data that simply can't be provided at all.
I can see what Wolfram was trying to do, but why did he have to couple it with immense hype that it's as important as Google? Why has he been going on and on about it to the media when it struggles to even do what it's supposed to absolutely excel at? I think they could've at least saved face if they'd stopped being so cocky about it and released it with a little less hype and fanfair and let it improve and become more useful and hence more greatly adopted over time.
I also tried various approaches to things it *should* be good at, but once again, not very impressive. I first tried "transfer function zero order hold" and variations on that. I expected to get something like "(1-e^-st)/s" and some words or a derivation. Should be right up it's alley, but no, it just failed, no results at all. I typed in "(1-e^-st)/s" and got a series expansion of that, several graphs of debatable accuracy of value, but nothing like "this is the equation of a zero-order hold" or even a question about s, "do you mean s=jw" or anything like that. I don't need the series expansion and I certainly wouldn't trust Mathematica to do it if I did. I still have a pencil and paper. Maybe there's something I was doing wrong, but it didn't give me results I would have expected.
On the topic of the immense hype, uh, duh, it's Wolfram, legend in his own mind and self-declared smartest man in the world. I am sure if he reads this his first reaction will be that "you guys are too stupid to grasp the brilliance, I am casting pearls before swine, I'm going to demonstrate the unified field theory with cellular automata". Insufferable even from his press releases - which I might add is common among quasi-geniuses. I have worked with some of the guys who *invented* most of the ideas behind satellite design, true geniuses whose names will never be widely known outside a few buildings at Lockheed Sunnyvale due to the nature of their work. One thing in common - the true geniuses are a lot like Feynman, personable, can explain and are willing to explain exactly why it works to anyone. The wannabe geniuses are like Wolfram seems to be - insufferably arrogant pains in the ass. Of course I only know Wolfram from his press releases, so I am making an unfair analogy or extrapolation from past experience, but I did read and understand, to the extent necessary, his book.
Anybody who has used Wolfram's products, such as Mathematica, for more than a few versions, knows that they don't have, how shall I say this? a very enlightened view of the relationship between the party that sells a product and the party that buys that product.
In fact, their user agreements have always been among the very worst in the software industry, that is, if you happen to believe that the consumer has any rights at all beyond the right to give money to the vendor.
They've always been pretty hostile toward their customers.
Not exactly the best in terms of the *users* either. The Mathematica syntax, for some things, just makes you want to cry it's so unwieldy. If you used it all day long every day, maybe the compact way you can do things might be better, but if you aren't using it all the time its painful.
Didn't come out in many consumer apps. I programmed for a system using bubble memory, and we had several GRiDs using bubble memory. It was very useful on satellite flight computers - there's still nothing equivalent, some programs still use metal tape. But it also had some problem, not the least of which was that it wasn't that fast - not fast at all, as a matter of fact, an had all sorts of duty cycle limitations that would have greatly limits it's applications.
I don't see the big deal here. People were always living on the edge of starvation. Why would anything be off the menu? The existence of kuru http://en.wikipedia.org/wiki/Kuru_(disease) certainly suggests that it was not at all unusual. Particularly when it was likely a case of simply seeing the neatherthals as another animal.
Not following them on that one, and they have the chronology completely wrong. Jobs, in particular, knew the Lisa was DOA and knew that the Mac was the way of the future for the company, and pulled people off it all the time to work on the Mac. They are right, in that the Lisa was a very nice machine (I wanted to get my father one to replace his typewriter a few years ago - he needed and wanted no more - instead he wound up with a $299 Officemax Dell shitbox that still barely functions from day to day) but I think it certainly doesn't deserve a Top 10 list. It wasn't a big enough deal to matter. I would have put the Newton on there before the Lisa.
Shaun Nichols: We're no doubt going to catch some flack for this one, but deep down even the hard-core evangelists will agree that Ubuntu has thus far been something of a disappointment. While Linux has definitely caught on in the enterprise server and database market, the open-source OS has never really been able to move into the greater market.
I don't know if I'm just easily offended or a fanboy, but I stopped reading the article at that point.
The question is, are they wrong? Ubuntu really has remained for Linux hobbyists. Maybe it shouldn't be that way, but it is, for the most part.
I disagree. I think you can combine "2.0" and "paradigm" to instantiate their synergy to create a diverse empowerment of all stakeholders. As long as you don't brick society in the process. Call it "English 2.0".
Possible - yes,. Practical, not at all. The telescope and ISS orbits have much different orbital inclinations. A quick back-of the envelope calculation suggests you need to adjust the Hubble's speed by something like 11,000 feet/second to get it into the same orbit with the ISS. That means a HUGE rocket with tremendous propulsion capability. The shuttle is nowhere near capable of that, and anything you did build would have to be gigantic and filled with fuel to be able to do it. The translunar boost stage of the Orion would be able to manage it, but that's a very long way off. So, no, that's not going to happen.
Even in pure functional languages like Haskell, the functional-programming dream of automatic parallelization is nowhere near here yet; in theory the compiler could just run a bunch of thunks of code in parallel, or speculatively, or whatever it wants, but in practice the overhead of figuring out which are worth splitting up has doomed all the efforts so far.
Forgive my profound ignorance on the topic, but am I to understand that the determination of which can run in parallel is done at run-time (and thus requires overhead to implement it)?! In my extremely limited experience, we determined which items could be parallel ahead of time and split them manually. I would have presumed that any automatic system would do the same as it was compiled and/or linked. But on-the-fly? I can easily see how that might have some issues.
Slashdot Mirror
Thanks, I'm just a dumb scientific simulation and real-time/embedded processor guy.
But just thinking about how it might be done as a general case, it looks like it would be very prone to circular logic or unintentional recursion. Clearly you have to have a list of the memory in use somewhere, but then you have to clean up that list, too, and it's not clear where it would end. I can think of explicit structures that partition it so when you leave a process it kills everything, but not an obvious general way.
Brett
I know I am profoundly ignorant and as thus should be modded into oblivion, but I don't do any programming where the memory allocation is variable. Is there a good reason are people really relying on hidden mechanisms to manage it? I would think it would be a lot more robust to keep track of allocation and deallocation explicitly, add when you need, and delete when you don't need, and not count on some generic mechanism. I know it happens, I see the memory leaks, but it would seem eminently avoidable.
Brett
Oh, he probably would have had no problem with that.
Jimmy Wales in an SP and in dire need of auditing while holding on to tuna cans and a voltmeter. Or has too many thetans, or needs a spaceflight in a DC-8. Or something.
Brett
Of course. I have a Trimline phone at my house, and have since it was legal to own a phone in about 1984 or so. Mine is touch-tone, not rotary. And at work we still have model 2500s all over the place.
Say what you want about them, but when they were ATT property they certainly highly motivated to build them to last. The only thing left after a nuclear war will be Twinkies, cockroaches, and Model 500 phones.
Brett
That seems petty, but you are absolutely right. Whenever this crap shows up on a site, I just move on. I don't want to be manipulated into watching a damn commercial.
Brett
Exactly...
So, for example, I get the series expansion for some function I type in. I have to credit Wolfram Alpha, apparently. Same situation, but I write it out on a pad of paper with a pencil. I have to credit Boise-Cascade paper company, and Faber-Castell for making the pencil? Suppose I take the square root of 5 with a slide rule - I attribute it to Aristo corporation? I don't see any significant difference.
Brett
I also tried various approaches to things it *should* be good at, but once again, not very impressive. I first tried "transfer function zero order hold" and variations on that. I expected to get something like "(1-e^-st)/s" and some words or a derivation. Should be right up it's alley, but no, it just failed, no results at all. I typed in "(1-e^-st)/s" and got a series expansion of that, several graphs of debatable accuracy of value, but nothing like "this is the equation of a zero-order hold" or even a question about s, "do you mean s=jw" or anything like that. I don't need the series expansion and I certainly wouldn't trust Mathematica to do it if I did. I still have a pencil and paper. Maybe there's something I was doing wrong, but it didn't give me results I would have expected.
On the topic of the immense hype, uh, duh, it's Wolfram, legend in his own mind and self-declared smartest man in the world. I am sure if he reads this his first reaction will be that "you guys are too stupid to grasp the brilliance, I am casting pearls before swine, I'm going to demonstrate the unified field theory with cellular automata". Insufferable even from his press releases - which I might add is common among quasi-geniuses. I have worked with some of the guys who *invented* most of the ideas behind satellite design, true geniuses whose names will never be widely known outside a few buildings at Lockheed Sunnyvale due to the nature of their work. One thing in common - the true geniuses are a lot like Feynman, personable, can explain and are willing to explain exactly why it works to anyone. The wannabe geniuses are like Wolfram seems to be - insufferably arrogant pains in the ass. Of course I only know Wolfram from his press releases, so I am making an unfair analogy or extrapolation from past experience, but I did read and understand, to the extent necessary, his book.
Brett
"W, H, Y, question mark! " WHY?"
Obscure reference noted and appreciated. I watched that one just the other day, one of the few really dated episodes.
Not exactly the best in terms of the *users* either. The Mathematica syntax, for some things, just makes you want to cry it's so unwieldy. If you used it all day long every day, maybe the compact way you can do things might be better, but if you aren't using it all the time its painful.
Brett
Didn't come out in many consumer apps. I programmed for a system using bubble memory, and we had several GRiDs using bubble memory. It was very useful on satellite flight computers - there's still nothing equivalent, some programs still use metal tape. But it also had some problem, not the least of which was that it wasn't that fast - not fast at all, as a matter of fact, an had all sorts of duty cycle limitations that would have greatly limits it's applications.
Brett
I don't see the big deal here. People were always living on the edge of starvation. Why would anything be off the menu? The existence of kuru http://en.wikipedia.org/wiki/Kuru_(disease) certainly suggests that it was not at all unusual. Particularly when it was likely a case of simply seeing the neatherthals as another animal.
Brett
That's true as well. Technology-wise its essentially of the same concept as the Mac turned out to be.
I agree 100%, modding that flamebait was ridiculous, I can't even imagine what someone could be thinking....
I agree that it's an arguable point. But it's not so completely off the scale to invalidate it to the point that it shouldn't be considered.
Brett
Not following them on that one, and they have the chronology completely wrong. Jobs, in particular, knew the Lisa was DOA and knew that the Mac was the way of the future for the company, and pulled people off it all the time to work on the Mac. They are right, in that the Lisa was a very nice machine (I wanted to get my father one to replace his typewriter a few years ago - he needed and wanted no more - instead he wound up with a $299 Officemax Dell shitbox that still barely functions from day to day) but I think it certainly doesn't deserve a Top 10 list. It wasn't a big enough deal to matter. I would have put the Newton on there before the Lisa.
Brett
The question is, are they wrong? Ubuntu really has remained for Linux hobbyists. Maybe it shouldn't be that way, but it is, for the most part.
Brett
Are you suggesting I am recycling comments from a prior existence?
I, for one, welcome our new large-eyed, nocturnal, gregarious and arboreal overlords!
I disagree. I think you can combine "2.0" and "paradigm" to instantiate their synergy to create a diverse empowerment of all stakeholders. As long as you don't brick society in the process. Call it "English 2.0".
Brett
Possible - yes,. Practical, not at all. The telescope and ISS orbits have much different orbital inclinations. A quick back-of the envelope calculation suggests you need to adjust the Hubble's speed by something like 11,000 feet/second to get it into the same orbit with the ISS. That means a HUGE rocket with tremendous propulsion capability. The shuttle is nowhere near capable of that, and anything you did build would have to be gigantic and filled with fuel to be able to do it. The translunar boost stage of the Orion would be able to manage it, but that's a very long way off. So, no, that's not going to happen.
Brett
Sales of Disco Records are up 400% for the 2-year period ending December 1979!
Disco Stu brings the facts to YOU!
Forgive my profound ignorance on the topic, but am I to understand that the determination of which can run in parallel is done at run-time (and thus requires overhead to implement it)?! In my extremely limited experience, we determined which items could be parallel ahead of time and split them manually. I would have presumed that any automatic system would do the same as it was compiled and/or linked. But on-the-fly? I can easily see how that might have some issues.
Brett
We are getting some very educational examples of it right now, in fact.
Brett