The poster implies we should all worry because half of the studies say it's a health risk...
But by that same logic none of us should worry because half of the studies say there is no damage.
So, if half of all my observations when searching for elephants show no elephants, but half of them do show elephants, you're saying I can conclude from this that elephants don't exist?
I forgot about the tutorial I promised. It doesn't take checking more than a few sources to find significant differences in the account of the Titanic's sinking. That's not uncommon, and means you have to do some research.
It's quite possible, on something like a historical point of detail, that Googling casually won't find you any answers. However, it will find you many sources, which have references, which you can track down.
In this particular case, the information you're looking for is actually more like informed speculation & analysis, since the available eyewitness reports are unreliable and inconsistent. So, applying epistemological principles, you might say to yourself "how can I verify whether the few eyewitness reports about the breakup of the ship make sense?" Some answers to that are likely to be found in a technical analysis of the sinking, so you look for those.
You can also ask yourself questions like "why didn't the stern make a huge wave which would have swamped the lifeboats?" Inconsistencies in your working hypothesis are useful for drilling down towards a more accurate model of the information you're looking for. You don't need to know in advance what you're looking for, but you need to know how to recognize when you don't know something.
You test the information you find against your current model(s), and there can be give and take on both sides, i.e. you might use a model to provisionally reject certain information.
There's no predetermined way of knowing when you've reached a final conclusion. Ultimately, it comes down to how much work you want to put into it, how much information is available, etc.
When you arrive at a final conclusion, you might even find that it doesn't match any single information source out there. Who's right, you or they? It's difficult to say. If you were intent on answering that question, you'd need to examine the processes used to arrive at other accounts, if possible. A less rigorous process is likely to arrive at a less reliable answer.
As an example, when Brian Williams reported on NBC news the other night that the lead judge in Saddam Hussein's tribunal had been killed, they had apparently received the information from multiple US officials, who in turn had received the information from government sources in Baghdad. Turns out they were wrong, it wasn't the lead judge. But NBC news reported the information as "confirmed", apparently based on having spoken to multiple US governement sources. Obviously, their grasp of these issues is rather limited -- even a superficial analysis would indicate that multiple sources in the US government might match merely because they all got their information from the same place. They made a mistake in reporting the identity of the victim as "confirmed", which they could have avoided if they had applied proper procedures, including asking how reliable their sources are, and whether their sources might be contaminated by common factors. If you don't apply that level of diligence to gathering information, you have to accept that the quality of your information will be lower.
Well, your origial post was rather concise, which made it tough to divine what was behind it. When it comes to authority, I think you've misunderstood what Google is. It's an index of the WWW, not an encyclopedia. Besides, encyclopedias, newspapers, TV news & documentaries etc. get things wrong too, sometimes spectacularly, and more often than you might think. To take the Titanic as an example, problems with the history of its sinking apparently go back to the original press coverage at the time, such as the way in which Hearst's media spun the story. Given that, it's hard to see how the problem you're describing has anything to do with Google, or even with the modern-day presence of blogs.
If you want ultimate authority, there are a number of gods I can introduce you to, although you have to agree to believe unquestioningly (have "faith") in the authority of whichever one you pick. I'll note that if you picked Google as your god in this sense, you wouldn't go as far wrong as with some of the other choices out there. Short of that, google for epistemology, and you'll find that the problem you're concerned about isn't going to be solved by any algorithm, ever.
That's right, blame the tools. It couldn't possibly be you. Tell me what you want to find and I'll give you a short tutorial on how to find it. Google's not an AI, you know.
Google recently went public. If you've drunk sufficient kool-aid, this makes them the last best hope of the tech industry. Imagine the the '90s Internet bubble all focused on one company.
Let's just hope Google doesn't decide to exploit this laser-like market focus by spinning off a host of baby Googles: then you'll be able to buy a GoogleBox PC running a browser-based GoogleOS and make phone calls over your Googlenet connection, using the Googlephone service, and you'll look up phone numbers using Whoogle (for any Google IP lawyers reading this, call me for a license on that last one).
Well, I'm an expert now, partly as a result of that research and subsequent work which I've done. And I'm talking about my preferred mode of study and research now. I also know many people in both academia and the commercial world who feel the same way - people for whom resources like Google, CiteSeer, ArXiv, and the electronic availability of papers, journals and books have meant a revolutionary advance in their ability to do serious research.
A key underlying assumption of Gorman's original argument was the assertion that "not many would choose to stare at a screen long enough" to read entire books. Unfortunately, that merely reflects Gorman's age and prejudices more than anything else (more on this below). Why should Gorman's Luddism inhibit those who are making knowledge available in useful formats, or those who find such formats useful? It shouldn't, and it won't. Gorman is basically saying "I don't like this personally, therefore it is bad for everyone". OTOH, many of the people who he is supposed to serve are voting with their feet (or fingers) and proving Gorman wrong. Unless he can produce arguments of more substance, then as I have said elsewhere, he deserves to be ignored.
Re the question of reading books on screens, Gorman makes a mistake common to those who argue against the deployment of technology -- he looks at the current and past situation to find reasons not to do it, rather than considering how things are changing and might change in future. I read books in various portable electronic formats, including on tablet PCs and PDAs, and current development seems to indicate that in future, we will quite likely be able to read on some kind of flexible, reflectively lit media, much like a sheet of paper. But should we wait until the perfect technological alternative to books is available before we begin digitizing the world's book collection? Obviously not, for many reasons, not least of which is that there are many ways in which that information could be useful in digital form, long before we reach some kind of electronic book nirvana.
It's not surprising that the people involved in the enterprise of digitizing a substantial fraction of humanity's books are enthusiastic about it. Seen in this context, it's clear that Gorman is primarily objecting to a shifting of the library limelight away from people such as himself.
OTOH, my girlfriend has severe dyslexia and likes audio sources. She depends heavily on audiobooks. The disability argument ultimately boils down to needing to make info available in multiple formats, not that text is better.
Goggle and Amazon have both made these sorts of claims for their projects.
Gorman is still using the "God's Mind" hype as a straw man to attack. Does he think that they mean those claims literally? Ironically, he seems to need a lesson in reading comprehension.
Actually the blog people are making the argument that quantity if more important that quality and that you don't need specialized skills.
Careful, you're doing the same sort of dangerous generalizing that Gorman is doing. The fact that most individual bloggers might not express their attitude to this issue in the most cogent, concise, and perfectly correct way isn't particularly relevant. Gorman has the same problem -- he's all over the map, attacking straw men and missing the real points.
One important point Gorman is missing is that Google's Pagerank system will have an amazing synergy with Google Scholar - already, you can use Google to easily identify works you might be interested in. All Google Scholar will do will make those works available online. Gorman is missing the big picture.
Further, it's not just about Google, but about the aggregation and interconnection of online resources, including digital libraries, specialist blogs and other online communities, all of which Gorman entirely ignores.
Gorman is essentially looking at a large ecosystem, but focusing only on particular parts of it -- as though he looked at the Earth and decided that most insects were useless and dangerous - ants get into people's picnic food, bees sting you. But in conjunction with the rest of the ecosystem, both end up creating something of significant value.
P.S. on the subject of the space computers take up in libraries, I would think wifi-enabled pads effectively solve that problem, taking up little more space than a book. The problems with that are just cost, damage, and theft, but certainly not space.
Re freedom of speech, I was referring to how that underlies the Internet, which is first and foremost a communication system. Physical libraries don't support the same kind of peer-to-peer communication. Gorman is apparently unused to communicating in such a wide-open forum, and is confusing all sorts of different issues, generalizing incorrectly, and responding to the wrong points.
To answer your questions, I was not an expert in the field. In this particular case, I was not reading many whole books, but used some quite extensively. However, if those books had been available online, I'd have been quite happy to do that. One can level a valid critique against online sources in that not every printed book is online, yet. But that's exactly what various digital library projects, including Google Scholar, are aiming at.
I appreciate a reasoned reply from a librarian to remind me that I was responding, in effect, to a troll, i.e. Gorman.
I'm a small rural library duhrector, and I like to think that my little library can satisfy a good majority of the needs of my patrons.
No question. However, part of where I was coming from is that I find that my reading needs are not usually served by sources that can "satisfy a good majority" of anything. Google, combined with various other online resources, is an exception, which works much better for me - even where a particular resource doesn't exist online, its existence can usually be identified online, unless one is operating in a particularly arcane field.
An AC respondent to my post suggested that professionally-maintained library catalogs (i.e. the index, not the material it represents) could not be duplicated by Google. However, my experience is the opposite. The catalog is the least interesting thing about physical libraries, to me. Part of the reason for that is that some library-like cataloging services are being done online, with services like CiteSeer (in my field), ArXiv, and various digital libraries. However, these all involve more or less complete decoupling from physical works. There are also collaborative catalogs, like CiteULike, and online communities which specialize in particular subject areas, which can be invaluable for finding out about relevant and interesting work.
The only remaining way in which I would find most physical libraries of use is if it happened to have a copy of a physical publication that I wanted to have in my hands. The reality, though, is that I rarely visit my local small-town library, even though it's within a short walking distance from my home, and I walk past it all the time. My occasional visits (perhaps three in the last five years) are never particularly fruitful. I don't think it's a particularly bad library, as small-town libraries go -- I live in a fairly wealthy town & county, so it's reasonably well-funded -- it's just not very relevant to me.
If I were the only kind of client that libraries had, then the solution would be obvious: move to an e-commerce model in which books are shipped, Netflix-DVD-rental-style, from wherever they are located, directly to people who order them online. But the round-trip cost of shipping books could approach or exceed the cost of purchasing in many cases - particularly since when you purchase, shipping is often "free", which would be difficult in the library model. (As I write this, I realize that I don't know what shipping/mailing options libraries might offer these days - perhaps I should check that out;)
BTW, I'm all in favor of the idea of shared public spaces -- community centers, theaters, etc. -- and libraries do serve that sort of purpose as well. However, look at what's happened with commercial spaces, e.g. their aggregation in the form of malls, which happens for a variety of economic and even social reasons. I think that most tax- and charitably-funded spaces have fallen behind in exploiting those kinds of overhead-sharing, people-concentrating trends. From my admittedly non-target-market perspective, it's not clear that most libraries in future ought really to be standalone buildings, as opposed to being part of some larger civic space.
I wish AC's wouldn't post thought-out replies (well, Gorman-class thought out, anyway). I'll answer anyway.
I bet that you used the catalogs in the NY Public Library during your research there. Those are maintained by professionals to help others researching information. Google (or any computer based system) can just not provide that service (yet?):
But that's precisely the point -- Google, along with other online resources, does a better job of this than any physical library ever could (without using the same online resources). Take a look at CiteSeer, ArXiv, CiteULike, etc. Then there are various "digital libraries", usually available by subscription, from organizations like ACM and IEEE (I'm focused on the computing field, which tends to be a bit ahead in this area, but that's just a question of time). Many of these services require people to maintain them, but the point just is that to a great degree, they obviate physical libraries.
Yes, you have restrictions in libraries: You are asked to behave in such a way that you do not damage the information stored there (no eating) and that you can get on well with the others working there (be quiet). Is not harassing the people around you and not destroying public properties to much to ask of you?
It doesn't matter that libraries have justifications for their rules, the point is remote electronic access to information provides a benefit - that you don't have to worry about that sort of thing.
Replacing expertice with passion -- something that seems to happen in blogs a lot -- is not helping anyone: In fact it hurts as the passion might blind others about the truth in the report.
This generalizes from the lowest common denominator of blogs in the same way Gorman does, and comes to an equally wrong conclusion. I participate in a blog that discusses research papers in my field. We have a passion and knowledge of the subject which no librarian could be expected to match.
Opinions are never irrelevant and no opinion should be ignored.
Nonsense, people ignore opinions all the time. There are loonies who believe aliens are visiting the earth, and most other people ignore them except as a source of entertainment. Gorman is free to ignore Google and bloggers, but he chose otherwise. However, Google and the bloggers would be quite right to ignore him.
As a society we could wonders on the library front for a fraction of the cost of projects like Google's; this is a point that no one questions. The real issue is what is the relative value of libraries as contrasted with digital information repositories.
Agreed. But we have thousands of years of experience with physical libraries, so we already know the answer to this question.
You have to live near a major urban center to have access to a decent physical library, if your needs go at all beyond the average. I once flew thousands of miles and spent weeks in NYC so that I could research a project at the NY Public Library, using materials I couldn't get anywhere near my home. This wasn't just a single book or journal that I could have had mailed to me -- I needed to track down cross-references, comb through original sources. Now I can get a signficant fraction of that material online, a fraction which increases every day.
That may be an extreme case, but it's just a scaled-up version of what every library-goer has to face -- you physically have to travel to the library, you have to deal with its constraints (quiet! don't eat in here! don't talk too loud!...), you often have to go through bureacracy to be allowed access. The list goes on. Physical libraries have served humanity well, but their time is rapidly passing.
Gorman is fighting a rearguard action along with every other member of a priest-like caste who made a career out of jealously guarding access to the world's books, music, movies, financial exchanges, etc. He, and others like him in the media and elsewhere, need to learn that the democratization of communication has both good and bad implications, just like the freedom of speech that makes it all possible. He should learn to stop generalizing about things that make him feel insecure. Until then, the opinions he's expressed are irrelevant and self-serving and should rightfully be ignored.
What's the difference between setjmp/lngjmp and goto? It seems to be a way of interrupting the expected control flow, and I've heard more than one argue that such is indicative of bad design.
Goto and setjmp/longjmp are similar, although the latter provides more support for saving an arbitrary context to return to. Both are still very low-level features, and using them directly in a program isn't usually considered a good idea. However, setjmp/longjmp can be used to implement capabilities, like the one in the example I gave, which are perfectly safe and valid to use. The key is that the use of the low-level feature is encapsulated in a library of functions which places some constraints on their behavior and ensures that they get used safely.
I'm not doubting your claim that "there are times that it's useful for a function not to return to its caller." Instead, it appears to me that it's a different way of thinking that is not neccessarily worse or better, just different.
The CS concept of continuations is a more general concept -- goto, setjmp/longjmp, and ordinary function call/return are all just instances of applied continuations. Understanding continuations gives a foundation for dealing with control flow of all kinds -- including for example in web applications, where continuations have gotten quite a lot of attention lately, even in mainstream systems.
This goes back to my question (which I now modify): How can learning Scheme benefit my C programming if it represents a fundamentally different way of thinking? The concrete example here seems to violate principles which are sacrosanct in "the C way" and would be labeled as "bad" by people who think in "the C way."
You wouldn't necessarily choose a solution like the one in the example just to do some ordinary programming task. However, if you're solving a more complex problem, then expressing it using a declarative non-deterministic approach - which is what that example supports - can make the difference between a relatively simple solution and a very complex one. Part of the point is to expand your mental bag of tricks so that you can choose the most effective solutions to problems. In that mex.pdf paper, Friedman writes "Programming languages are the box outside of which most programmers, architects and managers cannot think." (Perhaps a bit condescending, but this gets back to the problem of how to communicate these things if they're actually true.) One way to think outside the box of a single programming language is to learn a number of different programming languages. However, if you're going to do that, you can get a lot of bang for the buck by learning a language which embodies very general principles that can be applied in the context of many other languages.
...enable problems to be solved without worrying about low-level details.
You lost me. In every job I've worked in, I have had to worry about the low-level details in one way or another.
Sure. But you no longer usually have to worry about details at the level of assembly language. A lot of people using languages like Java and Python no longer worry about manual memory management. Over time, as language technology has advanced, low-level details get automated away. The academic languages do a very good job of this - much better, in many respects, than most of the mainstream languages. That doesn't mean you can or should throw out the mainstream languages (yet!), but it's worth having some knowledge of what the possibilities are. The main reason I raised the point, though, is that the abstraction away from low-level details is helpful for understanding things at a higher level. Not that you don't ever need the low level, but they are different levels, and are each equally important in their own way. C is primarily geared towards low-level thing
You can be charged $50,000 for downloading a 2MB MP3 from an internet source, but only $300 by driving 50Mph over the speed limit.
There's so many examples of "What the hell? This makes no sense!" that it's become a sad fact of life.
Some of these things make perfect sense: driving 50 mph above the speed limit, you're endangering the lives of a few other generally ordinary individuals - until you actually injure or kill some of them, you're OK. Downloading a 2MB MP3, you've actually committed the heinous theft of intellectual property, allegedly damaging the income of a large group of corporations. Follow the money...
I'm beginning to get the impression that you don't like elitist Lisp advocates...;)
You're the nicest LISP advocate I've met so far (though you did attempt to exploit my non-existent lack of self-esteem, which I forgive).
Thanks. Perhaps it's because I'm not a pure Lisp advocate. In the Lisp family of languages, I far prefer Scheme, which is different enough from Lisp that many pure Lisp advocates like to complain that "Scheme is not a Lisp". Re my question about why Lisp advocacy bugs you, I seriously wanted to know if somewhere underneath it all you didn't wonder what the fuss was about, whether there might be some basis for it (although I'm not defending assholism).
If it is so fundamentally different, then how did it help your programming in C?
There are some answers, admittedly anecdotal, about how these ideas can help your programming, in C in the mex.pdf paper that I linked to in my previous post. Some of it might not be entirely clear because it talks about exploiting things like continuations.
Here's how I would describe it: when you program in C and most ordinary languages, you're using a mental model that's driven by the design and limitations of real computers. Languages like Lisp, Scheme, and the functional languages like ML and Haskell use models that come from study of fundamental principles of programming and computing. These principles are much more general than any single programming language - a specific programming language like C is just a special case of these principles.
The academic languages are more closely modeled on these principles, so learning those languages tends to teach you at least some of the principles. It's really these principles that are the big deal - languages like Lisp/Scheme/ML/Haskell are just embodiments of them. Learning these principles allows you to understand C and other languages on a different level, and can lead you to think about problems in different ways, and to solutions that you might not have thought of otherwise. Once you've learned some of these general principles, it's easier to learn other languages. It's much easier to apply generalized knowledge to a specific case, than it is to generalize from a specific case (like C).
To take a concrete example, in C, functions normally return to their caller, because C only uses a single call stack per thread. However, there are times that it's useful for a function not to return to its caller, which is why setjmp & longjmp exist. But using setjmp and longjmp to do anything useful isn't the easiest thing in the world, because they're very low-level. By contrast, in some languages like Scheme and SML, there's a feature which allows you to manipulate the flow of control of a program more safely and easily, via the call/cc function. Here's a description and example of this in C terms. This is just a very specific implementation of the general concept of continuations, in C.
The linked example shows that if you know how, you can do this kind of thing in C. But most C programmers wouldn't necessarily think to solve a problem that way, because it's not an obvious solution in C. If you're familiar with these concepts from elsewhere, though, and you find somewhere where they could be useful in a C program, then it just becomes a question of how you can implement it in C. There's usually a way, and often it's well worth it. Again, that mex.pdf paper describes some examples.
Ideas from computer science and the academic languages have had a big impact on the mainstream languages: garbage collection, lexical closures, regular expressions, SQL, and many other things were developed in an academic context and eventually became mainstream. But the problem with this is that getting exposed to these features on a piecemeal basis doesn't give you the full picture of how they fit
Furthermore, if I wanted to help my programming career, then I'd learn C++. Why would you reccommend LISP over C++ if your goal is to help me in my programming career?
For the same kinds of reasons that they teach languages like Scheme in universities. What I personally recommend is learning more about the basic principles of programming, which is why I mentioned SICP.
But if you learn a language like Lisp or Scheme properly, you'll be exposed to many of the ideas covered in books like SICP, so learning the language isn't a bad substitute for studying the underlying concepts.
The reason I would recommend doing that over learning C++ is that it provides a foundation for understanding any programming language - it'll be easier to learn and understand C++ afterwards. This has been studied in universities - students who go through the more theoretical introduction to CS concepts using Scheme do better on subsequent courses in Java and C++ than students who start out learning the latter languages (I can dig up references if you care).
One problem Lisp advocates face is, how do you explain to someone that learning Lisp will teach them important things about programming that they don't yet understand? People don't want to believe that, so they respond negatively, and things devolve from there because both sides are human.
My personal experience is that I spent almost 20 years programming in languages like C, C++, Java, and various scripting languages. I developed some successful commercial products in C and C++. But when I read SICP a few years ago, I was intrigued enough to learn more about Scheme, and later some of the more functional languages like ML and OCaml. I only wish I had known about these languages, and the ideas behind them, a long time ago.
That may well be what the guy meant (certainly the guy quoted after him meant that), but it's not what he said. He mentioned generating an MP3 from these mysterious data files, for example, implying that MP3 is a format by his own definition, and thus by extension, his format that is no format must actually be a format.
If what the guy meant to say is that the idea of selling the physical media will go away, well, that was a done deal by the time iTunes and the legally authorized version of Napster appeared. If all he's saying is that the future of music is digital downloads, then he's about 5 years behind the curve.
That totally changes the emphasis for the Klingon translation. Not to mention the fact that there's more than one translation of any given word into l337.
Proof that Slashdot has jumped a bay full of sharks - no-one's pointing out that "The new format is no format" is a meaningless piece of marketing fluff uttered by someone who has no clue of what he's saying.
Where did all the "nerds" go? Seems like they've been replaced by good consumers who laugh slightly cynically at anything they're told, but end up believing it anyway, no matter how ludicrous it is.
Slashdot Mirror
I forgot about the tutorial I promised. It doesn't take checking more than a few sources to find significant differences in the account of the Titanic's sinking. That's not uncommon, and means you have to do some research.
It's quite possible, on something like a historical point of detail, that Googling casually won't find you any answers. However, it will find you many sources, which have references, which you can track down.
In this particular case, the information you're looking for is actually more like informed speculation & analysis, since the available eyewitness reports are unreliable and inconsistent. So, applying epistemological principles, you might say to yourself "how can I verify whether the few eyewitness reports about the breakup of the ship make sense?" Some answers to that are likely to be found in a technical analysis of the sinking, so you look for those.
You can also ask yourself questions like "why didn't the stern make a huge wave which would have swamped the lifeboats?" Inconsistencies in your working hypothesis are useful for drilling down towards a more accurate model of the information you're looking for. You don't need to know in advance what you're looking for, but you need to know how to recognize when you don't know something.
You test the information you find against your current model(s), and there can be give and take on both sides, i.e. you might use a model to provisionally reject certain information.
There's no predetermined way of knowing when you've reached a final conclusion. Ultimately, it comes down to how much work you want to put into it, how much information is available, etc.
When you arrive at a final conclusion, you might even find that it doesn't match any single information source out there. Who's right, you or they? It's difficult to say. If you were intent on answering that question, you'd need to examine the processes used to arrive at other accounts, if possible. A less rigorous process is likely to arrive at a less reliable answer.
As an example, when Brian Williams reported on NBC news the other night that the lead judge in Saddam Hussein's tribunal had been killed, they had apparently received the information from multiple US officials, who in turn had received the information from government sources in Baghdad. Turns out they were wrong, it wasn't the lead judge. But NBC news reported the information as "confirmed", apparently based on having spoken to multiple US governement sources. Obviously, their grasp of these issues is rather limited -- even a superficial analysis would indicate that multiple sources in the US government might match merely because they all got their information from the same place. They made a mistake in reporting the identity of the victim as "confirmed", which they could have avoided if they had applied proper procedures, including asking how reliable their sources are, and whether their sources might be contaminated by common factors. If you don't apply that level of diligence to gathering information, you have to accept that the quality of your information will be lower.
Well, your origial post was rather concise, which made it tough to divine what was behind it. When it comes to authority, I think you've misunderstood what Google is. It's an index of the WWW, not an encyclopedia. Besides, encyclopedias, newspapers, TV news & documentaries etc. get things wrong too, sometimes spectacularly, and more often than you might think. To take the Titanic as an example, problems with the history of its sinking apparently go back to the original press coverage at the time, such as the way in which Hearst's media spun the story. Given that, it's hard to see how the problem you're describing has anything to do with Google, or even with the modern-day presence of blogs.
If you want ultimate authority, there are a number of gods I can introduce you to, although you have to agree to believe unquestioningly (have "faith") in the authority of whichever one you pick. I'll note that if you picked Google as your god in this sense, you wouldn't go as far wrong as with some of the other choices out there. Short of that, google for epistemology, and you'll find that the problem you're concerned about isn't going to be solved by any algorithm, ever.
That's right, blame the tools. It couldn't possibly be you. Tell me what you want to find and I'll give you a short tutorial on how to find it. Google's not an AI, you know.
Alternatively, you could learn how to search properly.
Google recently went public. If you've drunk sufficient kool-aid, this makes them the last best hope of the tech industry. Imagine the the '90s Internet bubble all focused on one company.
Let's just hope Google doesn't decide to exploit this laser-like market focus by spinning off a host of baby Googles: then you'll be able to buy a GoogleBox PC running a browser-based GoogleOS and make phone calls over your Googlenet connection, using the Googlephone service, and you'll look up phone numbers using Whoogle (for any Google IP lawyers reading this, call me for a license on that last one).
I, for one, welcome our new over-Googlords.
Those of us with sites that can handle it want Google to index us! Bring it on, Google! Make my server your little Google-bitch!
"Realtime black hole list" is a cool-sounding phrase when it just refers to a blacklist of names. But when real black holes get involved, watch out!
Well, I'm an expert now, partly as a result of that research and subsequent work which I've done. And I'm talking about my preferred mode of study and research now. I also know many people in both academia and the commercial world who feel the same way - people for whom resources like Google, CiteSeer, ArXiv, and the electronic availability of papers, journals and books have meant a revolutionary advance in their ability to do serious research.
A key underlying assumption of Gorman's original argument was the assertion that "not many would choose to stare at a screen long enough" to read entire books. Unfortunately, that merely reflects Gorman's age and prejudices more than anything else (more on this below). Why should Gorman's Luddism inhibit those who are making knowledge available in useful formats, or those who find such formats useful? It shouldn't, and it won't. Gorman is basically saying "I don't like this personally, therefore it is bad for everyone". OTOH, many of the people who he is supposed to serve are voting with their feet (or fingers) and proving Gorman wrong. Unless he can produce arguments of more substance, then as I have said elsewhere, he deserves to be ignored.
Re the question of reading books on screens, Gorman makes a mistake common to those who argue against the deployment of technology -- he looks at the current and past situation to find reasons not to do it, rather than considering how things are changing and might change in future. I read books in various portable electronic formats, including on tablet PCs and PDAs, and current development seems to indicate that in future, we will quite likely be able to read on some kind of flexible, reflectively lit media, much like a sheet of paper. But should we wait until the perfect technological alternative to books is available before we begin digitizing the world's book collection? Obviously not, for many reasons, not least of which is that there are many ways in which that information could be useful in digital form, long before we reach some kind of electronic book nirvana.
It's not surprising that the people involved in the enterprise of digitizing a substantial fraction of humanity's books are enthusiastic about it. Seen in this context, it's clear that Gorman is primarily objecting to a shifting of the library limelight away from people such as himself.
OTOH, my girlfriend has severe dyslexia and likes audio sources. She depends heavily on audiobooks. The disability argument ultimately boils down to needing to make info available in multiple formats, not that text is better.
See if you can find out how much bandwidth this Slashdotting uses up. In fact, you could do a report on it and submit it to Slashdot, and be famous! :)
Gorman is still using the "God's Mind" hype as a straw man to attack. Does he think that they mean those claims literally? Ironically, he seems to need a lesson in reading comprehension.
Careful, you're doing the same sort of dangerous generalizing that Gorman is doing. The fact that most individual bloggers might not express their attitude to this issue in the most cogent, concise, and perfectly correct way isn't particularly relevant. Gorman has the same problem -- he's all over the map, attacking straw men and missing the real points.
One important point Gorman is missing is that Google's Pagerank system will have an amazing synergy with Google Scholar - already, you can use Google to easily identify works you might be interested in. All Google Scholar will do will make those works available online. Gorman is missing the big picture.
Further, it's not just about Google, but about the aggregation and interconnection of online resources, including digital libraries, specialist blogs and other online communities, all of which Gorman entirely ignores.
Gorman is essentially looking at a large ecosystem, but focusing only on particular parts of it -- as though he looked at the Earth and decided that most insects were useless and dangerous - ants get into people's picnic food, bees sting you. But in conjunction with the rest of the ecosystem, both end up creating something of significant value.
P.S. on the subject of the space computers take up in libraries, I would think wifi-enabled pads effectively solve that problem, taking up little more space than a book. The problems with that are just cost, damage, and theft, but certainly not space.
Re freedom of speech, I was referring to how that underlies the Internet, which is first and foremost a communication system. Physical libraries don't support the same kind of peer-to-peer communication. Gorman is apparently unused to communicating in such a wide-open forum, and is confusing all sorts of different issues, generalizing incorrectly, and responding to the wrong points.
To answer your questions, I was not an expert in the field. In this particular case, I was not reading many whole books, but used some quite extensively. However, if those books had been available online, I'd have been quite happy to do that. One can level a valid critique against online sources in that not every printed book is online, yet. But that's exactly what various digital library projects, including Google Scholar, are aiming at.
No question. However, part of where I was coming from is that I find that my reading needs are not usually served by sources that can "satisfy a good majority" of anything. Google, combined with various other online resources, is an exception, which works much better for me - even where a particular resource doesn't exist online, its existence can usually be identified online, unless one is operating in a particularly arcane field.
An AC respondent to my post suggested that professionally-maintained library catalogs (i.e. the index, not the material it represents) could not be duplicated by Google. However, my experience is the opposite. The catalog is the least interesting thing about physical libraries, to me. Part of the reason for that is that some library-like cataloging services are being done online, with services like CiteSeer (in my field), ArXiv, and various digital libraries. However, these all involve more or less complete decoupling from physical works. There are also collaborative catalogs, like CiteULike, and online communities which specialize in particular subject areas, which can be invaluable for finding out about relevant and interesting work.
The only remaining way in which I would find most physical libraries of use is if it happened to have a copy of a physical publication that I wanted to have in my hands. The reality, though, is that I rarely visit my local small-town library, even though it's within a short walking distance from my home, and I walk past it all the time. My occasional visits (perhaps three in the last five years) are never particularly fruitful. I don't think it's a particularly bad library, as small-town libraries go -- I live in a fairly wealthy town & county, so it's reasonably well-funded -- it's just not very relevant to me.
If I were the only kind of client that libraries had, then the solution would be obvious: move to an e-commerce model in which books are shipped, Netflix-DVD-rental-style, from wherever they are located, directly to people who order them online. But the round-trip cost of shipping books could approach or exceed the cost of purchasing in many cases - particularly since when you purchase, shipping is often "free", which would be difficult in the library model. (As I write this, I realize that I don't know what shipping/mailing options libraries might offer these days - perhaps I should check that out ;)
BTW, I'm all in favor of the idea of shared public spaces -- community centers, theaters, etc. -- and libraries do serve that sort of purpose as well. However, look at what's happened with commercial spaces, e.g. their aggregation in the form of malls, which happens for a variety of economic and even social reasons. I think that most tax- and charitably-funded spaces have fallen behind in exploiting those kinds of overhead-sharing, people-concentrating trends. From my admittedly non-target-market perspective, it's not clear that most libraries in future ought really to be standalone buildings, as opposed to being part of some larger civic space.
You have to live near a major urban center to have access to a decent physical library, if your needs go at all beyond the average. I once flew thousands of miles and spent weeks in NYC so that I could research a project at the NY Public Library, using materials I couldn't get anywhere near my home. This wasn't just a single book or journal that I could have had mailed to me -- I needed to track down cross-references, comb through original sources. Now I can get a signficant fraction of that material online, a fraction which increases every day.
That may be an extreme case, but it's just a scaled-up version of what every library-goer has to face -- you physically have to travel to the library, you have to deal with its constraints (quiet! don't eat in here! don't talk too loud!
Gorman is fighting a rearguard action along with every other member of a priest-like caste who made a career out of jealously guarding access to the world's books, music, movies, financial exchanges, etc. He, and others like him in the media and elsewhere, need to learn that the democratization of communication has both good and bad implications, just like the freedom of speech that makes it all possible. He should learn to stop generalizing about things that make him feel insecure. Until then, the opinions he's expressed are irrelevant and self-serving and should rightfully be ignored.
I've just emailed the yahoo address on your web page, at about 2:23 pm Eastern time. Respond here if that doesn't work.
Goto and setjmp/longjmp are similar, although the latter provides more support for saving an arbitrary context to return to. Both are still very low-level features, and using them directly in a program isn't usually considered a good idea. However, setjmp/longjmp can be used to implement capabilities, like the one in the example I gave, which are perfectly safe and valid to use. The key is that the use of the low-level feature is encapsulated in a library of functions which places some constraints on their behavior and ensures that they get used safely.
The CS concept of continuations is a more general concept -- goto, setjmp/longjmp, and ordinary function call/return are all just instances of applied continuations. Understanding continuations gives a foundation for dealing with control flow of all kinds -- including for example in web applications, where continuations have gotten quite a lot of attention lately, even in mainstream systems.
You wouldn't necessarily choose a solution like the one in the example just to do some ordinary programming task. However, if you're solving a more complex problem, then expressing it using a declarative non-deterministic approach - which is what that example supports - can make the difference between a relatively simple solution and a very complex one. Part of the point is to expand your mental bag of tricks so that you can choose the most effective solutions to problems. In that mex.pdf paper, Friedman writes "Programming languages are the box outside of which most programmers, architects and managers cannot think." (Perhaps a bit condescending, but this gets back to the problem of how to communicate these things if they're actually true.) One way to think outside the box of a single programming language is to learn a number of different programming languages. However, if you're going to do that, you can get a lot of bang for the buck by learning a language which embodies very general principles that can be applied in the context of many other languages.
Sure. But you no longer usually have to worry about details at the level of assembly language. A lot of people using languages like Java and Python no longer worry about manual memory management. Over time, as language technology has advanced, low-level details get automated away. The academic languages do a very good job of this - much better, in many respects, than most of the mainstream languages. That doesn't mean you can or should throw out the mainstream languages (yet!), but it's worth having some knowledge of what the possibilities are. The main reason I raised the point, though, is that the abstraction away from low-level details is helpful for understanding things at a higher level. Not that you don't ever need the low level, but they are different levels, and are each equally important in their own way. C is primarily geared towards low-level thing
Thanks. Perhaps it's because I'm not a pure Lisp advocate. In the Lisp family of languages, I far prefer Scheme, which is different enough from Lisp that many pure Lisp advocates like to complain that "Scheme is not a Lisp". Re my question about why Lisp advocacy bugs you, I seriously wanted to know if somewhere underneath it all you didn't wonder what the fuss was about, whether there might be some basis for it (although I'm not defending assholism).
There are some answers, admittedly anecdotal, about how these ideas can help your programming, in C in the mex.pdf paper that I linked to in my previous post. Some of it might not be entirely clear because it talks about exploiting things like continuations.
Here's how I would describe it: when you program in C and most ordinary languages, you're using a mental model that's driven by the design and limitations of real computers. Languages like Lisp, Scheme, and the functional languages like ML and Haskell use models that come from study of fundamental principles of programming and computing. These principles are much more general than any single programming language - a specific programming language like C is just a special case of these principles.
The academic languages are more closely modeled on these principles, so learning those languages tends to teach you at least some of the principles. It's really these principles that are the big deal - languages like Lisp/Scheme/ML/Haskell are just embodiments of them. Learning these principles allows you to understand C and other languages on a different level, and can lead you to think about problems in different ways, and to solutions that you might not have thought of otherwise. Once you've learned some of these general principles, it's easier to learn other languages. It's much easier to apply generalized knowledge to a specific case, than it is to generalize from a specific case (like C).
To take a concrete example, in C, functions normally return to their caller, because C only uses a single call stack per thread. However, there are times that it's useful for a function not to return to its caller, which is why setjmp & longjmp exist. But using setjmp and longjmp to do anything useful isn't the easiest thing in the world, because they're very low-level. By contrast, in some languages like Scheme and SML, there's a feature which allows you to manipulate the flow of control of a program more safely and easily, via the call/cc function. Here's a description and example of this in C terms. This is just a very specific implementation of the general concept of continuations, in C.
The linked example shows that if you know how, you can do this kind of thing in C. But most C programmers wouldn't necessarily think to solve a problem that way, because it's not an obvious solution in C. If you're familiar with these concepts from elsewhere, though, and you find somewhere where they could be useful in a C program, then it just becomes a question of how you can implement it in C. There's usually a way, and often it's well worth it. Again, that mex.pdf paper describes some examples.
Ideas from computer science and the academic languages have had a big impact on the mainstream languages: garbage collection, lexical closures, regular expressions, SQL, and many other things were developed in an academic context and eventually became mainstream. But the problem with this is that getting exposed to these features on a piecemeal basis doesn't give you the full picture of how they fit
For the same kinds of reasons that they teach languages like Scheme in universities. What I personally recommend is learning more about the basic principles of programming, which is why I mentioned SICP.
But if you learn a language like Lisp or Scheme properly, you'll be exposed to many of the ideas covered in books like SICP, so learning the language isn't a bad substitute for studying the underlying concepts.
The reason I would recommend doing that over learning C++ is that it provides a foundation for understanding any programming language - it'll be easier to learn and understand C++ afterwards. This has been studied in universities - students who go through the more theoretical introduction to CS concepts using Scheme do better on subsequent courses in Java and C++ than students who start out learning the latter languages (I can dig up references if you care).
The foundational ideas I'm talking about help you to design better programs, even when you're using mainstream languages. To give you some idea of what I'm referring to, I recommend reading The Role of the Study of Programming Languages in the Education of a Programmer (PDF).
One problem Lisp advocates face is, how do you explain to someone that learning Lisp will teach them important things about programming that they don't yet understand? People don't want to believe that, so they respond negatively, and things devolve from there because both sides are human.
My personal experience is that I spent almost 20 years programming in languages like C, C++, Java, and various scripting languages. I developed some successful commercial products in C and C++. But when I read SICP a few years ago, I was intrigued enough to learn more about Scheme, and later some of the more functional languages like ML and OCaml. I only wish I had known about these languages, and the ideas behind them, a long time ago.
That may well be what the guy meant (certainly the guy quoted after him meant that), but it's not what he said. He mentioned generating an MP3 from these mysterious data files, for example, implying that MP3 is a format by his own definition, and thus by extension, his format that is no format must actually be a format.
If what the guy meant to say is that the idea of selling the physical media will go away, well, that was a done deal by the time iTunes and the legally authorized version of Napster appeared. If all he's saying is that the future of music is digital downloads, then he's about 5 years behind the curve.
Nice try. But you have to start with the original inscription:
Ash nazg durbatulûk,
Ash nazg gimbatul,
Ash nazg thrakatulûk,
Agh burzum-ishi krimpatul.
That totally changes the emphasis for the Klingon translation. Not to mention the fact that there's more than one translation of any given word into l337.
Proof that Slashdot has jumped a bay full of sharks - no-one's pointing out that "The new format is no format" is a meaningless piece of marketing fluff uttered by someone who has no clue of what he's saying.
Where did all the "nerds" go? Seems like they've been replaced by good consumers who laugh slightly cynically at anything they're told, but end up believing it anyway, no matter how ludicrous it is.