I can understand a personal journal of some 13-year-old angsty kid being considered a "blog". But is Slashdot considered a "blog"? Is the news listing on Borland's site considered a "blog"?
In the context of Google's new search: Yes, and Yes.
Perhaps the downfall of this service is that what it is supposed to be searching is not very well defined. One cannot do exact searches when the search medium is so undefined
I don't think it's a downfall, you're just misunderstanding the intent. The intent is to search RSS and Atom feeds; sites which publish information frequently.
Google don't have separate search pages for "company websites", "news websites", "forums" and "personal homepages" for example, and it would be equally silly to have a separate one for "journals".
This is about searching a specific medium (information available over feeds) which *is* well defined. It's not about searching a particular topic - the topics should be filtered by the search words you use.
Indeed. And furthermore, just as Slashdot has Journals, things are also confused in that LiveJournal has communities (shared groups where several people can post, and then people discuss them, just like Slashdot).
I'd say the key thing is any site that publishes frequent new pieces of information - it's about the type of technology, and not about what it's used for. So Slashdot, a personal journal, and a company's news items all count.
And a good way to handle this is through things like RSS feeds - which is exactly what Google are going for here. It's about handling a specific piece of technology, which is worthy of a separate search to take advantage of the usefulness of RSS (just as it would probably be confusing to mix usenet posts and webpages together).
Unfortunately, people see the word "blog", and think that it offers some magical way to exclude all the posts related to "things they don't want to read about".
I doubt it, because this would mean removing any website which had an Atom or RSS feed. Well, to be sure it's not entirely clear how Google decide to count something as a "blog", whether they go solely by the presence of a feed or not.
But it would be annoying to have to search twice, just because you weren't sure whether what you were looking for was counted as a "blog" or not.
Slashdot for example would be excluded (yes, Google count Slashdot as a "blog", no matter how many people claim it isn't, since it has an RSS feed). Great if you don't want to stumble across whiney troll comments, not so great if a Slashdot article contains what you are looking for.
Also, the problem of blog threads cluttering up search results also occurs with webforum posts being included - but these tend not to have a feed, so this wouldn't solve the problem.
When it comes to anything else, such as ghosts, unicorns, aliens or fairies, it's simple: Do you believe in x? Yes or No, or perhaps Not Sure.
I never hear of people trying to distinguishing between "no belief in pixies" and "belief that pixies don't exist". I never hear people claiming that those who answer "No" have a belief, and therefore they require faith for that belief, or they are being irrational, or that their "belief" is a religion.
It occurs to me that all this talk of "one program with several documents" and "several instances of a program" is missing the point.
The issue was, if a user tries to run an application when its already running, should:
- The original application window be activated, or,
- A new application window be opened?
The point is that whether the latter occurs by running a new instance of the program, or whether it is clever to spawn multiple documents/windows/whatever from the same instance, is nothing to do with the user interface issue.
Even if every program has the functionality for multiple documents, it is still quicker to doubleclick and get a new document, rather than also having to go File->New, and that is what UI design is about.
If every program doesn't have this functionaltiy though (and I'm still curious to know how MacOS solves the problem for those who mentioned that platform), then the only way to open a new "document" is to run a new instance of the program, but that's more of a behind the scenes thing than UI.
If Windows was a document-centric UI, why did they invent the horrible kludge that is MDI? What is the point of having document windows that are inside and clipped by the application window?
I actually prefer this. It's tidier to group documents belonging to the same application together, rather than having them scattered about the task bar. Conceivably there are better ways of doing things, eg, having workspaces where documents of the same type are grouped together, but it's also possible to mix documents from different applications if appropriate (eg, a text editor and command prompt) - AmigaOS can do something along these lines. I've no idea what MacOS does. Or indeed, what classic MacOS does, if we're making comparisons to old versions of Windows.
Either way, this is nothing to do with being application or document centric. Even with MDI, if a user doubleclicks the application icon again, a sensibly written application will detect this and simply open a new document in the main application window (eg, Opera does this).
Running two instances of an application is the difference between Mac OS and Windows. Mac OS is document-centric, Windows is application-centric. Application windows in Mac OS ideally represent a document, while application windows in Windows represent the application.
Strange, all the different Opera windows on my Windows machine represent one application. All the different Word windows represent one application. Windows is most certainly "document-centric". The exception are applications which don't have support for more than one "document" - I would be curious to know how MacOS copes with these situations?
It's not even hypothetical. What you describe is exactly how Mac OS does it.
Okay, several people have made comparisons to MacOS, so I'm genuinely curious: Does this means it's impossible to write a one-document-only style program? That if I wrote say a calculator program with only one calculator-window, MacOS would have some clever way to allow multiple calculator windows, despite only having one instance of the application running?
for the record, that's how macs have worked since they've been able to run more than one program at the same time
I'm not sure that forcing such behaviour is a good idea - there are some programs that you might to run more than one instance of (any simple program which doesn't support multiple windows itself, eg, a command prompt or telnet client). Popping up the existing program when I want to run a new instance would be rather annoying.
The failure is that instead of building a device with sufficient intelligence to perform some task (such as operating a vacuum cleaner), we redefine the task in such a way that the device no longer needs intelligence to accomplish the task (we fuse the vacuum cleaner with the device to such an extent that it can't function as anything but a vacuum cleaner).
I see what you're saying, but the problem is that the vacuum cleaner isn't some naturally occuring device that humans managed to use through sheer intelligence; it is a device that has been specifically designed with humans in mind. It is an unfair comparison to expect a robot to use a device which wasn't specifically designed with it in mind.
These are issues of dexterity (ie, making a robot which can grip such a thing) and economics (what's the most efficient way to design it). Furthermore, the "intelligence" that a human uses in using a vacuum cleaner is not conscious thought, but comes from "brute force" computation in the brain. But of course, a "brute force" hardware approach only counts as unintelligent when a computer does it...
It's the most rational expectation. We can recognize intelligence. We know that humans are intelligent. We know that animals have some degree of intelligence. To postulate that there are other kinds of intelligence as well is a matter of faith.
Intelligence should be about end results. If someone else solved a puzzle more quickly than me, it would seem odd to regard him as unintelligent just because he used a different method to me.
Even if we wanted to look at the methods, we should be consistent. If we want to look at conscious thought rather than brute force algorithms, then yes chess computers are not quite so intelligent, but neither is a human using a vacuum cleaner, or recognising an image.
No, it's just that people are consistently underwhelmed by what AI delivers. The AI community promises a household robot
Do the people want such a robot? If it cost 10 times as much as all the separate robots put together, could only do one thing at once, had to be supplied with extra things like vacuum cleaners, took up 10 times the space, I doubt it. This isn't anything to do with AI.
The AI community promises speech recognition, delivers error-prone transcription software.
Ah yes, speech recognition, another example which humans do not do consciously through thought. If in x years time we have proper speech recognition, but it comes primarily through advancements in hardware rather than more clever algorithms, will speech recognition join chess playing as examples of AI which aren't really intelligent?
The AI community promises autonomous cars, delivers a car which completes all of 7 miles in the DARPA grand challenge.
And once this is achieved, people like you will be saying it's not AI, because it doesn't come in the form of a humanoid robot which turns the steering wheel and pushes the pedals!
The original expectation was of a robot which could do household chores, like the robot from the Jetsons. That is, a robot which could operate a vacuum cleaner, answer the door and feed the dog.
So all that means is the original expectation was flawed. If it's more efficient to have several dedicated devices rather than an all-in-one, why does this mean AI has failed? I don't see what that has to do with intelligence: putting all the different AI programs into one small computer isn't the problem, it's more an issue of robotics (ie, it's a lot harder to build something that can move about to both feed the dog and clean the floor), convenience (several devices means you can use the all at once, independently) or economics (people prefer to buy a dedicated device for what they need, rather than an expensive all-in-one).
Yes, but the expectation was that the computer would be using more or less the same mental processes as a human grandmaster. Instead we got a really fast tic-tac-toe solver which had to be repeatedly rebooted in order to perform its job.
So how come it's fair for people point to things like image recognition when "proving" the lack of decent AI? A human has a huge advantage over a computer, in that we have dedicated brain cells to perform the job, rather than having to consciously think. Why isn't this considered brute force cheating?
There is no expectation that a computer should work in the same way as a human; that's obviously a rather unfair and biased expectation.
The more accurate criticism is to point out that being intelligent at chess playing doesn't imply being intelligent at other things. However, the point is that "intelligence" covers many areas, and just as humans are good in some areas and bad in others, this is also the case with computers. It's hardly going to be the case that overnight there'll be some magic breakthrough, and we'll have "achieved AI". Rather, we'll gradually improve AI in various areas.
The problem is that people make poor predictions of the future, and fail to predict which bits of AI will be achieved first. Also, once we've achieved some area of AI, it's viewed as "obviously trivial", and people point to the remaining things computers can't do, as "proof" that computers aren't intelligent at all. The OP is correct - the problem is that AI is always being redefined.
It takes no intelligence to run all possible outcomes and pick the one most likely to succeed. Intelligence is the chess master that evaluates trees of choices and prunes them, guessing what the other person will do (rather than calculating for all possible moves) to reduce the computations necessary. I don't know of any computer programs that prune choices based on being below some threshold of probability based off likely moves. They just throw more itterations into it.
I disagree. The computation power simply doesn't exist to run "all possible outcomes". Chess programs most certainly do try to prune choices to reduce the number of moves they have to search through, and they also have to make evaluations of how good a position is (even if you can see 10 moves into the future, it's not immediately clear which of those outcomes is most likely to win).
It's easier to improve the hardware than programming.
It's not that easy - the number of possible outcomes increases exponentially with the number of turns you want to look into the future, so it requires huge advances in computing power to just get an extra turn.
The main argument for saying that today's computer programs aren't really intelligent is, I would say, that their algorithms had to be explicitly programmed, and they did not learn this themselves. I guess it depends on whether you think AI can mean any intelligent behaviour, or that it must include the ability to learn that behaviour.
It's an interesting idea, but I still believe a PIN is better:
- If someone finds out my PIN or signature, I can easily change my PIN, but not my signature.
- Given that I use my signature for all sorts of occasions, where people will see my signature, I'd say it's a lot more insecure than a PIN which I don't have to show anyone, and I can have a different one for every card.
I mean, how many people would trust PIN if you have to tell the cashier your number in order for him to check? Your system would be more secure if it was possible for a machine to compare signatures, but even then, I need to show my signature for all sorts of other things.
But is it really commonplace to involve the police and court system for when students break school rules?
I don't know how the US works, but here in the UK we have things like detention for cases like this, even when they may technically be breaking the law (eg, tampering with property they are given such as defacing a textbook might be technically vandalism, but charging them as such would be laughable). The police are only involved if there is serious injury to other pupils, or severe damage to property which they were not supposed to be using at all.
Or is this the old case of "A computer was involved, therefore it must be much more serious than actual real life damage"?
I remember one teacher at my school, talking about Victorian times (or some time in the past, at least), saying "I could have you hanged if you stole a piece of chalk". Now sure, anything these students would have got is incomparable to the death sentence. But the ludicrous thing was not so much the death sentence (I mean, hanging was far more common in general in the past for all sorts of trivial crimes, so that's nothing new), it was the idea of involving the legal system to impose a harsh sentence, rather than dealing with it in the school, just because a law had technically been broken.
The correct question to ask, if you are interested in why there are Opera users, is: Why ought an Opera user migrate to Firefox?
The thing to remember is that Opera was around long before Firefox (yes, some of us ditched IE long before it became fashionable, and whilst I'm glad that Firefox is around, to be honest it gets tiring to hear Firefox-fans telling us to ditch Opera, when we've been happily enjoying an IE-beater for years...) so even if Opera and Firefox are equally matched, and even if all new users go for Firefox, there is still not much of a reason for Opera users to switch, and no point to going through the hassle of downloading and getting used to a new browser. Having adverts isn't really a problem - I've now used my free registration, but the increase in useable screen size was negligible.
Another consideration is that (imo) Opera works better than Firefox out of the box. Firefox may or may not be able to do the things I like about Opera if I install a load of extensions, but it's a hassle to have to try and find these extensions and download them all.
Really, the ads are very small and don't bother me. I did use the free registration to see what it looks like without the ads. I still don't think the ads are a big deal.
I agree - I tried this free registration, and then thought "Oh, I now have a blank space where before there was an ad". I suppose I could fill the space with extra toolbar icons, but there's nothing I really need.
Honest question: What does Firefox have (either natively or via extensions) which Opera doesn't?
It may or may not be true that the things I love about Opera (from proper tabbed browsing, where the tabs are saved, to an "Undo" function to open pages you accidently closed) are available in Firefox via extensions. But whilst it's easy to install Firefox (indeed, I have it installed), it's a hassle to have to hunt around to see what extensions are available.
Firefox may be great, but this is why I continue to use Opera.
We look forward to federating with any service provider who shares our belief in enabling user choice and open communications. We do believe, however, that it is important to balance openness with ensuring that we maintain a safe and reliable service that protects user privacy and blocks spam and other abuses. We are using the federation opportunity with EarthLink and Sipphone to develop a set of guidelines by which all members of the federated network can work together to ensure that we protect our users while maximizing the reach of the network.
So, you'd better hope that they allow your Jabber server to join their network.
Also, I'm pretty sure that if you actually read TFA, he mentions:
While Google is busy cheerleading their openness and freedom of choice the closest thing to an open network they seem to be considering is pre-arranged formal peering agreements with just a select few of the large Jabber service providers. There are some popular ones out there and it's probably fair to expect that people who use the public @jabber.org Jabber server may someday be able to communicate with @gmail.com Jabber users. But it doesn't sound like Google is considering embracing the true spirit of open communications by using Jabber as it was designed. It's no help for those of us who already have a connection to the public Jabber network.
On a similar note, I remember when Windows 95 came out a PC fanatic friend of mine was bragging to the Amiga users about how it was now a 32 bit OS.
It was only some time later I realised that AmigaOS had been always 32 bit, since 1985 - again, we didn't brag about these things, or think about them as being anything special, so I didn't at first realise that the Amiga already had these things.
blogs serve as an "echo chamber" for like-minded people. they link to each other, post in agreement with each other, and then count the posts and post about how much they agree with each other some more.
it is inevitable that bloggers will eventually gain an understanding of the true scale of this echo chamber. once they begin to realize that they have dedicated several megabytes of bandwidth preaching mostly to the choir, their numbers will fall off, and blogging will give way to the next internet fad.
The same could be said of Slashdot. The "group think" mentality is far worse on Slashdot than on any blog, and the articles posted are often posted with a bias, yet will mostly only be read by people who agree with that point of view.
What you describe is a problem at all levels: from the national newspapers who post biased articles that are only read by people who agree with them, to a group of friends chatting in a pub, who all have the same opinions.
There is still however an advantage, in that even if you're preaching to the choir (be it on Slashdot or a blog), it is useful to convey information about events that are happening (I read Slashdot because it generally informs me of things I want to know about).
Also, even though my friends have similar views to me, I still have interesting and useful discussions through blogs - minus the idiots/trolls you find on webforums such as Slashdot.
The author seems to be confusing two different styles of blog.
The first is someone who writes (often on a standalone website) with the intention of being read by and being interesting to complete strangers. This corresponds to the first two generations.
The second, what he calls "consumer bloggers". These may use a blog for various reasons, such as personal journalling, or communicating with friends, but it's rarely intended that what they write is targetted to people who don't know them. Similarly, such people are unlikely to read blogs other than those of their friends.
Whilst there are crossovers, these are very distinct usages (so much so, that I always feel it's misleading to group them under the term "blogger" - "blog" is just a medium, and says nothing about the usage or intention of the writing).
If the first has given way to the second, I guess it's because few people want to read things written by strangers, even if they are quite interesting, and the second usage of blogs is far more powerful. But I see no evidence that the first style of blogging is in decline, and even if it is, this may not be related to "consumer blogging" at all.
I also feel the author has the timelines wrong for "consumer blogging" - LiveJournal for example has been around since 1999, which always made it easy to set up a blog (the author claims it was "a damn site harder to set up a blog than it is now" even in 2002!) and since about 2002, the vast majority of people I know have had blogs, and used them as "consumer blogs".
The term "geek blogger" is a bit misleading too - most of the people I know with blogs could be considered "geeks", but they're using them in the style of consumer blogging, rather than the first style of blogging.
I have thought about tossing up a web site I have had many good ideas for content that still is hard to find in a clear consice way on the net, but I fear that any website made after 2004 that belongs to an individual will automaticly be labled a "blog"
The term blogger to me is offencive because most blogs are written by people who, quite frankly, have NOTHING to say, have no design skills, and think that they are 1337 because they can hit a few buttons on blogger.com.
It sounds to me like you're the one insulting people who write on a website...
By the same reasoning, no one owns an employee, but we still have everything from overexcessive "we own everything you think" IP contracts, to no-compete contracts.
That's the thing I don't understand - by all means have a laissez-faire approach if you really think that works better, but that should work both ways, in the employee's favour as well and not just the employer's.
Slashdot Mirror
I can understand a personal journal of some 13-year-old angsty kid being considered a "blog". But is Slashdot considered a "blog"? Is the news listing on Borland's site considered a "blog"?
In the context of Google's new search: Yes, and Yes.
Perhaps the downfall of this service is that what it is supposed to be searching is not very well defined. One cannot do exact searches when the search medium is so undefined
I don't think it's a downfall, you're just misunderstanding the intent. The intent is to search RSS and Atom feeds; sites which publish information frequently.
Google don't have separate search pages for "company websites", "news websites", "forums" and "personal homepages" for example, and it would be equally silly to have a separate one for "journals".
This is about searching a specific medium (information available over feeds) which *is* well defined. It's not about searching a particular topic - the topics should be filtered by the search words you use.
Indeed. And furthermore, just as Slashdot has Journals, things are also confused in that LiveJournal has communities (shared groups where several people can post, and then people discuss them, just like Slashdot).
I'd say the key thing is any site that publishes frequent new pieces of information - it's about the type of technology, and not about what it's used for. So Slashdot, a personal journal, and a company's news items all count.
And a good way to handle this is through things like RSS feeds - which is exactly what Google are going for here. It's about handling a specific piece of technology, which is worthy of a separate search to take advantage of the usefulness of RSS (just as it would probably be confusing to mix usenet posts and webpages together).
Unfortunately, people see the word "blog", and think that it offers some magical way to exclude all the posts related to "things they don't want to read about".
I doubt it, because this would mean removing any website which had an Atom or RSS feed. Well, to be sure it's not entirely clear how Google decide to count something as a "blog", whether they go solely by the presence of a feed or not.
But it would be annoying to have to search twice, just because you weren't sure whether what you were looking for was counted as a "blog" or not.
Slashdot for example would be excluded (yes, Google count Slashdot as a "blog", no matter how many people claim it isn't, since it has an RSS feed). Great if you don't want to stumble across whiney troll comments, not so great if a Slashdot article contains what you are looking for.
Also, the problem of blog threads cluttering up search results also occurs with webforum posts being included - but these tend not to have a feed, so this wouldn't solve the problem.
Why does God have to be nice?
Well that's the point - he doesn't. But nonetheless, millions (if not billions) of people believe he is both nice, and capable of intervening.
Then we need a new word.
But why do we?
When it comes to anything else, such as ghosts, unicorns, aliens or fairies, it's simple: Do you believe in x? Yes or No, or perhaps Not Sure.
I never hear of people trying to distinguishing between "no belief in pixies" and "belief that pixies don't exist". I never hear people claiming that those who answer "No" have a belief, and therefore they require faith for that belief, or they are being irrational, or that their "belief" is a religion.
So why does it happen with the issue of God?
You appear to misunderstand the following concepts: Theory, fact, law, big bang, statistics, life.
Theory - something backed up with evidence, and has made successful predictions.
Intelligent Design - not a theory, but instead conjecture; a made up idea.
It occurs to me that all this talk of "one program with several documents" and "several instances of a program" is missing the point.
The issue was, if a user tries to run an application when its already running, should:
- The original application window be activated, or,
- A new application window be opened?
The point is that whether the latter occurs by running a new instance of the program, or whether it is clever to spawn multiple documents/windows/whatever from the same instance, is nothing to do with the user interface issue.
Even if every program has the functionality for multiple documents, it is still quicker to doubleclick and get a new document, rather than also having to go File->New, and that is what UI design is about.
If every program doesn't have this functionaltiy though (and I'm still curious to know how MacOS solves the problem for those who mentioned that platform), then the only way to open a new "document" is to run a new instance of the program, but that's more of a behind the scenes thing than UI.
If Windows was a document-centric UI, why did they invent the horrible kludge that is MDI? What is the point of having document windows that are inside and clipped by the application window?
I actually prefer this. It's tidier to group documents belonging to the same application together, rather than having them scattered about the task bar. Conceivably there are better ways of doing things, eg, having workspaces where documents of the same type are grouped together, but it's also possible to mix documents from different applications if appropriate (eg, a text editor and command prompt) - AmigaOS can do something along these lines. I've no idea what MacOS does. Or indeed, what classic MacOS does, if we're making comparisons to old versions of Windows.
Either way, this is nothing to do with being application or document centric. Even with MDI, if a user doubleclicks the application icon again, a sensibly written application will detect this and simply open a new document in the main application window (eg, Opera does this).
Running two instances of an application is the difference between Mac OS and Windows. Mac OS is document-centric, Windows is application-centric. Application windows in Mac OS ideally represent a document, while application windows in Windows represent the application.
Strange, all the different Opera windows on my Windows machine represent one application. All the different Word windows represent one application. Windows is most certainly "document-centric". The exception are applications which don't have support for more than one "document" - I would be curious to know how MacOS copes with these situations?
It's not even hypothetical. What you describe is exactly how Mac OS does it.
Okay, several people have made comparisons to MacOS, so I'm genuinely curious: Does this means it's impossible to write a one-document-only style program? That if I wrote say a calculator program with only one calculator-window, MacOS would have some clever way to allow multiple calculator windows, despite only having one instance of the application running?
for the record, that's how macs have worked since they've been able to run more than one program at the same time
I'm not sure that forcing such behaviour is a good idea - there are some programs that you might to run more than one instance of (any simple program which doesn't support multiple windows itself, eg, a command prompt or telnet client). Popping up the existing program when I want to run a new instance would be rather annoying.
The failure is that instead of building a device with sufficient intelligence to perform some task (such as operating a vacuum cleaner), we redefine the task in such a way that the device no longer needs intelligence to accomplish the task (we fuse the vacuum cleaner with the device to such an extent that it can't function as anything but a vacuum cleaner).
I see what you're saying, but the problem is that the vacuum cleaner isn't some naturally occuring device that humans managed to use through sheer intelligence; it is a device that has been specifically designed with humans in mind. It is an unfair comparison to expect a robot to use a device which wasn't specifically designed with it in mind.
These are issues of dexterity (ie, making a robot which can grip such a thing) and economics (what's the most efficient way to design it). Furthermore, the "intelligence" that a human uses in using a vacuum cleaner is not conscious thought, but comes from "brute force" computation in the brain. But of course, a "brute force" hardware approach only counts as unintelligent when a computer does it...
It's the most rational expectation. We can recognize intelligence. We know that humans are intelligent. We know that animals have some degree of intelligence. To postulate that there are other kinds of intelligence as well is a matter of faith.
Intelligence should be about end results. If someone else solved a puzzle more quickly than me, it would seem odd to regard him as unintelligent just because he used a different method to me.
Even if we wanted to look at the methods, we should be consistent. If we want to look at conscious thought rather than brute force algorithms, then yes chess computers are not quite so intelligent, but neither is a human using a vacuum cleaner, or recognising an image.
No, it's just that people are consistently underwhelmed by what AI delivers. The AI community promises a household robot
Do the people want such a robot? If it cost 10 times as much as all the separate robots put together, could only do one thing at once, had to be supplied with extra things like vacuum cleaners, took up 10 times the space, I doubt it. This isn't anything to do with AI.
The AI community promises speech recognition, delivers error-prone transcription software.
Ah yes, speech recognition, another example which humans do not do consciously through thought. If in x years time we have proper speech recognition, but it comes primarily through advancements in hardware rather than more clever algorithms, will speech recognition join chess playing as examples of AI which aren't really intelligent?
The AI community promises autonomous cars, delivers a car which completes all of 7 miles in the DARPA grand challenge.
And once this is achieved, people like you will be saying it's not AI, because it doesn't come in the form of a humanoid robot which turns the steering wheel and pushes the pedals!
The original expectation was of a robot which could do household chores, like the robot from the Jetsons. That is, a robot which could operate a vacuum cleaner, answer the door and feed the dog.
So all that means is the original expectation was flawed. If it's more efficient to have several dedicated devices rather than an all-in-one, why does this mean AI has failed? I don't see what that has to do with intelligence: putting all the different AI programs into one small computer isn't the problem, it's more an issue of robotics (ie, it's a lot harder to build something that can move about to both feed the dog and clean the floor), convenience (several devices means you can use the all at once, independently) or economics (people prefer to buy a dedicated device for what they need, rather than an expensive all-in-one).
Yes, but the expectation was that the computer would be using more or less the same mental processes as a human grandmaster. Instead we got a really fast tic-tac-toe solver which had to be repeatedly rebooted in order to perform its job.
So how come it's fair for people point to things like image recognition when "proving" the lack of decent AI? A human has a huge advantage over a computer, in that we have dedicated brain cells to perform the job, rather than having to consciously think. Why isn't this considered brute force cheating?
There is no expectation that a computer should work in the same way as a human; that's obviously a rather unfair and biased expectation.
The more accurate criticism is to point out that being intelligent at chess playing doesn't imply being intelligent at other things. However, the point is that "intelligence" covers many areas, and just as humans are good in some areas and bad in others, this is also the case with computers. It's hardly going to be the case that overnight there'll be some magic breakthrough, and we'll have "achieved AI". Rather, we'll gradually improve AI in various areas.
The problem is that people make poor predictions of the future, and fail to predict which bits of AI will be achieved first. Also, once we've achieved some area of AI, it's viewed as "obviously trivial", and people point to the remaining things computers can't do, as "proof" that computers aren't intelligent at all. The OP is correct - the problem is that AI is always being redefined.
It takes no intelligence to run all possible outcomes and pick the one most likely to succeed. Intelligence is the chess master that evaluates trees of choices and prunes them, guessing what the other person will do (rather than calculating for all possible moves) to reduce the computations necessary. I don't know of any computer programs that prune choices based on being below some threshold of probability based off likely moves. They just throw more itterations into it.
I disagree. The computation power simply doesn't exist to run "all possible outcomes". Chess programs most certainly do try to prune choices to reduce the number of moves they have to search through, and they also have to make evaluations of how good a position is (even if you can see 10 moves into the future, it's not immediately clear which of those outcomes is most likely to win).
It's easier to improve the hardware than programming.
It's not that easy - the number of possible outcomes increases exponentially with the number of turns you want to look into the future, so it requires huge advances in computing power to just get an extra turn.
The main argument for saying that today's computer programs aren't really intelligent is, I would say, that their algorithms had to be explicitly programmed, and they did not learn this themselves. I guess it depends on whether you think AI can mean any intelligent behaviour, or that it must include the ability to learn that behaviour.
It's an interesting idea, but I still believe a PIN is better:
- If someone finds out my PIN or signature, I can easily change my PIN, but not my signature.
- Given that I use my signature for all sorts of occasions, where people will see my signature, I'd say it's a lot more insecure than a PIN which I don't have to show anyone, and I can have a different one for every card.
I mean, how many people would trust PIN if you have to tell the cashier your number in order for him to check? Your system would be more secure if it was possible for a machine to compare signatures, but even then, I need to show my signature for all sorts of other things.
But is it really commonplace to involve the police and court system for when students break school rules?
I don't know how the US works, but here in the UK we have things like detention for cases like this, even when they may technically be breaking the law (eg, tampering with property they are given such as defacing a textbook might be technically vandalism, but charging them as such would be laughable). The police are only involved if there is serious injury to other pupils, or severe damage to property which they were not supposed to be using at all.
Or is this the old case of "A computer was involved, therefore it must be much more serious than actual real life damage"?
I remember one teacher at my school, talking about Victorian times (or some time in the past, at least), saying "I could have you hanged if you stole a piece of chalk". Now sure, anything these students would have got is incomparable to the death sentence. But the ludicrous thing was not so much the death sentence (I mean, hanging was far more common in general in the past for all sorts of trivial crimes, so that's nothing new), it was the idea of involving the legal system to impose a harsh sentence, rather than dealing with it in the school, just because a law had technically been broken.
To rephrase, why ought I migrate to Opera?
That's not rephrasing the question.
The correct question to ask, if you are interested in why there are Opera users, is: Why ought an Opera user migrate to Firefox?
The thing to remember is that Opera was around long before Firefox (yes, some of us ditched IE long before it became fashionable, and whilst I'm glad that Firefox is around, to be honest it gets tiring to hear Firefox-fans telling us to ditch Opera, when we've been happily enjoying an IE-beater for years...) so even if Opera and Firefox are equally matched, and even if all new users go for Firefox, there is still not much of a reason for Opera users to switch, and no point to going through the hassle of downloading and getting used to a new browser. Having adverts isn't really a problem - I've now used my free registration, but the increase in useable screen size was negligible.
Another consideration is that (imo) Opera works better than Firefox out of the box. Firefox may or may not be able to do the things I like about Opera if I install a load of extensions, but it's a hassle to have to try and find these extensions and download them all.
Really, the ads are very small and don't bother me. I did use the free registration to see what it looks like without the ads. I still don't think the ads are a big deal.
I agree - I tried this free registration, and then thought "Oh, I now have a blank space where before there was an ad". I suppose I could fill the space with extra toolbar icons, but there's nothing I really need.
Honest question: What does Firefox have (either natively or via extensions) which Opera doesn't?
It may or may not be true that the things I love about Opera (from proper tabbed browsing, where the tabs are saved, to an "Undo" function to open pages you accidently closed) are available in Firefox via extensions. But whilst it's easy to install Firefox (indeed, I have it installed), it's a hassle to have to hunt around to see what extensions are available.
Firefox may be great, but this is why I continue to use Opera.
(from http://www.google.com/talk/developer.html#service)
We look forward to federating with any service provider who shares our belief in enabling user choice and open communications. We do believe, however, that it is important to balance openness with ensuring that we maintain a safe and reliable service that protects user privacy and blocks spam and other abuses. We are using the federation opportunity with EarthLink and Sipphone to develop a set of guidelines by which all members of the federated network can work together to ensure that we protect our users while maximizing the reach of the network.
So, you'd better hope that they allow your Jabber server to join their network.
Also, I'm pretty sure that if you actually read TFA, he mentions:
While Google is busy cheerleading their openness and freedom of choice the closest thing to an open network they seem to be considering is pre-arranged formal peering agreements with just a select few of the large Jabber service providers. There are some popular ones out there and it's probably fair to expect that people who use the public @jabber.org Jabber server may someday be able to communicate with @gmail.com Jabber users. But it doesn't sound like Google is considering embracing the true spirit of open communications by using Jabber as it was designed. It's no help for those of us who already have a connection to the public Jabber network.
On a similar note, I remember when Windows 95 came out a PC fanatic friend of mine was bragging to the Amiga users about how it was now a 32 bit OS.
It was only some time later I realised that AmigaOS had been always 32 bit, since 1985 - again, we didn't brag about these things, or think about them as being anything special, so I didn't at first realise that the Amiga already had these things.
blogs serve as an "echo chamber" for like-minded people. they link to each other, post in agreement with each other, and then count the posts and post about how much they agree with each other some more.
it is inevitable that bloggers will eventually gain an understanding of the true scale of this echo chamber. once they begin to realize that they have dedicated several megabytes of bandwidth preaching mostly to the choir, their numbers will fall off, and blogging will give way to the next internet fad.
The same could be said of Slashdot. The "group think" mentality is far worse on Slashdot than on any blog, and the articles posted are often posted with a bias, yet will mostly only be read by people who agree with that point of view.
What you describe is a problem at all levels: from the national newspapers who post biased articles that are only read by people who agree with them, to a group of friends chatting in a pub, who all have the same opinions.
There is still however an advantage, in that even if you're preaching to the choir (be it on Slashdot or a blog), it is useful to convey information about events that are happening (I read Slashdot because it generally informs me of things I want to know about).
Also, even though my friends have similar views to me, I still have interesting and useful discussions through blogs - minus the idiots/trolls you find on webforums such as Slashdot.
The author seems to be confusing two different styles of blog.
The first is someone who writes (often on a standalone website) with the intention of being read by and being interesting to complete strangers. This corresponds to the first two generations.
The second, what he calls "consumer bloggers". These may use a blog for various reasons, such as personal journalling, or communicating with friends, but it's rarely intended that what they write is targetted to people who don't know them. Similarly, such people are unlikely to read blogs other than those of their friends.
Whilst there are crossovers, these are very distinct usages (so much so, that I always feel it's misleading to group them under the term "blogger" - "blog" is just a medium, and says nothing about the usage or intention of the writing).
If the first has given way to the second, I guess it's because few people want to read things written by strangers, even if they are quite interesting, and the second usage of blogs is far more powerful. But I see no evidence that the first style of blogging is in decline, and even if it is, this may not be related to "consumer blogging" at all.
I also feel the author has the timelines wrong for "consumer blogging" - LiveJournal for example has been around since 1999, which always made it easy to set up a blog (the author claims it was "a damn site harder to set up a blog than it is now" even in 2002!) and since about 2002, the vast majority of people I know have had blogs, and used them as "consumer blogs".
The term "geek blogger" is a bit misleading too - most of the people I know with blogs could be considered "geeks", but they're using them in the style of consumer blogging, rather than the first style of blogging.
I have thought about tossing up a web site I have had many good ideas for content that still is hard to find in a clear consice way on the net, but I fear that any website made after 2004 that belongs to an individual will automaticly be labled a "blog"
The term blogger to me is offencive because most blogs are written by people who, quite frankly, have NOTHING to say, have no design skills, and think that they are 1337 because they can hit a few buttons on blogger.com.
It sounds to me like you're the one insulting people who write on a website...
By the same reasoning, no one owns an employee, but we still have everything from overexcessive "we own everything you think" IP contracts, to no-compete contracts.
That's the thing I don't understand - by all means have a laissez-faire approach if you really think that works better, but that should work both ways, in the employee's favour as well and not just the employer's.