You've never heard of pie menus, did you? Sure, the exact implementation seen in the video was shit (no text, didn't capture the radial position of the cursor) but research done on real pie menus shows that they're much better for a contextual menu that always has the same amount of elements.
Basically, the only thing the stupid algorithm knows about Go is the simple rules and how to score the board. It knows nothing of strategy, tactics, strong shapes, living shapes, dead shapes, etc. Of course, it may be doing some sophisticated analysis to determine fruitful branches so as to not waste time on bad ones, but that doesn't defeat my point; it just means that with more computing power, you don't have to be so choosy. Knowledge about the complexities of the game is not required for the machine to win with this method, and that makes me call bullshit.
For me, it makes me wonder if this is exactly what the human brain does to play go. Both approaches look remarkably similar (and far more flexible than a rule-based approach). That would make the brain a device for massively parallel computation on extremely simple rules, which seems plausible.
I wonder why MMOs don't use flash-like minigames to solve the grinding problem. They are massively popular, after all, and they typically don't suffer to bots.
CWmike writes "Ever try to remember who you bumped into at the store a few days back? Well, you're not alone. And IBM researchers are working on software that just may help you better recollect all the forgotten pieces of your life. This week, the company unveiled Pensieve, software that stores images, sounds, and text on everyday mobile devices, then allows the user extract them later on, to help them recall names, faces, conversations and events. IBM's project is akin to one that Gordon Bell and other scientists at Microsoft Research have been working on for the past nine years."
it's a consequence of aligning edges with the nearest pixel. You can either have blurry fonts with precise spacing or clear fonts with the sort of sub-pixel spacing variations you describe.
Or also you could have both, but that would require rewriting the font rendering engine, as the GP suggested.
The man was complaining that his team spends too much time reviewing lots of detailed screenshots. And you want him to also review lots of detailed guidelines? How is that supposed to help?
Design is subjective, but user efficiency is measurable. If the application has well defined goals, there's a way to find whether the current implementation matches them.
It's just that more of those measures aren't technical - i.e. don't think maths, in How many clicks? - think in social sciences, field research, satisfaction surveys.
Damn, even *some* important measures are mathematical (how long does it take to buy the product, how many errors the user makes, how many users give up without completing), you just need a human expert to discover which measures are the good ones.
somebody would have to publish a high quality interface guideline
You're thinking in developer terms. But an interface MUST NOT be designed by the developer, even if he has good guidelines to follow - those are just for achieving consistency. They're just like coding standards, not requisites.
The overall process of each application function must be designed by someone who is watching the user following - it's exactly *debugging* the process, only with an expert in "human processors" doing the debugging. You wouldn't program a persistence layer without running the classes against a database engine. The "human2computer" "interface" layer is just the same - the design must be provided by an expert, and refined through real usage.
it would require a lot of people revisiting finished elements of their apps because somebody said it doesn't look pretty enough.
If you only revisit *finished* elements, you're doing it wrong. What about if the finished element shows the "input your credit card" form before showing the final price, but *real people* don't give their card numbers if they don't know how much will the shopping total cost? That requisite should be discovered in the early stages of development.
Finally, in the topic of aesthetics, beautiful works better. So it is one of the important requirements in the human API, just like batch throughput is important in the DDBB API.
Ah, but solving a series of linear equations is also a collection of various algorithms: those for addition, product, substraction and division, with some extra rules to combine them properly.
If some other website flat-out reproduced Slashdot's appearance, changing the green to orange, would a person with basic familiarity with Slashdot look at the new site and consider that the two might originate from the same people?
What! You're implying that this is not written by the/. editors?
Also the Balance Board. Although it sells tied to WiiFit, it's making its way into several of the next-iteration games (Rayman, snowboarding, Wii Sports Resort, Wii Music). And it's a big success.
The Wii Motion will sell tied to Wii Sports Resort and used by Clone Wars, so it can repeat that formula.
The only feature necessary for 4.0 was fixing the binary compatibility. 4.0 has that, so yes, it was feature complete according to the planned feature set.
That doesn't preclude adding MORE features in later releases. Each one will be published when their planned features are finished - but some other features can be postponed for later.
The idea is that, after the system has determined what your preferred range is, it can go to the edge of that range and select a new title. The goal is to find you a new book that you probably wouldn't read on your own, but is not so far outside your preference that you'll never make it through. Reward exploration.:) You can then adjust how far outside your preferred range you care to stray - mildly different, or total radical?
I suggest you to include the search terms -buy and -price. That makes wonders in getting Google to show you the actually relevant pieces of information.
You misunderstand the scope of the GPL. Despite its wording, the real effect it has is not on freedom of code, but on freedom of whole software projects.
The intent never was to provide "free code", but to gather a group of people around the goals for which that code was created, and ensuring that those people has no artificial restrictions to use the code created to fulfill that goal. In this context, there is harm in someone modifying the code and releasing closed versions. Even if the original, unmodified code is still present, the new code is competing in the same environment, and people in the project has no longer access to the enhancements - they must rebuild it for themselves. The net result is that the free rider benefited from the community work, and now the community is left worse than it began.
This is what the GPL restrictions avoid, and it *does* protect freedom for the people in the project (although it reduces freedom for the people outside it. Yes, the GPL is selfish in its nature; and that's why it works so well in practice, its economical principles are sound).
If you value code created so that "a person could do anything s/he wanted with that code, including making internal improvements and releasing it under a proprietary license", be free to release under MIT and BSD. The GPL was NEVER intended for that purpose. If you find that offensive, well, what can one say about people being offended because others don't have the same exact intentions than oneself?
The delicious hypocrisy is that GPL zealots are also often the loudest preachers of "copyright infringement doesn't hurt anyone"
Now that was pure trolling. GPL supporters are ALL for copyright enforcement (at least within the present body of law), since the GPL requires it to work. We can hope for a reduced length truer to its original intent, or for enforcing the (legal) fair use provisions, but that's a different beast than "copyright infringement".
Slashdot Mirror
If your account number were 2000 units lower, I'd ask you to return your geek card.
You've never heard of pie menus, did you? Sure, the exact implementation seen in the video was shit (no text, didn't capture the radial position of the cursor) but research done on real pie menus shows that they're much better for a contextual menu that always has the same amount of elements.
For me, it makes me wonder if this is exactly what the human brain does to play go. Both approaches look remarkably similar (and far more flexible than a rule-based approach). That would make the brain a device for massively parallel computation on extremely simple rules, which seems plausible.
I wonder why MMOs don't use flash-like minigames to solve the grinding problem. They are massively popular, after all, and they typically don't suffer to bots.
Mozilla Summit rocks!
CWmike writes "Ever try to remember who you bumped into at the store a few days back? Well, you're not alone. And IBM researchers are working on software that just may help you better recollect all the forgotten pieces of your life. This week, the company unveiled Pensieve, software that stores images, sounds, and text on everyday mobile devices, then allows the user extract them later on, to help them recall names, faces, conversations and events. IBM's project is akin to one that Gordon Bell and other scientists at Microsoft Research have been working on for the past nine years."
So your problem is with the "share and share-alike" spirit of the GPL, not with the companies dual licensing.
Is that MS stock or Apple stock?
And second error, what you think looks strange is called kerning.
Actually no, kerning would be solving that strange look. It looks weird because it isn't doing proper kerning.
it's a consequence of aligning edges with the nearest pixel. You can either have blurry fonts with precise spacing or clear fonts with the sort of sub-pixel spacing variations you describe.
Or also you could have both, but that would require rewriting the font rendering engine, as the GP suggested.
The man was complaining that his team spends too much time reviewing lots of detailed screenshots. And you want him to also review lots of detailed guidelines? How is that supposed to help?
Guidelines don't a UI-design technique make. (Hint - developers shouldn't design interfaces, only implement them).
Design is subjective, but user efficiency is measurable. If the application has well defined goals, there's a way to find whether the current implementation matches them.
It's just that more of those measures aren't technical - i.e. don't think maths, in How many clicks? - think in social sciences, field research, satisfaction surveys.
Damn, even *some* important measures are mathematical (how long does it take to buy the product, how many errors the user makes, how many users give up without completing), you just need a human expert to discover which measures are the good ones.
somebody would have to publish a high quality interface guideline
You're thinking in developer terms. But an interface MUST NOT be designed by the developer, even if he has good guidelines to follow - those are just for achieving consistency. They're just like coding standards, not requisites.
The overall process of each application function must be designed by someone who is watching the user following - it's exactly *debugging* the process, only with an expert in "human processors" doing the debugging. You wouldn't program a persistence layer without running the classes against a database engine. The "human2computer" "interface" layer is just the same - the design must be provided by an expert, and refined through real usage.
it would require a lot of people revisiting finished elements of their apps because somebody said it doesn't look pretty enough.
If you only revisit *finished* elements, you're doing it wrong. What about if the finished element shows the "input your credit card" form before showing the final price, but *real people* don't give their card numbers if they don't know how much will the shopping total cost? That requisite should be discovered in the early stages of development.
Finally, in the topic of aesthetics, beautiful works better. So it is one of the important requirements in the human API, just like batch throughput is important in the DDBB API.
Well I, for one, am not amused.
Ah, but solving a series of linear equations is also a collection of various algorithms: those for addition, product, substraction and division, with some extra rules to combine them properly.
What! You're implying that this is not written by the /. editors?
Yes, but that only work up the point where you don't feel respected anymore. Then it degenerates into the situation described by the grandparent.
Also the Balance Board. Although it sells tied to WiiFit, it's making its way into several of the next-iteration games (Rayman, snowboarding, Wii Sports Resort, Wii Music). And it's a big success.
The Wii Motion will sell tied to Wii Sports Resort and used by Clone Wars, so it can repeat that formula.
The only feature necessary for 4.0 was fixing the binary compatibility. 4.0 has that, so yes, it was feature complete according to the planned feature set.
That doesn't preclude adding MORE features in later releases. Each one will be published when their planned features are finished - but some other features can be postponed for later.
3. They call a ghoul to perform a level drain.
Shouldn't be a problem, they have many in-house.
What can I say? Your ideas about the ideal interface differ from mine.
Except that, this algorithm can be used to find books that broke away from what you'd read before.
Actually, the algorithm at Booklamp could be used to put to have your horizons expanded. According to the site's FAQ ("Can I have BookLamp find me books that are different than a certain book, if I want to?"), they plan a feature to find books just out of your comfort zone, to "Press your limits":
I suggest you to include the search terms -buy and -price. That makes wonders in getting Google to show you the actually relevant pieces of information.
You misunderstand the scope of the GPL. Despite its wording, the real effect it has is not on freedom of code, but on freedom of whole software projects.
The intent never was to provide "free code", but to gather a group of people around the goals for which that code was created, and ensuring that those people has no artificial restrictions to use the code created to fulfill that goal. In this context, there is harm in someone modifying the code and releasing closed versions. Even if the original, unmodified code is still present, the new code is competing in the same environment, and people in the project has no longer access to the enhancements - they must rebuild it for themselves. The net result is that the free rider benefited from the community work, and now the community is left worse than it began.
This is what the GPL restrictions avoid, and it *does* protect freedom for the people in the project (although it reduces freedom for the people outside it. Yes, the GPL is selfish in its nature; and that's why it works so well in practice, its economical principles are sound).
If you value code created so that "a person could do anything s/he wanted with that code, including making internal improvements and releasing it under a proprietary license", be free to release under MIT and BSD. The GPL was NEVER intended for that purpose. If you find that offensive, well, what can one say about people being offended because others don't have the same exact intentions than oneself?
Now that was pure trolling. GPL supporters are ALL for copyright enforcement (at least within the present body of law), since the GPL requires it to work. We can hope for a reduced length truer to its original intent, or for enforcing the (legal) fair use provisions, but that's a different beast than "copyright infringement".