I will counter your question with one of my own. Why wait until there's a trend, and millions of people are affected, instead of just Missoula?
I might have agreed with you on the grounds that slashdot isn't the right forum for this message to spread. But I'm not sure about that either. Slashdot's demographics tend to be very far left leaning (socialists, anarchists) and very far right leaning (libertarians, reactionaries, other kinds of anarchists). Some of these groups are very well organized and can perhaps help Missoula residents.
Perhaps you aren't interested. But no one put a gun to your head and made you read the summary. I don't mean to be flippant. There are plenty of other interesting stories on the main page that fit your interests better.
Parent is not flamebait. Parent was asking an interesting question that has major consequences for AppleTV's utility/hackability. Does it run a stripped down OS X? An embedded iPod-like OS? I would imagine the latter, but if the former, it'll only be a matter of time before VLC and other open source media tools support it.
MS Team: We are very serious! Me: Shall I call an ambulance? MS Team: No no no.... we are very serious about you, a Linux user. Me: I see... take your chairs..... ooops take your seats and let's talk seriously. MS Team: We are serious about you. Which category Linux user are you? Me: I'm just a Linux user. MS Team: Are you an experimenter, follower, aficionado, transitioner or... Me: I'm just a normal Linux user. MS Team: Before we talk among ourselves in the presentation we must classify you. Why do you like Linux?
Wrong already. They won't ask why you like Linux. They'll first try to establish repoire. Then they'll "level with you", and get you to talk about the things you dislike about Linux or what you would like to be more productive. Finally, they'll tell you that they have a product that would serve your needs better.
This is Sales 101, and figuring out how to approach that final step, based on your answers to the previous portion of the conversation, is what this website is for.
This isn't a reason to be careful. It's an opportunity for anyone invested.
Microsoft has clearly done a lot of market research. And has published a summary, meant more-or-less for internal consumption, publicly. This is the perfect chance for distro leaders to read what Microsoft recommends to their sales people, gather legitimate issues with Linux, and serve their customers better. I'm pointing at you Ubuntu, SUSE, RedHat -- though anyone can take this opportunity.
Heh, posting anonymously isn't going to prove much.
But assuming you're telling the truth: Why don't you try to clarify what you meant by 'usability' as mentioned in other threads on the topic?
Here's what I'd like to see: A simple, elegant GUI with full 3d acceleration (perhaps beryl/compiz based) without gimmicky, useless eye candy. (Some gimmicky eye candy is useful). Perhaps with a GNUStep back end for running Cocoa applications like TextMate. And a great package management system. Something like a cross between apt and portage would be fantastic.
Sun has the resources to make this happen. Hop to it!:-)
I love your work on Naruto Shippudden. The scene where Naruto learns that Gaara became Kazekage is great. How did you manage to make such great art for your fanfic?
Sure, there is that old argument, "the people buying from Tony probably wouldn't have bought it retail anyway".
There's another side to this argument. People that are willing to buy it retail wouldn't have bought from Tony.
Let me offer a parallel to the phenomenon I'm trying to describe. We regularly hear about people trying to convince their bosses that open source software is a worth while investment. And often, the bosses veto the suggestion. Open source software is often better than proprietary software, and yet the bosses feel the need to buy proprietary. Why? Lots of reasons: support, CYA, fanboyism, etc.
Pirated software does little to satisfy those reasons. Private enterprise makes up the majority of IT spending.
There's no such thing as a "high gain omni-directional antenna". An antenna's gain is defined to be the logarithm of the ratio of signal strength and what an omnidirectional antenna's signal strength would be. (To be precise, the definition involves "perfect" omnidirectional antennas. The so-called 'isotropic' antennae)
Obviously, the ratio of what an omni puts out and what an omni "would" put out is 1/1. The gain is therefore 0 dBi.
Your lightbulb analogy is the perfect example of this idea. A plain old lightbulb will put out 100J every second, but "split up" evenly in all directions. A 100W spotlight will put out the same power, but illuminate the spot much more brightly. It is a "high gain" light. The plain old light bulb is a "no gain" light, because gain is defined in terms of what it outputs. Indeed, high gain antennas are inherently more directional. Just a minor detail.
This is a good point for me to respond to the Ask Slashdotter. You already have two fairly capable routers. I know the Linksys will easily run OpenWRT, HyperWRT-thibor, etc. HyperWRT allows (*ahem*, I forget exactly what they call it) wireless "bridging", which lets you run multiple routers as access points to the same wireless network. (So instead of having two separate WLANs, you have a "roaming" cell phone-like effect). HyperWRT also gives you some control over how much power the router's transmitter can use.
Depending on your home's architecture, you might want to look into using high gain directional antennae. You want to use directional antennae to help eliminate RF leakage. If your neighbors can see your wireless network, you're radiating power that could be useful to you. In this case, your router is also picking up your neighbor's noise.
So figure out where you actually use your wireless network. Use high gain antennae to "illuminate" those spots. Look into "cardioid" or "log directional" antennas. Also, check out http://en.wikipedia.org/wiki/Log-periodic_antenna. It has a nice picture of the energy distribution of a log directional antenna over North America. Obviously, you're working on a smaller scale, but the idea remains the same. Put the antenna in a corner, radiating towards the longest dimension in your home.
I suspect that a firmware upgrade and some smart antenna placement will solve your problem.
I can imagine a solution to clean up the former (although this would make nuclear fuel even more expensive), but I haven't yet seen a (proven) solution for the latter*
Solution for the latter: materials recycling using pebble bed and newer reactors. Followed by disposing unrecyclable materials in the largest reactor in the solar system, Super Man style.
That's a fair point. Mine was that if we're late for this hypothetical 600,000 year cycle, we're late for a reason. That reason might cause an eruption tomorrow. Or never. Or anywhere in between. Saying that we're merely late is not alarmist.
As someone else joked, if a girl misses her period, she's probably pregnant. We shouldn't expect her period to start tomorrow just because she's late.
I'm gonna have to start watching it. I've caught up with my other shows: One Piece, Naruto Shippuden, Dexter, Lost. Who are the best Death Note fansubbers?
Imagine this scenario: massive asteroid hits Earth along a fault line, perhaps somewhere in Asia. And sets of Yellowstone. Which uproots the nuclear waste facilities in Nevada. Which future people will worship due to it's invisible deadliness. Eventually, teratoma becomes a sign of reverence and power. Monsters will rule the world.
Every picture will be have to compared with the rest of pictures in the folder, that is a very big number of comparison to make, which will require a lot of processing power. This might not be feasible to run on a regular hardware as it would require more time then manual sorting.
Not true. That's the absolute worst case naive data miner. Things can be much better.
Actually, I was a mathematics major and now work as a professional mathematician. Thanks for playing!
I meant any of: 1. Characterized by or acting with rigor: a rigorous program to restore physical fitness. 2. Full of rigors; harsh: a rigorous climate. 3. Rigidly accurate; precise.
The first two would be satisfied by a good political science program. Since you mentioned 'numbers' and objectivity, I'll say that you're making yourself sound like a huge idiot to the several thousand of us who know anything about mathematical logic and how it relates to philosophy. That debate has been going on for centuries and has not been settled, even though the anti-Realist claims are true.
Hint: Is the Continuum Hypothesis true? Is it false? Or does it depend on who's asking?
Luckily, if a product isn't good enough for the courts, it's not good enough for the police. Even if the open source product is actually better. Yes, this is a good thing.
So what does the interpreted bit have to do with this discussion?
I was unclear. I hoped that the relation would be clear, but it's my fault it wasn't. There are two issues: the one I was originally responding to, and my smart one. First, the code the GGP posted is trivially "decidable" in the sense the GGP meant. The inner block is not going to run, and we can know that very quickly just by running/reading it. Obviously, Rice's theorem stops us from being able to always rely on being able to decide if a program has a non-trivial property. But Google doesn't have to decide that.
JavaScript is an interpreted language with full, open specifications (barring browser silliness). Decidability is irrelevant in this context because Google is in a position where they can implement a JavaScript interpreter with hard limits on the resources available to the JavaScript script. After all, if a.js fails to satisfy Google, it's the author's loss and not Google's. So it shouldn't matter if there's code out there with while (1) { sleep 30; } (or whatever the JavaScript equivalent is).
Granted, Google could do this even if JavaScript was a compiled language. But the solution would either be brittle (making changes to the compiler to build hard limits into each executable, write files to create a file system based data structure, etc) or messy and slow (using some kind of scripting and IPC to control parallel tasks) or both. Perhaps other schemes are possible. Basically, the "interpreted bit" is relevant to the discussion because it facilitates Google's hypothetical implementation.
All true. However, I'd say that your first and second points are essentially a non-issue. Google already scans the "dark web" if it's accessible to the public, even if only accessible in non-trivial ways. Which is to say, they're already requesting massive numbers of documents and generating enormous data structures to mine. Dealing with JavaScript would be more of the same in this respect.
Your other points are better. Malicious JavaScript could easily tie the GoogleBot up, if Google's hypothetical JavaScript interpreter didn't have built-in runtime limits. Time limits are one option. Disallowing certain JavaScript constructs is another possibility.
Decidability is a non-issue in this context. Your example falls flat, because JavaScript is an interpreted language. All Google would have to do is run an interpreter and data mine the results.
Slashdot Mirror
Please mod this post +5, Troll.
You know you want to.
I will counter your question with one of my own. Why wait until there's a trend, and millions of people are affected, instead of just Missoula?
I might have agreed with you on the grounds that slashdot isn't the right forum for this message to spread. But I'm not sure about that either. Slashdot's demographics tend to be very far left leaning (socialists, anarchists) and very far right leaning (libertarians, reactionaries, other kinds of anarchists). Some of these groups are very well organized and can perhaps help Missoula residents.
Perhaps you aren't interested. But no one put a gun to your head and made you read the summary. I don't mean to be flippant. There are plenty of other interesting stories on the main page that fit your interests better.
Sleep's power draw is actually quite negligible. I've woken up my PowerBook G4 after sleeping, on battery power, for a week.
I run OS X and Linux, and even I found your troll boring.
NEXT!
Parent is not flamebait. Parent was asking an interesting question that has major consequences for AppleTV's utility/hackability. Does it run a stripped down OS X? An embedded iPod-like OS? I would imagine the latter, but if the former, it'll only be a matter of time before VLC and other open source media tools support it.
MS Team: We are very serious!
Me: Shall I call an ambulance?
MS Team: No no no.... we are very serious about you, a Linux user.
Me: I see... take your chairs..... ooops take your seats and let's talk seriously.
MS Team: We are serious about you. Which category Linux user are you?
Me: I'm just a Linux user.
MS Team: Are you an experimenter, follower, aficionado, transitioner or...
Me: I'm just a normal Linux user.
MS Team: Before we talk among ourselves in the presentation we must classify you. Why do you like Linux?
Wrong already. They won't ask why you like Linux. They'll first try to establish repoire. Then they'll "level with you", and get you to talk about the things you dislike about Linux or what you would like to be more productive. Finally, they'll tell you that they have a product that would serve your needs better.
This is Sales 101, and figuring out how to approach that final step, based on your answers to the previous portion of the conversation, is what this website is for.
This isn't a reason to be careful. It's an opportunity for anyone invested.
Microsoft has clearly done a lot of market research. And has published a summary, meant more-or-less for internal consumption, publicly. This is the perfect chance for distro leaders to read what Microsoft recommends to their sales people, gather legitimate issues with Linux, and serve their customers better. I'm pointing at you Ubuntu, SUSE, RedHat -- though anyone can take this opportunity.
Heh, posting anonymously isn't going to prove much.
:-)
But assuming you're telling the truth: Why don't you try to clarify what you meant by 'usability' as mentioned in other threads on the topic?
Here's what I'd like to see: A simple, elegant GUI with full 3d acceleration (perhaps beryl/compiz based) without gimmicky, useless eye candy. (Some gimmicky eye candy is useful). Perhaps with a GNUStep back end for running Cocoa applications like TextMate. And a great package management system. Something like a cross between apt and portage would be fantastic.
Sun has the resources to make this happen. Hop to it!
Hello timecop!
I love your work on Naruto Shippudden. The scene where Naruto learns that Gaara became Kazekage is great. How did you manage to make such great art for your fanfic?
Sure, there is that old argument, "the people buying from Tony probably wouldn't have bought it retail anyway".
There's another side to this argument. People that are willing to buy it retail wouldn't have bought from Tony.
Let me offer a parallel to the phenomenon I'm trying to describe. We regularly hear about people trying to convince their bosses that open source software is a worth while investment. And often, the bosses veto the suggestion. Open source software is often better than proprietary software, and yet the bosses feel the need to buy proprietary. Why? Lots of reasons: support, CYA, fanboyism, etc.
Pirated software does little to satisfy those reasons. Private enterprise makes up the majority of IT spending.
There's no such thing as a "high gain omni-directional antenna". An antenna's gain is defined to be the logarithm of the ratio of signal strength and what an omnidirectional antenna's signal strength would be. (To be precise, the definition involves "perfect" omnidirectional antennas. The so-called 'isotropic' antennae)
. It has a nice picture of the energy distribution of a log directional antenna over North America. Obviously, you're working on a smaller scale, but the idea remains the same. Put the antenna in a corner, radiating towards the longest dimension in your home.
Obviously, the ratio of what an omni puts out and what an omni "would" put out is 1/1. The gain is therefore 0 dBi.
Your lightbulb analogy is the perfect example of this idea. A plain old lightbulb will put out 100J every second, but "split up" evenly in all directions. A 100W spotlight will put out the same power, but illuminate the spot much more brightly. It is a "high gain" light. The plain old light bulb is a "no gain" light, because gain is defined in terms of what it outputs. Indeed, high gain antennas are inherently more directional. Just a minor detail.
This is a good point for me to respond to the Ask Slashdotter. You already have two fairly capable routers. I know the Linksys will easily run OpenWRT, HyperWRT-thibor, etc. HyperWRT allows (*ahem*, I forget exactly what they call it) wireless "bridging", which lets you run multiple routers as access points to the same wireless network. (So instead of having two separate WLANs, you have a "roaming" cell phone-like effect). HyperWRT also gives you some control over how much power the router's transmitter can use.
Depending on your home's architecture, you might want to look into using high gain directional antennae. You want to use directional antennae to help eliminate RF leakage. If your neighbors can see your wireless network, you're radiating power that could be useful to you. In this case, your router is also picking up your neighbor's noise.
So figure out where you actually use your wireless network. Use high gain antennae to "illuminate" those spots. Look into "cardioid" or "log directional" antennas. Also, check out http://en.wikipedia.org/wiki/Log-periodic_antenna
I suspect that a firmware upgrade and some smart antenna placement will solve your problem.
I can imagine a solution to clean up the former (although this would make nuclear fuel even more expensive), but I haven't yet seen a (proven) solution for the latter*
Solution for the latter: materials recycling using pebble bed and newer reactors. Followed by disposing unrecyclable materials in the largest reactor in the solar system, Super Man style.
That's a fair point. Mine was that if we're late for this hypothetical 600,000 year cycle, we're late for a reason. That reason might cause an eruption tomorrow. Or never. Or anywhere in between. Saying that we're merely late is not alarmist.
As someone else joked, if a girl misses her period, she's probably pregnant. We shouldn't expect her period to start tomorrow just because she's late.
I'm gonna have to start watching it. I've caught up with my other shows: One Piece, Naruto Shippuden, Dexter, Lost. Who are the best Death Note fansubbers?
Imagine this scenario: massive asteroid hits Earth along a fault line, perhaps somewhere in Asia. And sets of Yellowstone. Which uproots the nuclear waste facilities in Nevada. Which future people will worship due to it's invisible deadliness. Eventually, teratoma becomes a sign of reverence and power. Monsters will rule the world.
Eh, it's better than Armaggedon.
No, it's kind of like saying that it erupts at a 600,000 year cycle, and that it's 40,000 years late this time.
Every picture will be have to compared with the rest of pictures in the folder, that is a very big number of comparison to make, which will require a lot of processing power. This might not be feasible to run on a regular hardware as it would require more time then manual sorting.
Not true. That's the absolute worst case naive data miner. Things can be much better.
Actually, I was a mathematics major and now work as a professional mathematician. Thanks for playing!
I meant any of:
1. Characterized by or acting with rigor: a rigorous program to restore physical fitness.
2. Full of rigors; harsh: a rigorous climate.
3. Rigidly accurate; precise.
The first two would be satisfied by a good political science program. Since you mentioned 'numbers' and objectivity, I'll say that you're making yourself sound like a huge idiot to the several thousand of us who know anything about mathematical logic and how it relates to philosophy. That debate has been going on for centuries and has not been settled, even though the anti-Realist claims are true.
Hint: Is the Continuum Hypothesis true? Is it false? Or does it depend on who's asking?
Because if you know that a Linux distro works on the hardware, they all will in a short span of time.
Or a slashdotter...
Thanks! I'll send you a card, if you'd like.
Luckily, if a product isn't good enough for the courts, it's not good enough for the police. Even if the open source product is actually better. Yes, this is a good thing.
A bachelor's degree is a requirement to join the force in many major cities.
So what does the interpreted bit have to do with this discussion?
.js fails to satisfy Google, it's the author's loss and not Google's. So it shouldn't matter if there's code out there with while (1) { sleep 30; } (or whatever the JavaScript equivalent is).
I was unclear. I hoped that the relation would be clear, but it's my fault it wasn't. There are two issues: the one I was originally responding to, and my smart one. First, the code the GGP posted is trivially "decidable" in the sense the GGP meant. The inner block is not going to run, and we can know that very quickly just by running/reading it. Obviously, Rice's theorem stops us from being able to always rely on being able to decide if a program has a non-trivial property. But Google doesn't have to decide that.
JavaScript is an interpreted language with full, open specifications (barring browser silliness). Decidability is irrelevant in this context because Google is in a position where they can implement a JavaScript interpreter with hard limits on the resources available to the JavaScript script. After all, if a
Granted, Google could do this even if JavaScript was a compiled language. But the solution would either be brittle (making changes to the compiler to build hard limits into each executable, write files to create a file system based data structure, etc) or messy and slow (using some kind of scripting and IPC to control parallel tasks) or both. Perhaps other schemes are possible. Basically, the "interpreted bit" is relevant to the discussion because it facilitates Google's hypothetical implementation.
All true. However, I'd say that your first and second points are essentially a non-issue. Google already scans the "dark web" if it's accessible to the public, even if only accessible in non-trivial ways. Which is to say, they're already requesting massive numbers of documents and generating enormous data structures to mine. Dealing with JavaScript would be more of the same in this respect.
Your other points are better. Malicious JavaScript could easily tie the GoogleBot up, if Google's hypothetical JavaScript interpreter didn't have built-in runtime limits. Time limits are one option. Disallowing certain JavaScript constructs is another possibility.
Decidability is a non-issue in this context. Your example falls flat, because JavaScript is an interpreted language. All Google would have to do is run an interpreter and data mine the results.