Meanwhile, GPU developers have created a component that processes floating point math very quickly, sold for much less $:FLOPS than Sparcs (or any other CPU). Combining a T1 and GPGPU offers "best of breed" economies of scale appropriate to each component, like installing 3rd party memory and HD rather than the expensive Sun brands.
That's why GPGPU is an interesting strategy. GPU APIs offer parallelism, too. When those APIs can be harnessed with bus signalling that's high-enough level symbolically to exploit the processing speed without bottlenecking on the bus data bandwidth, it's quite a compelling architecture. It definitely needs a lot of work to make truly general purpose (or enough to get critical mass from a lot of niches). But that's why encouraging people like the poster to try it is worth doing.
I wasn't criticizing developers in my original post. But I am certainly criticizing you, now that you've crawled out from under your rock.
You are the poster child for developers who can't communicate. Who posts obnoxious hyperbole and strawman arguments. Using my post as a random excuse to insult users. Thanks for proving my point. Now go back to your flatfood cube and stop bothering the humans - they might not use the software you're cranking out when they realize what might be lurking behind it. You're giving us developers who can communicate with users a bad name.
Because the users, including the developers, are the reason we have software.
Because product development driven by "what can we make, then who will want it" ("if we build it, they will come") is a proven recipe for failure, all day long, every day. While "what do people want, then how do we give it to them" is the recipe for success of every product.
Asking for software that prioritizes users, including developers who use it, isn't pissing on those people. Bitterness like yours, which invents insult where there is none, shows one of the ways in which developers are often not qualified to communicate with the people who use the software, including deciding what those users want.
This is the first time Ubuntu is trying the "let the developers decide everything" model. What makes you so obtuse that you can't tell that this new attempt is a step backwards for their strategy?
My "off the cuff development model" is to put users' requirements first, before developers'. I never claimed to invent that "model", just to favor it. Who the hell are you?
I am a developer, with decades of success forming my attitude. Now you should bow down and worship me, the way you worship the Ubuntu developers.
Even with Linux, the vast majority of users will never touch the code, except perhaps to compile the source for installation. Even then, most users install binaries. Ubuntu's a desktop Linux, so that ratio is even more pronounced. Even server distros have more users than developers, especially when you consider all the users of the servers' services, compared to the relatively few developers.
It's true that Linux developers are users. And it's true that Linux users can develop - that's the open source definition. But most don't exercise the option - they only execute the code.
What is true about Linux is the attitude of many developers that users are just like developers. That's really wrong. And holds back the development of Linux by often ignoring the requirements of nondeveloper users, the majority of the userbase, the majority of the user cases. That doesn't have to be a priority of developers, whose own use cases are their priority. But for commercial developers, and any developers who want the critical mass benefits of a majority OS's userbase, it's essential.
For me, who is primarily a user of Linux, though a developer of apps and services that run on Linux, user requirements drive the value of the product much more than developer requirements. That Linux is so valuable, though its requirements have been driven largely by developers, shows just how much more valuable Linux could be if user requirements were met in proportion to their relevance.
"Dapper is billed as the solid, long term support release"
Nothing so temporary as a permanent solution.
"You can submit feature requests through various channels."
Except that those channels won't be determining Eft's features - the developer channels will be.
"The Ubuntu devs try very hard. It's not as easy as you might think."
Who says they don't? Who says it's easy? All I said was that delivering a product driven by what the producers want to make isn't as good as what the consumers want to consume.
Your post is wishful spinning, and a guilt trip. My post is constructive product development advice.
Someone else I talked with last night in meatspace was also similarly concerned. It was useful to write/publish it. I neglected to mention that it is possible to blast the moon hard enough with an initial antimatter blast on the far side that raises a cloud of (matter) dust on the near side, and remaining unconverted antimatter into orbit from the far side, which settles on the near side in microscopic total conversions with the matter dust. Which could completely revise the image on the near side.
Odds are astronomically (pun intended) against it. But a properly controlled blast might make a helluva billboard. Read more about it in my upcoming SF story, "Unpublished";).
The only way to affect the Moon that would change the tides, or any of its effects on its co-orbit with the Earth is to reduce the Moon's mass, or to move its orbit around our common center of gravity (located inside the Earth, off-center).
The total amount of matter lost in the energy conversion from combination with antimatter is only the amount equal to the antimatter. So 10mg anti consumes 10mg matter. Even extremely large amounts of antimatter, like kilograms, would consume only kilograms of matter. Not going to "matter" (pun intended:).
The explosion probably couldn't move much of the Moon's mass out of the orbital system. 10mg is enough only to move a rocket back and forth between Earth and Mars, which is a tiny mass at low acceleration for human comfort. Escape velocity from the Earth/Moon system might be achieved by some matter, but again, probably not enough to matter. Moving the whole Moon enough to disturb its orbit also seems out of the question. Though perhaps kilograms of antimatter might offer enough energy in the conversion to move it, injecting that much energy into the whole Earth/Moon system would be absorbed by both worlds, so probably be dampened. I'm not certain that such a change couldn't stress Earth faults, triggering earthquakes and volcanoes, but I still doubt it - and that's at hundreds of times the masses we're discussing in this story.
Remember that even if the Moon cracked in half - not bloody likely - the effect wouldn't change the mass and orbits of the two worlds, around each other or around the Sun. I'd hope that the large spaces on the Moon and in orbit near it would be used to maintain separation so that there'd be no possibility of more than a tiny fraction of the total antimatter converting with matter before the planned time, under control. And that they'd put the whole shebang across the far side, so any conversion would release high energy rays towards the Earth only with the whole Moon shielding us.
So it's not completely impossible for such large energies released in the wrong way to affect us. But it's extremely unlikely, especially when the advantages of distance and local masses are exploited.
I think I understand you English, which seems OK. But I don't understand why you think the incoming queue will get overloaded if people can read the rejected pile (basically equivalent to reading the queue after rejection). FWIW, reading the queue will likely reduce the amount of duplicate submissions, and offer a way to avoid the kinds of submissions that get rejected.
I don't think Slashdot's incoming queue will be any larger without an increase in readership. That would mean an increase in ad rates and income, which would mean more money to pay queue editors. Slashdot is already scaled to hundreds of thousands, even millions, of active users and dozens of stories. If they can't handle a larger queue, especialy with redundancy tools matching URLs and quoted summaries, the problem is management. Which could be very real - I like Taco, but he might not be able to keep his job at a larger scale demanding better management. I'd like to think he could grow into it with the right assistance.
How does getting rejected by Slashdot humiliate the submitter, when their submissions are there for the public to see? The submitter's identity might show a conflict of interest preventing it from publication - like the astroturf campaigns Slashdot probably gets all day long, which occasionally make it to publication. Submitters can submit anonymously, or even with a onetime userID they create just to submit.
The problem is that there's no convenient single character symbol for "per". The "p" or "P" symbol also means pico or peta, and can combine with other unit symbols. The "*" symbol is not commonly used to mean "per", though it's most appropriate. The ":" symbol is a direct synonym for "÷" - it's the original Greek notation that meant "is to" in ratios, particularly "is to / as" ratios like "a:b::b:a+b". As such, it reflects both the numerator/denominator relationship, and the implication of multiplication, as Greeks didn't actually reduce ratios by division. The "dot" symbol might be better, as it literally means "per", but it's not that convenient to type, its HTML entity #183 ("") is not rendered by all browsers, and it's not that much more likely recognized by the people who won't understand ":".
I don't like "km/h" because I've had people on Slashdot get confused by the "/" and start dividing, especially when I'm discussing some formula not often discussed (or completely novel), especially when abbreviating may ratios/rates in a single post. I wouldn't care, except they often get "me too" replies, which means lots more silent people making the same mistake.
That post was the first I've gotten where people didn't understand the ":". The poster actually ignored it, which seems the least sensible way to react. But then, they had already decided to argue with my conclusion, deciding I had just exaggerated the numbers. In a post they didn't even bother completing to make any sense at all. Then they returned to the thread with an obnoxious argument. I have no reason whatsoever to respect their inability to infer what ":" means, or anything else, except their ability to muddy the waters for other similarly less observant readers. So I might clarify, but only for the benefit of those reading who are actually interested in a discussion, rather than a rant.
I've got plenty of rant available, but I'd always rather talk it out - then I get to learn something. So I've done so in this thread, when the other posters have offered the same level of respect. But I'm just about done.
10mg of antimatter will initially consume only 10mg of matter.
The amount of matter consumed in the next few seconds by the blast will be extremely large.
The radiation released in that blast will kill everything for hundreds of miles around the center.
The huge explosion can cause vast wildfires and even an earthquake if close enough to a fault.
There is no reason to believe that such a facility can be shielded in any meaningful way - the technology is too new, there's not nearly enough testing of the materials.
The risk of this event happening will be continuous while the facility is in operation. Theft can move that risk to anywhere on Earth. And military applications can make it happen again and again.
Meanwhile, even nuclear technology has proven to be too powerful for humans to manage without severely disrupting our security. Right now we're almost as close to a nuclear war as we've ever been - except when a single Soviet soldier said "no" to orders to launch during the Cuban Missile Crisis, or when we actually nuked Japan. The actual hysteria we'd get with antimatter facilities on Earth could very well push us over the edge we've never left.
Antimatter production/consumption in space to store solar or nuclear energy would put human energy industry in the next higher category. The kind of power scales we could engineer would offer a replacement for most of the dirty power we transform on Earth, so we'd be back within the ecosystem's capacity to recycle the pollution. We'd get a space travel industry, and lots more.
That is the reality. The fantasy that we can just start making antimatter on Earth without those unacceptable risks totally ignores the facts of the past 50 years. And the facts of the past few months, past few days with Iran and the rest of the "nuclear club". You can fool yourself because you want the shiny new toy, but you can't fool me.
I posted about an entire antimatter industry, not just the single mission. That scope necessitates the use of an actual cost per gram, a useful number beyond just the single Mars mission discussed in the article.
Converse to your argument, keeping the costs in terms of the arbitrary 10mg would have artifically reduced the number, without real utility. So I never assumed the readers were stupid - that inference comes soley from your own mind.
I expected to talk about the economics of an orbital antimatter industry. Instead I had a largely annoying argument about ratio symbols with a largely obnoxious group of readers - read the Subject residue of just this tangential message. I'd say I grossly overestimated my audience's intellect, with a few exceptions.
Even today, Microsoft has been revealed to cover up insecurity problems. But I don't see any backlash. Except by people like me who already know better than to trust Microsoft.
Over the course of my posting history, I've acknowledged whenever I've been proven wrong. You can find whatever you please, you can just make summary judgements without offering evidence or argument, you can call me names. Because you're a fool, as you've just demonstrated with your otherwise useless post. Don't offer me any advice with an attitude like that - don't pretend to do me any favors. Don't call me a liar or anything else you can't back up.
If you're such a weirdo that you want to come out of the woodwork to lie about my history, you should just keep it to yourself. You're the one who looks like a fool.
Microsoft has been reported to turn over hosted email. But I'm talking about the track records of the two companies. Microsoft continuously exposes users' data to egregious security holes, exposes private info on a daily basis. I don't know how you could offer an example of a compromised high profile MS server as hypothetical: the news is full of those stories every month. FWIW, the hosting and banking industries are not majority Microsoft servers - certainly not the ones that process secure info. While the only news of Google's insecurity is a government attempt to get search histories, which was fought vigorously. And successfully, turning over at most aggregate stats that don't seem to compromise security.
Microsoft also has baggage of much more complex software, including much more client cases exposed to higher risk and long histories of unrevisited code.
Google has a much more fragile brand than Microsoft's well abused one, based entirely on free, uninstalled software that people can stop using the second a competitor does better. Microsoft has a monopoly that has trained people to accept severe insecurity. And Microsoft has many other breaches of public and private trust - too numerable and well known to list, while Google has none. So I might trust Google (my exact word), though not necessarily. While there's not a chance I would trust Microsoft - I'd trust MS to betray me, as they always have.
Slashdot Mirror
Meanwhile, GPU developers have created a component that processes floating point math very quickly, sold for much less $:FLOPS than Sparcs (or any other CPU). Combining a T1 and GPGPU offers "best of breed" economies of scale appropriate to each component, like installing 3rd party memory and HD rather than the expensive Sun brands.
That's why GPGPU is an interesting strategy. GPU APIs offer parallelism, too. When those APIs can be harnessed with bus signalling that's high-enough level symbolically to exploit the processing speed without bottlenecking on the bus data bandwidth, it's quite a compelling architecture. It definitely needs a lot of work to make truly general purpose (or enough to get critical mass from a lot of niches). But that's why encouraging people like the poster to try it is worth doing.
I wasn't criticizing developers in my original post. But I am certainly criticizing you, now that you've crawled out from under your rock.
You are the poster child for developers who can't communicate. Who posts obnoxious hyperbole and strawman arguments. Using my post as a random excuse to insult users. Thanks for proving my point. Now go back to your flatfood cube and stop bothering the humans - they might not use the software you're cranking out when they realize what might be lurking behind it. You're giving us developers who can communicate with users a bad name.
Because the users, including the developers, are the reason we have software.
Because product development driven by "what can we make, then who will want it" ("if we build it, they will come") is a proven recipe for failure, all day long, every day. While "what do people want, then how do we give it to them" is the recipe for success of every product.
Asking for software that prioritizes users, including developers who use it, isn't pissing on those people. Bitterness like yours, which invents insult where there is none, shows one of the ways in which developers are often not qualified to communicate with the people who use the software, including deciding what those users want.
My brain makes me amazing - thanks for asking.
This is the first time Ubuntu is trying the "let the developers decide everything" model. What makes you so obtuse that you can't tell that this new attempt is a step backwards for their strategy?
My "off the cuff development model" is to put users' requirements first, before developers'. I never claimed to invent that "model", just to favor it. Who the hell are you?
I am a developer, with decades of success forming my attitude. Now you should bow down and worship me, the way you worship the Ubuntu developers.
I thought the "Ken Burns Effect" is the effect that popular US history documentaries have on US national archives. Like the Smithsonian giving Viacom's Showtime cable station a monopoly on access to the archives. Kinda like burning the public archives, without burning the money it makes a private corporation.
Even with Linux, the vast majority of users will never touch the code, except perhaps to compile the source for installation. Even then, most users install binaries. Ubuntu's a desktop Linux, so that ratio is even more pronounced. Even server distros have more users than developers, especially when you consider all the users of the servers' services, compared to the relatively few developers.
It's true that Linux developers are users. And it's true that Linux users can develop - that's the open source definition. But most don't exercise the option - they only execute the code.
What is true about Linux is the attitude of many developers that users are just like developers. That's really wrong. And holds back the development of Linux by often ignoring the requirements of nondeveloper users, the majority of the userbase, the majority of the user cases. That doesn't have to be a priority of developers, whose own use cases are their priority. But for commercial developers, and any developers who want the critical mass benefits of a majority OS's userbase, it's essential.
For me, who is primarily a user of Linux, though a developer of apps and services that run on Linux, user requirements drive the value of the product much more than developer requirements. That Linux is so valuable, though its requirements have been driven largely by developers, shows just how much more valuable Linux could be if user requirements were met in proportion to their relevance.
"Dapper is billed as the solid, long term support release"
Nothing so temporary as a permanent solution.
"You can submit feature requests through various channels."
Except that those channels won't be determining Eft's features - the developer channels will be.
"The Ubuntu devs try very hard. It's not as easy as you might think."
Who says they don't? Who says it's easy? All I said was that delivering a product driven by what the producers want to make isn't as good as what the consumers want to consume.
Your post is wishful spinning, and a guilt trip. My post is constructive product development advice.
So we'll determine the next version of Ubuntu that practically everyone uses by what the developers want.
How about an experiment where the users determine the features of the leading desktop Linux distro?
It was a terrorist attack.
Someone else I talked with last night in meatspace was also similarly concerned. It was useful to write/publish it. I neglected to mention that it is possible to blast the moon hard enough with an initial antimatter blast on the far side that raises a cloud of (matter) dust on the near side, and remaining unconverted antimatter into orbit from the far side, which settles on the near side in microscopic total conversions with the matter dust. Which could completely revise the image on the near side.
;).
Odds are astronomically (pun intended) against it. But a properly controlled blast might make a helluva billboard. Read more about it in my upcoming SF story, "Unpublished"
If vwe can't reply to the queue, only view the rejects, how would that increase troll stories?
The only way to affect the Moon that would change the tides, or any of its effects on its co-orbit with the Earth is to reduce the Moon's mass, or to move its orbit around our common center of gravity (located inside the Earth, off-center).
:).
The total amount of matter lost in the energy conversion from combination with antimatter is only the amount equal to the antimatter. So 10mg anti consumes 10mg matter. Even extremely large amounts of antimatter, like kilograms, would consume only kilograms of matter. Not going to "matter" (pun intended
The explosion probably couldn't move much of the Moon's mass out of the orbital system. 10mg is enough only to move a rocket back and forth between Earth and Mars, which is a tiny mass at low acceleration for human comfort. Escape velocity from the Earth/Moon system might be achieved by some matter, but again, probably not enough to matter. Moving the whole Moon enough to disturb its orbit also seems out of the question. Though perhaps kilograms of antimatter might offer enough energy in the conversion to move it, injecting that much energy into the whole Earth/Moon system would be absorbed by both worlds, so probably be dampened. I'm not certain that such a change couldn't stress Earth faults, triggering earthquakes and volcanoes, but I still doubt it - and that's at hundreds of times the masses we're discussing in this story.
Remember that even if the Moon cracked in half - not bloody likely - the effect wouldn't change the mass and orbits of the two worlds, around each other or around the Sun. I'd hope that the large spaces on the Moon and in orbit near it would be used to maintain separation so that there'd be no possibility of more than a tiny fraction of the total antimatter converting with matter before the planned time, under control. And that they'd put the whole shebang across the far side, so any conversion would release high energy rays towards the Earth only with the whole Moon shielding us.
So it's not completely impossible for such large energies released in the wrong way to affect us. But it's extremely unlikely, especially when the advantages of distance and local masses are exploited.
I think I understand you English, which seems OK. But I don't understand why you think the incoming queue will get overloaded if people can read the rejected pile (basically equivalent to reading the queue after rejection). FWIW, reading the queue will likely reduce the amount of duplicate submissions, and offer a way to avoid the kinds of submissions that get rejected.
I don't think Slashdot's incoming queue will be any larger without an increase in readership. That would mean an increase in ad rates and income, which would mean more money to pay queue editors. Slashdot is already scaled to hundreds of thousands, even millions, of active users and dozens of stories. If they can't handle a larger queue, especialy with redundancy tools matching URLs and quoted summaries, the problem is management. Which could be very real - I like Taco, but he might not be able to keep his job at a larger scale demanding better management. I'd like to think he could grow into it with the right assistance.
How does getting rejected by Slashdot humiliate the submitter, when their submissions are there for the public to see? The submitter's identity might show a conflict of interest preventing it from publication - like the astroturf campaigns Slashdot probably gets all day long, which occasionally make it to publication. Submitters can submit anonymously, or even with a onetime userID they create just to submit.
The problem is that there's no convenient single character symbol for "per". The "p" or "P" symbol also means pico or peta, and can combine with other unit symbols. The "*" symbol is not commonly used to mean "per", though it's most appropriate. The ":" symbol is a direct synonym for "÷" - it's the original Greek notation that meant "is to" in ratios, particularly "is to / as" ratios like "a:b::b:a+b". As such, it reflects both the numerator/denominator relationship, and the implication of multiplication, as Greeks didn't actually reduce ratios by division. The "dot" symbol might be better, as it literally means "per", but it's not that convenient to type, its HTML entity #183 ("") is not rendered by all browsers, and it's not that much more likely recognized by the people who won't understand ":".
I don't like "km/h" because I've had people on Slashdot get confused by the "/" and start dividing, especially when I'm discussing some formula not often discussed (or completely novel), especially when abbreviating may ratios/rates in a single post. I wouldn't care, except they often get "me too" replies, which means lots more silent people making the same mistake.
That post was the first I've gotten where people didn't understand the ":". The poster actually ignored it, which seems the least sensible way to react. But then, they had already decided to argue with my conclusion, deciding I had just exaggerated the numbers. In a post they didn't even bother completing to make any sense at all. Then they returned to the thread with an obnoxious argument. I have no reason whatsoever to respect their inability to infer what ":" means, or anything else, except their ability to muddy the waters for other similarly less observant readers. So I might clarify, but only for the benefit of those reading who are actually interested in a discussion, rather than a rant.
I've got plenty of rant available, but I'd always rather talk it out - then I get to learn something. So I've done so in this thread, when the other posters have offered the same level of respect. But I'm just about done.
10mg of antimatter will initially consume only 10mg of matter.
The amount of matter consumed in the next few seconds by the blast will be extremely large.
The radiation released in that blast will kill everything for hundreds of miles around the center.
The huge explosion can cause vast wildfires and even an earthquake if close enough to a fault.
There is no reason to believe that such a facility can be shielded in any meaningful way - the technology is too new, there's not nearly enough testing of the materials.
The risk of this event happening will be continuous while the facility is in operation. Theft can move that risk to anywhere on Earth. And military applications can make it happen again and again.
Meanwhile, even nuclear technology has proven to be too powerful for humans to manage without severely disrupting our security. Right now we're almost as close to a nuclear war as we've ever been - except when a single Soviet soldier said "no" to orders to launch during the Cuban Missile Crisis, or when we actually nuked Japan. The actual hysteria we'd get with antimatter facilities on Earth could very well push us over the edge we've never left.
Antimatter production/consumption in space to store solar or nuclear energy would put human energy industry in the next higher category. The kind of power scales we could engineer would offer a replacement for most of the dirty power we transform on Earth, so we'd be back within the ecosystem's capacity to recycle the pollution. We'd get a space travel industry, and lots more.
That is the reality. The fantasy that we can just start making antimatter on Earth without those unacceptable risks totally ignores the facts of the past 50 years. And the facts of the past few months, past few days with Iran and the rest of the "nuclear club". You can fool yourself because you want the shiny new toy, but you can't fool me.
I'd love to see Slashdot's "rejected" queue. That would really be a testament to "open source", of the journalistic kind.
I posted about an entire antimatter industry, not just the single mission. That scope necessitates the use of an actual cost per gram, a useful number beyond just the single Mars mission discussed in the article.
Converse to your argument, keeping the costs in terms of the arbitrary 10mg would have artifically reduced the number, without real utility. So I never assumed the readers were stupid - that inference comes soley from your own mind.
I expected to talk about the economics of an orbital antimatter industry. Instead I had a largely annoying argument about ratio symbols with a largely obnoxious group of readers - read the Subject residue of just this tangential message. I'd say I grossly overestimated my audience's intellect, with a few exceptions.
Fuck you.
Even today, Microsoft has been revealed to cover up insecurity problems. But I don't see any backlash. Except by people like me who already know better than to trust Microsoft.
$5:g
3g
$15
You don't divide by the g amount, you multiply. "/" means divide.
Over the course of my posting history, I've acknowledged whenever I've been proven wrong. You can find whatever you please, you can just make summary judgements without offering evidence or argument, you can call me names. Because you're a fool, as you've just demonstrated with your otherwise useless post. Don't offer me any advice with an attitude like that - don't pretend to do me any favors. Don't call me a liar or anything else you can't back up.
If you're such a weirdo that you want to come out of the woodwork to lie about my history, you should just keep it to yourself. You're the one who looks like a fool.
You need to look more closely at how rates are ratios of even unlike units.
Microsoft has been reported to turn over hosted email. But I'm talking about the track records of the two companies. Microsoft continuously exposes users' data to egregious security holes, exposes private info on a daily basis. I don't know how you could offer an example of a compromised high profile MS server as hypothetical: the news is full of those stories every month. FWIW, the hosting and banking industries are not majority Microsoft servers - certainly not the ones that process secure info. While the only news of Google's insecurity is a government attempt to get search histories, which was fought vigorously. And successfully, turning over at most aggregate stats that don't seem to compromise security.
Microsoft also has baggage of much more complex software, including much more client cases exposed to higher risk and long histories of unrevisited code.
Google has a much more fragile brand than Microsoft's well abused one, based entirely on free, uninstalled software that people can stop using the second a competitor does better. Microsoft has a monopoly that has trained people to accept severe insecurity. And Microsoft has many other breaches of public and private trust - too numerable and well known to list, while Google has none. So I might trust Google (my exact word), though not necessarily. While there's not a chance I would trust Microsoft - I'd trust MS to betray me, as they always have.