I just looked over on my bookcase, and indeed there is the manual for my HP48G. It is about an inch thick or so - small type. The irony is that I probably paid the same price for it as people pay now, so it isn't like they need to save money on paper.
Then again, the same was true of MS Office, Operating Systems, and everything else back in the 80s and early 90s. Expectations have changed.
Uh, the students are in fact required to buy certain models. Most classes restrict what kinds of calculators you're allowed to use on exams. Most would prohibit the use of netbooks.
So, the customer's ability to get what they really want is limited.
It looks like it discharges cleaner water into the ocean. That might make it illegal in the US, due to some bizarre interpretation of EPA regulations.
I was reading that European and Middle-East cleanup ships were turned away from helping with cleanup. This was due to their principle of operation - they would skim muck from the surface, separate the oil and water, and discharge the water back into the ocean, keeping the oil in a holding tank. The problem is that they didn't clean the water they discharged up to EPA standards. Cleanup ships were required to store the mostly-clean water they would otherwise discharge, which means they would have to make frequent trips to dump their tanks.
This is of course absurd, since a ship that takes in a 50% oil solution and outputs a 0.1% oil solution can operate indefinitely and only make things better. The regulation was meant to apply to devices that prevent oil spills in the first place (dumping nothing into a clean ocean is better than dumping 0.1% oil solution).
Utterly amazing. Clearly regulation is necessary (otherwise there is no incentive to not pollute in the first place), but such strict application of law is something right out of Kefka or Wikipedia...:)
This wasn't a new clinical trial as far as I can tell, but it was mining past trials for data and applying them to new hypothesis. A recent xkcd showed some of the problems with this kind of argument. If the data didn't support permanent loss of sexual function, perhaps it would support heart attacks, cancer, blindness, or perhaps cures for any of the above.
If you think this is a possible side-effect, then hypothesize that, conduct a large trial, and see what the results are.
Sure, mining past data can generate ideas for new trials, but it is not a substitute for them. It is far too susceptible to finding patterns that aren't really there - especially with the amount of noise in a typical clinical trial dataset.
What we really need are open source cloud solutions. There are very few out there now and they're generally weak.
How does open source software help if you still hand your plain text data to a third party running the service? The only way to protect data that you want an untrustworthy third party to hold for you is to encrypt it such that they do not have access. Just encrypt your messages before sending them to Google.
If there were an open-source equivalent to Gmail I wouldn't have to send my mail to anybody - I'd run my own cloud solution on my own hardware under my own control.
Short of that, there is no point in encrypting email and sending it to Google, unless I just want to use them as a backup solution. I couldn't read the emails via Gmail without some kind of browser plugin to decrypt them again, which defeats most of the purpose of having them in the cloud in the first place.
For backups, encrypting before uploading makes a lot more sense - as long as you don't need easy remote access to the backups.
the idea of building a phone for the gigantic foreign market never hit them, apparently.
That is pretty typical Japanese culture. Foreign markets are usually of secondary concern to them. The only reason they were as successful as they were was that American manufacturers were SO reluctant to change that even an unenthusiastic foreign corporation could clean house.
Why would anyone use that? Won't it get finger prints all over it?
I know somebody who works in Pharmaceuticals and they had to buy special equipment to make pills for the Japanese market. To be accepted by a Japanese consumer, a pill needs to be perfectly white without any sign of blemish or visible imperfection. While this is a general concern in any manufacturing process, an American won't ask their doctor to prescribe something different if they find one pill with a ding or a slight discoloration in the bottle.
Yup, the reality is that little that is done in major companies is done primarily for the company's benefit. What happens, happens because it will make a manager somewhere better off in some way (financially, psychologically, whatever). To the extent that company well-being is actually tied to rewards that may happen to also benefit the company, but that is usually a secondary consideration.
When some manager comes up with some crazy mandate that everybody knows will hurt the company, do they all raise a big protest? No way! They all get in there and write up at the end of the year how enthusiastically they implemented it so that they can collect their annual bonus. Then, when the next manager enthusiastically reverses the policy of the first one, everybody enacts that policy with equal vigor. Being bad for a company will never slow down a big initiative. Now, if the big initiative is bad for individual employees, then it will meet with resistance regardless of impact on the company.
Right, which doesn't prove any new hypothesis - it merely disproves their original hypothesis. Likely at best they can show the standard model is wrong, and that is if there is no systematic source of error, as a statistical test only accounts for random error.
Don't get me wrong - I'm interested in where this goes. However, this is just the start of a long process.
And again, I never claimed that the people doing the test didn't understand statistics, or that they hadn't accounted for this sort of problem. As I said, I am not a statistician. However, I see this stuff all the time in science - it is really easy to spot a pattern and think that it means something.
No disagreements about the reality of how hypothesis are actually formed.
However, this is being presented as "scientists discovered something" when your argument amounts to "scientists found some data that suggests that in the future they may or may not discover something."
I agree that bumps in noise are how a lot of science starts out. However, that is the beginning of science, and not the end.
I've got my own mail server, and it forwards my mail to gmail. I've been experimenting with going more cloud-based, and right now gmail is the best option for this out there. The next closest options I've found are Roundcube and Zimbra, and they're not nearly as nice.
All my mail is safely stored on my own servers, and a copy is sent to gmail. So I'm safe if they ever go under. However, I'm in no way protected from anybody with access to Google's servers.
What we really need are open source cloud solutions. There are very few out there now and they're generally weak.
When I read things like "In about 250 times more cases than expected, the total energy of the jets clustered around a value of about 144 billion electron volts" I get nervous.
This is like saying that in a series of 1M coin tosses the sequence HTTHHTTTTHHTTHHH came up 100x more often than would be expected by chance. Does that mean that any particular sequence of 8 tosses should come up 1/65536th of the time, and this one came up 1/655th of the time, or does it mean that some random sequence of results should come up 100% of the time in a random series of 16 coin tosses, and we happened to pick the random series that came up the most often in that particular set of data?
If I mine a big set of data against 100 random hypothesis I'll be able to find about 5 that I can show to be true with 95% confidence, despite the fact that there is nothing really going on.
The real test is to come up with the hypothesis first, then collect the data.
Now, these guys are probably smart, and hopefully control for this. If you want to test for 100 hypotheses and REALLY have 95% confidence, then you need to target a confidence of 1-0.05^100 for each test - at least that is how I see it (being a complete novice at statistics).
When it comes to software development, spending more doesn't necessarily get you more.
According to your own post it got the guy responsible a promotion. How is that not getting you more? I deliver good solutions all the time and I rarely get promoted.
People make the mistake of viewing things from the perspective of "the company" - as if "the company" had a brain and a mouth. Companies are run by managers, who often make decisions that have exactly the results they were planned to have - the betterment of the manager.
Yeah, but they make such nice powerpoint presentations. Everybody who hires consultants gets big bonuses, because their slides look better than those generated by their peers...
As long as executive leadership is swayed by pretty pictures, and senior managers are forbidden from hiring graphics artists, they will instead hire consultants to do the work of graphics artists at 10X the cost.
Yup - at work we have documented processes, and nobody ever wants to point out a problem with them, because to fix one line in a process requires going through a horrible process as you re-negotiate the other 499 lines in the process with all the people who fought about them the first time they got approved, but without the help of whatever VP rammed the process through the first time.
As a result we either ignore our documented processes, or just follow non-optimal ones. Fixing them is just not worth the effort.
And of course, the guy who speaks up gets the action item...
Certainly true where I work. Typically senior management takes the first six months of the year to figure out our strategic plans for the year (during which time half of the general staff idles), then projects launch and start out in panic mode to meet EOY deadlines. Of course, EOY means in time for year-end ratings which is something like Oct-Nov, and everybody is gone in Dec. So, we spend about 5 months of the year doing productive work, and senior management spends the whole year locked in a conference room.
Oh, and you can't change a light bulb without a business case, near-VP approval, a project plan, and 10 people each doing exactly one part of the project while trying as hard as possible not to collaborate. That is, unless you're one of those groups that just breaks all the rules, in which case you could probably just run a whole CRM system out of a 850MB access database on a network share.
What are you going to do - not do business with any of the 100 companies that were compromised? All of their competitors were compromised as well.
It is like complaining about SMS prices on US cell carriers - as long as everybody offers lousy service and the FTC refuses to regulate, customers get to choose between various levels of crappiness...
He could require GPL v2+ on all new patches, so that at least the 2+-compatible codebase grows over time - thus fixing itself eventually once v2-only code is phased out, or becomes small enough to be replaceable or get a few author approvals here and there.
I'd say dozens of distros is the reason linux is as successful as it is. Imagine of slackware were your only choice, and things like RHEL, Arch, Debian, Gentoo, or Ubuntu were legally forbidden?
The issue with android isn't that it is fragmenting, but rather that carriers make forks and then prevent customers from having a choice. What Android really needs is GPLv3.
Hmm, I run linux so I don't use the installer all that much (just on windows installs).
However, it does duplicate half the so's on the system, which obviously can't help its memory use.
For whatever reason it feels like it isn't using a ton of RAM. Perhaps firefox is just that bad on my install. The biggest pain I've had with firefox is that for whatever reason a delay in loading one tab can cause all the other tabs not to load anything either.
I agree that cars will need places to park. However, as you said the parking doesn't need to be within walking distance of the destination. Fewer constraints on a solution almost always results in a solution that is better optimized to meet other goals.
Traditional parking lots use huge amounts of ground surface area, which results in less green space in general, more surface heating (albedo of asphalt), and more water runoff (which is full of oil and stuff like that). We make up for that with big retention basins and stuff like that, which wastes space as well.
Probably a better model would be parking garages and things like that. They waste far less space and could be located anywhere. They could meet the need for both commercial and residential parking as well. Maybe there are other models that are even better - I'm not a traffic engineer.
I do agree that it might result in more mileage and road utilization, but there is obviously a balance between every store having 20,000 square feet of asphalt in front of it and driving 15 miles to a skyscraper-sized parking garage.
Also, automated cars could have much higher utilization, which means fewer cars to park. In fact, on-demand rental might actually work out as a better model than ownership when a car can be at your door 3 minutes after you call for one. On-demand rental could mean that cars wouldn't even need to park much at all during the day. That would require solutions for the fact that many cars double as portable storage lockers. That isn't an unsolvable problem however.
Again, I'm not talking about now. I'm talking about years from now - the long-term. I'm not saying that nobody will buy mp3 players next Tuesday, or a year from next Tuesday. I'll be surprised if they're all that big five years from now, however.
The whole reason I switched to chrome was that it fixes many of these fundamental flaws. It is a lot more like what Firefox used to be - lean and efficient. Granted, they're adding on the features as well, but at least tabs are self-contained and it doesn't suck down RAM by the hundreds of MB.
You can already buy them decoupled from data plans - the selection is just very limited and the cost is $550. You can get them used for under $100, but generally only for old models.
In a couple of years, smartphones will be like digital cameras - $90 at walmart will get you a decent one. Data plans will eventually get cheaper as well.
I'm not talking about next Tuesday or anything. Think long term.
In any case, why would you buy an mp3 player when you can just use your cell phone to do the same thing and have one less device to carry. I could see there being some market for a small dedicated device, but it will no longer be mainstream...
Slashdot Mirror
That is a bit of a crime - at those price points.
I just looked over on my bookcase, and indeed there is the manual for my HP48G. It is about an inch thick or so - small type. The irony is that I probably paid the same price for it as people pay now, so it isn't like they need to save money on paper.
Then again, the same was true of MS Office, Operating Systems, and everything else back in the 80s and early 90s. Expectations have changed.
Uh, the students are in fact required to buy certain models. Most classes restrict what kinds of calculators you're allowed to use on exams. Most would prohibit the use of netbooks.
So, the customer's ability to get what they really want is limited.
It looks like it discharges cleaner water into the ocean. That might make it illegal in the US, due to some bizarre interpretation of EPA regulations.
I was reading that European and Middle-East cleanup ships were turned away from helping with cleanup. This was due to their principle of operation - they would skim muck from the surface, separate the oil and water, and discharge the water back into the ocean, keeping the oil in a holding tank. The problem is that they didn't clean the water they discharged up to EPA standards. Cleanup ships were required to store the mostly-clean water they would otherwise discharge, which means they would have to make frequent trips to dump their tanks.
This is of course absurd, since a ship that takes in a 50% oil solution and outputs a 0.1% oil solution can operate indefinitely and only make things better. The regulation was meant to apply to devices that prevent oil spills in the first place (dumping nothing into a clean ocean is better than dumping 0.1% oil solution).
Utterly amazing. Clearly regulation is necessary (otherwise there is no incentive to not pollute in the first place), but such strict application of law is something right out of Kefka or Wikipedia... :)
This wasn't a new clinical trial as far as I can tell, but it was mining past trials for data and applying them to new hypothesis. A recent xkcd showed some of the problems with this kind of argument. If the data didn't support permanent loss of sexual function, perhaps it would support heart attacks, cancer, blindness, or perhaps cures for any of the above.
If you think this is a possible side-effect, then hypothesize that, conduct a large trial, and see what the results are.
Sure, mining past data can generate ideas for new trials, but it is not a substitute for them. It is far too susceptible to finding patterns that aren't really there - especially with the amount of noise in a typical clinical trial dataset.
What we really need are open source cloud solutions. There are very few out there now and they're generally weak.
How does open source software help if you still hand your plain text data to a third party running the service? The only way to protect data that you want an untrustworthy third party to hold for you is to encrypt it such that they do not have access. Just encrypt your messages before sending them to Google.
If there were an open-source equivalent to Gmail I wouldn't have to send my mail to anybody - I'd run my own cloud solution on my own hardware under my own control.
Short of that, there is no point in encrypting email and sending it to Google, unless I just want to use them as a backup solution. I couldn't read the emails via Gmail without some kind of browser plugin to decrypt them again, which defeats most of the purpose of having them in the cloud in the first place.
For backups, encrypting before uploading makes a lot more sense - as long as you don't need easy remote access to the backups.
the idea of building a phone for the gigantic foreign market never hit them, apparently.
That is pretty typical Japanese culture. Foreign markets are usually of secondary concern to them. The only reason they were as successful as they were was that American manufacturers were SO reluctant to change that even an unenthusiastic foreign corporation could clean house.
Why would anyone use that? Won't it get finger prints all over it?
I know somebody who works in Pharmaceuticals and they had to buy special equipment to make pills for the Japanese market. To be accepted by a Japanese consumer, a pill needs to be perfectly white without any sign of blemish or visible imperfection. While this is a general concern in any manufacturing process, an American won't ask their doctor to prescribe something different if they find one pill with a ding or a slight discoloration in the bottle.
Yup, the reality is that little that is done in major companies is done primarily for the company's benefit. What happens, happens because it will make a manager somewhere better off in some way (financially, psychologically, whatever). To the extent that company well-being is actually tied to rewards that may happen to also benefit the company, but that is usually a secondary consideration.
When some manager comes up with some crazy mandate that everybody knows will hurt the company, do they all raise a big protest? No way! They all get in there and write up at the end of the year how enthusiastically they implemented it so that they can collect their annual bonus. Then, when the next manager enthusiastically reverses the policy of the first one, everybody enacts that policy with equal vigor. Being bad for a company will never slow down a big initiative. Now, if the big initiative is bad for individual employees, then it will meet with resistance regardless of impact on the company.
Right, which doesn't prove any new hypothesis - it merely disproves their original hypothesis. Likely at best they can show the standard model is wrong, and that is if there is no systematic source of error, as a statistical test only accounts for random error.
Don't get me wrong - I'm interested in where this goes. However, this is just the start of a long process.
And again, I never claimed that the people doing the test didn't understand statistics, or that they hadn't accounted for this sort of problem. As I said, I am not a statistician. However, I see this stuff all the time in science - it is really easy to spot a pattern and think that it means something.
No disagreements about the reality of how hypothesis are actually formed.
However, this is being presented as "scientists discovered something" when your argument amounts to "scientists found some data that suggests that in the future they may or may not discover something."
I agree that bumps in noise are how a lot of science starts out. However, that is the beginning of science, and not the end.
Actually, as has become high-profile in recent months, they call that medicine, including drug discovery.
I've got my own mail server, and it forwards my mail to gmail. I've been experimenting with going more cloud-based, and right now gmail is the best option for this out there. The next closest options I've found are Roundcube and Zimbra, and they're not nearly as nice.
All my mail is safely stored on my own servers, and a copy is sent to gmail. So I'm safe if they ever go under. However, I'm in no way protected from anybody with access to Google's servers.
What we really need are open source cloud solutions. There are very few out there now and they're generally weak.
When I read things like "In about 250 times more cases than expected, the total energy of the jets clustered around a value of about 144 billion electron volts" I get nervous.
This is like saying that in a series of 1M coin tosses the sequence HTTHHTTTTHHTTHHH came up 100x more often than would be expected by chance. Does that mean that any particular sequence of 8 tosses should come up 1/65536th of the time, and this one came up 1/655th of the time, or does it mean that some random sequence of results should come up 100% of the time in a random series of 16 coin tosses, and we happened to pick the random series that came up the most often in that particular set of data?
If I mine a big set of data against 100 random hypothesis I'll be able to find about 5 that I can show to be true with 95% confidence, despite the fact that there is nothing really going on.
The real test is to come up with the hypothesis first, then collect the data.
Now, these guys are probably smart, and hopefully control for this. If you want to test for 100 hypotheses and REALLY have 95% confidence, then you need to target a confidence of 1-0.05^100 for each test - at least that is how I see it (being a complete novice at statistics).
When it comes to software development, spending more doesn't necessarily get you more.
According to your own post it got the guy responsible a promotion. How is that not getting you more? I deliver good solutions all the time and I rarely get promoted.
People make the mistake of viewing things from the perspective of "the company" - as if "the company" had a brain and a mouth. Companies are run by managers, who often make decisions that have exactly the results they were planned to have - the betterment of the manager.
Yeah, but they make such nice powerpoint presentations. Everybody who hires consultants gets big bonuses, because their slides look better than those generated by their peers...
As long as executive leadership is swayed by pretty pictures, and senior managers are forbidden from hiring graphics artists, they will instead hire consultants to do the work of graphics artists at 10X the cost.
Yup - at work we have documented processes, and nobody ever wants to point out a problem with them, because to fix one line in a process requires going through a horrible process as you re-negotiate the other 499 lines in the process with all the people who fought about them the first time they got approved, but without the help of whatever VP rammed the process through the first time.
As a result we either ignore our documented processes, or just follow non-optimal ones. Fixing them is just not worth the effort.
And of course, the guy who speaks up gets the action item...
Certainly true where I work. Typically senior management takes the first six months of the year to figure out our strategic plans for the year (during which time half of the general staff idles), then projects launch and start out in panic mode to meet EOY deadlines. Of course, EOY means in time for year-end ratings which is something like Oct-Nov, and everybody is gone in Dec. So, we spend about 5 months of the year doing productive work, and senior management spends the whole year locked in a conference room.
Oh, and you can't change a light bulb without a business case, near-VP approval, a project plan, and 10 people each doing exactly one part of the project while trying as hard as possible not to collaborate. That is, unless you're one of those groups that just breaks all the rules, in which case you could probably just run a whole CRM system out of a 850MB access database on a network share.
Simple - there is no reason not to.
What are you going to do - not do business with any of the 100 companies that were compromised? All of their competitors were compromised as well.
It is like complaining about SMS prices on US cell carriers - as long as everybody offers lousy service and the FTC refuses to regulate, customers get to choose between various levels of crappiness...
He could require GPL v2+ on all new patches, so that at least the 2+-compatible codebase grows over time - thus fixing itself eventually once v2-only code is phased out, or becomes small enough to be replaceable or get a few author approvals here and there.
I'd say dozens of distros is the reason linux is as successful as it is. Imagine of slackware were your only choice, and things like RHEL, Arch, Debian, Gentoo, or Ubuntu were legally forbidden?
The issue with android isn't that it is fragmenting, but rather that carriers make forks and then prevent customers from having a choice. What Android really needs is GPLv3.
Well, it probably doesn't legally void the warranty, but it will take you a few years in court to settle that one...
Hmm, I run linux so I don't use the installer all that much (just on windows installs).
However, it does duplicate half the so's on the system, which obviously can't help its memory use.
For whatever reason it feels like it isn't using a ton of RAM. Perhaps firefox is just that bad on my install. The biggest pain I've had with firefox is that for whatever reason a delay in loading one tab can cause all the other tabs not to load anything either.
I agree that cars will need places to park. However, as you said the parking doesn't need to be within walking distance of the destination. Fewer constraints on a solution almost always results in a solution that is better optimized to meet other goals.
Traditional parking lots use huge amounts of ground surface area, which results in less green space in general, more surface heating (albedo of asphalt), and more water runoff (which is full of oil and stuff like that). We make up for that with big retention basins and stuff like that, which wastes space as well.
Probably a better model would be parking garages and things like that. They waste far less space and could be located anywhere. They could meet the need for both commercial and residential parking as well. Maybe there are other models that are even better - I'm not a traffic engineer.
I do agree that it might result in more mileage and road utilization, but there is obviously a balance between every store having 20,000 square feet of asphalt in front of it and driving 15 miles to a skyscraper-sized parking garage.
Also, automated cars could have much higher utilization, which means fewer cars to park. In fact, on-demand rental might actually work out as a better model than ownership when a car can be at your door 3 minutes after you call for one. On-demand rental could mean that cars wouldn't even need to park much at all during the day. That would require solutions for the fact that many cars double as portable storage lockers. That isn't an unsolvable problem however.
Again, I'm not talking about now. I'm talking about years from now - the long-term. I'm not saying that nobody will buy mp3 players next Tuesday, or a year from next Tuesday. I'll be surprised if they're all that big five years from now, however.
The whole reason I switched to chrome was that it fixes many of these fundamental flaws. It is a lot more like what Firefox used to be - lean and efficient. Granted, they're adding on the features as well, but at least tabs are self-contained and it doesn't suck down RAM by the hundreds of MB.
You can already buy them decoupled from data plans - the selection is just very limited and the cost is $550. You can get them used for under $100, but generally only for old models.
In a couple of years, smartphones will be like digital cameras - $90 at walmart will get you a decent one. Data plans will eventually get cheaper as well.
I'm not talking about next Tuesday or anything. Think long term.
In any case, why would you buy an mp3 player when you can just use your cell phone to do the same thing and have one less device to carry. I could see there being some market for a small dedicated device, but it will no longer be mainstream...