I agree that objectivity is not impossible, and that it's a desirable goal. However you made it sound a little too simple - just report the facts.
What about context? The kind of context given (which is almost always "facts") shows the agenda of the reporter.
Adam punches Bob in the nose. If we learn that Adam frequently punches people for fun, how does that affect our perception? What if learn that Bob had smashed the windows of Adam's car last week?
By reporting one contextual fact or the other, the journalist changes our perception of who "caused" the problem. And this really happens!
That's why the internet is having a huge effect - we can now search for a person or organization and see where they came up in the past.
Generally, NSA doesn't feed info to law enforcement. Their domain is national security. However I have no idea how the war on terror has affected this boundary.
NSA is not allowed to knowingly listen to US citizens inside the US. Also, disclosing intercepts would jeopardize technical sources that are more important than a criminal conviction.
In other words, your idea is good; all signals intelligence agencies have already thought it through; therefore the idea is no longer good. Stephenson's Cryptonomicon has a fascinating subplot on this topic.
Actually, AutoCAD grew from a DOS mentality - they just happened to get it very right. The Unix ports were always secondary. Now, John Walker was very impressed by X on Unix and wanted to bring similar windowing power to DOS. So when Microsoft released Windows 1.0, Autodesk was an early and enthusiastic adopter. They never expressed awareness that they were moving away from a truly great UI - in truth, most of the good things remain in the Windows version.
When I've watched mouse-oriented AutoCAD drafters, they are definitely slower than keyboard users. At least the ones I've seen have 2-5 seconds between each of those clicks. Remember, each click requires at least positioning over a pretty small icon - sometimes going down one level in a menu.
Comments highlight a blind spot of geeks
on
The Evil in E-Mail
·
· Score: 1
Many of the slashdot comments here were completely predictable and reflect a mental problem of geeks - a near-autism that wants things to be boolean rather than analog. I've seen this tendency on many other stories, and I think it is probably hurting some of you in the workplace or school. One poster worries that so many emails will be flagged that we'll create a huge pool of federal employees to read them. Others think they'll be unjustly accused of crimes due to some "false positive".
This kind of system generally calculates one or more scores. Common sense tells me that messages with scores exceeding a threshold would be manually inspected. Weighting factors and thresholds would be adjusted over time to tune the system.
This is how many things work presently. Anti-spam, anti-abuse, anti-fraud. You triage the stream of transactions into black, gray and white. The gray should be the smallest band by far, and is inspected by humans. Those humans tune the system to better handle the ambiguity.
If you took any widely used heuristic tagging system like this and attempted to explain it on slashdot, it would be met with the same torrent of near-autistic objections. Essentially, "it can't work and shouldn't work because it's not 100% perfect." But the real world runs on many tools that are not 100%. Because the alternative is manual inspection of every single transaction, which is cost-prohibitive.
Maybe statistics is the key mental tool for geek-autists to grapple with ambiguity. Statistics can help us explore correlation between A and B without asserting rigid linkage between A and B.
The main package he's maintaining is xscreensaver. I don't think xscreensaver can work on the Mac - it's not just an X application (those do work on the Mac) but a framework that plugs rather intimately into X to take over the display and run a screensaver. It only works on desktop Unix running X.
Jamie apparently plans to have headless Linux boxes, but without a desktop X workstation there is no place to use or test xscreensaver, and probably no interest in doing so.
The majority of the "hacks" - that is actual screensavers - were written by others. Jamie is primarily responsible for the magic that lets these programs work as screensavers.
I see both sides of this one. The common problem in Linux is that you try to do something fairly common and obvious and it doesn't work. The answer is to install this or upgrade that - something completely non-obvious. When you get that answer, especially if it took a lot of tries and research, there's a certain irritation. You tend to think, "Apparently this state of affairs is just fine with everyone, right?" Simple stuff, obvious stuff, doesn't work, and you have to spend time digging up secret recipes. How much time is wasted by thousands of people digging up this info in parallel?
The problem is not with the software; it's more with the distro packaging. There's a total lack of ownership. Nobody seems to have looked at the whole distro and made sure it worked for even the most common cases.
Case in point: on Red Hat, lynx tries to use xli to display images. Xli is not included in Red Hat (as far as I can tell) so this causes an error message until you fix it. This is not lynx's fault - it's Red Hat's fault.
I love using Linux, but I hate configuring it. Those are completely wasted hours I could have used to develop something new and interesting.
I see your point about exceptions. But as for the C++ spec not requiring thread-safeness - what level of thread-safeness does java provide? Does it automagically lock container classes? If so, couldn't this be an unnecessary performance impediment in some cases - for example when an operation is defined to affect two containers, or the same container twice, and the C++ programmer might use one lock over the entire operation?
And not just Java, but Swing. It is a wondefully powerful language, just like C, and it has native threading, exceptions, and class extensions, so it blows C++ out of the water.
Exceptions - C++ has them.
Threading - why should it be built into the language? Why not build in TCP/IP and graphics while you're at it? (Yes I know - to meet its cross-platform goals, java has to abstract threading. But I don't see this as a datapoint blowing C++ out of the water.)
Well, you're a step closer to reason than the person you're responding to. But what makes anyone posting on slashdot think they're qualified to advise the military on setting up roadblocks?
Chances are the military is doing exactly the right thing given the circumstances.
It is obvious that the military leadership is poor.
Not to me. Are you basing that impression on our media's tendency to highlight every screw-up?
I think the only area where the military needs advice from slashdotters is IT. For example, knowing what a PDF is and isn't.
Anyone who infringes a patent is liable, even if he bought an infringing item or method from a vendor. The vendor is liable for contributory infringement. In practice, the inventor usually prefers to sue the vendor rather than pursue a huge number of customers whose liability will be quite small.
For an example of the patent holder pursuing the end user, look up the Solaia case.
By calling these machines "robots", we imply that they're autonomous and mobile. But from the article, they will be placed every kilometer along the border. And there is no mention of autonomy.
These are more like remote-controlled guns with video cameras. Maybe they'll be mounted on telephone poles. No new technology really needed here - it could be built with 1970's tech.
The driver here is not to have a more effective border patrol, but to free up lots of troops so they can spend time training rather than guarding the border.
I agree that we have a problem with obvious patents being granted. But why "small"?
If someone discovers how to make an oil refinery 0.1% more efficient by sticking a brass wire into a certain pipe, should that be unpatentable because it's "small"?
I've read a few software patents, and they don't look anything like pure math. They are usually recipes for accomplishing a specific improvement in a specific kind of software.
I have seen software patents that shouldn't have been issued, in my opinion. But it wasn't because they were "pure math". It was, sometimes, because the claims were too broadly worded, covering prior art. Other times, the invention was obvious.
The "pure math" argument is illogical. Many electronic circuit inventions are essentially data processors, and can be simulated in a computer. If someone spends a year of research, and develops a huge PC board, a machine vision module that takes the output of a video camera and senses whether a human being is present, is that patentable? What if he reduces the board to a single FPGA? Still patentable? What if he turns the method into software?
Smedley, I have scrutinized this radar screen and see absolutely no penguins!
Umm, yes sir, the thing about this radar screen you're holding...
No evasions, Smedley! You told me Linux was the next big thing, but my radar shows no trace of it!
But sir, the thing is, this radar screen is actually an etch-a-sketch. Remember, we gave it to you after you went through three notebooks in a week. If you want Linux to show up, just twiddle those knobs...
Big picture, Smedley! No implementation details! Linux is NOT showing up on the radar!
So cautious that they're all but useless. If they blacklisted your colo facility, it was probably after a very long period of fruitless negotiations. Odds are the facility should have been blacklisted years ago.
It's interesting that you don't choose to tell us the affected netblock. Is it by any chance a notorious hellhole vomiting spam into everyone's inbox?
You complain that MAPS wasn't around on the weekend waiting for your call. Tell me, are the spammers in your netblock manning the phones every weekend for complaints?
It would be amusing to see a few heads exploding around here as people see Linus working for a "pure-IP" company. Of course there's no real contradiction - Linus believes in IP.
I think a lot of slashdotters haven't faced up to the fact that IP makes the tech industry possible.
If they would like to become a research company working for others to develop tech , then fair enough but not an IP company
So it's OK with you if they develop new technology; they just can't own it? In which case, how will they make money? Or are you saying they must sell off their IP as fast as they generate it?
Slashdot Mirror
I agree that objectivity is not impossible, and that it's a desirable goal. However you made it sound a little too simple - just report the facts.
What about context? The kind of context given (which is almost always "facts") shows the agenda of the reporter.
Adam punches Bob in the nose. If we learn that Adam frequently punches people for fun, how does that affect our perception? What if learn that Bob had smashed the windows of Adam's car last week?
By reporting one contextual fact or the other, the journalist changes our perception of who "caused" the problem. And this really happens!
That's why the internet is having a huge effect - we can now search for a person or organization and see where they came up in the past.
Generally, NSA doesn't feed info to law enforcement. Their domain is national security. However I have no idea how the war on terror has affected this boundary.
NSA is not allowed to knowingly listen to US citizens inside the US. Also, disclosing intercepts would jeopardize technical sources that are more important than a criminal conviction.
In other words, your idea is good; all signals intelligence agencies have already thought it through; therefore the idea is no longer good. Stephenson's Cryptonomicon has a fascinating subplot on this topic.
Actually, AutoCAD grew from a DOS mentality - they just happened to get it very right. The Unix ports were always secondary. Now, John Walker was very impressed by X on Unix and wanted to bring similar windowing power to DOS. So when Microsoft released Windows 1.0, Autodesk was an early and enthusiastic adopter. They never expressed awareness that they were moving away from a truly great UI - in truth, most of the good things remain in the Windows version.
When I've watched mouse-oriented AutoCAD drafters, they are definitely slower than keyboard users. At least the ones I've seen have 2-5 seconds between each of those clicks. Remember, each click requires at least positioning over a pretty small icon - sometimes going down one level in a menu.
Many of the slashdot comments here were completely predictable and reflect a mental problem of geeks - a near-autism that wants things to be boolean rather than analog. I've seen this tendency on many other stories, and I think it is probably hurting some of you in the workplace or school. One poster worries that so many emails will be flagged that we'll create a huge pool of federal employees to read them. Others think they'll be unjustly accused of crimes due to some "false positive".
This kind of system generally calculates one or more scores. Common sense tells me that messages with scores exceeding a threshold would be manually inspected. Weighting factors and thresholds would be adjusted over time to tune the system.
This is how many things work presently. Anti-spam, anti-abuse, anti-fraud. You triage the stream of transactions into black, gray and white. The gray should be the smallest band by far, and is inspected by humans. Those humans tune the system to better handle the ambiguity.
If you took any widely used heuristic tagging system like this and attempted to explain it on slashdot, it would be met with the same torrent of near-autistic objections. Essentially, "it can't work and shouldn't work because it's not 100% perfect." But the real world runs on many tools that are not 100%. Because the alternative is manual inspection of every single transaction, which is cost-prohibitive.
Maybe statistics is the key mental tool for geek-autists to grapple with ambiguity. Statistics can help us explore correlation between A and B without asserting rigid linkage between A and B.
The main package he's maintaining is xscreensaver. I don't think xscreensaver can work on the Mac - it's not just an X application (those do work on the Mac) but a framework that plugs rather intimately into X to take over the display and run a screensaver. It only works on desktop Unix running X.
Jamie apparently plans to have headless Linux boxes, but without a desktop X workstation there is no place to use or test xscreensaver, and probably no interest in doing so.
The majority of the "hacks" - that is actual screensavers - were written by others. Jamie is primarily responsible for the magic that lets these programs work as screensavers.
I see both sides of this one. The common problem in Linux is that you try to do something fairly common and obvious and it doesn't work. The answer is to install this or upgrade that - something completely non-obvious. When you get that answer, especially if it took a lot of tries and research, there's a certain irritation. You tend to think, "Apparently this state of affairs is just fine with everyone, right?" Simple stuff, obvious stuff, doesn't work, and you have to spend time digging up secret recipes. How much time is wasted by thousands of people digging up this info in parallel?
The problem is not with the software; it's more with the distro packaging. There's a total lack of ownership. Nobody seems to have looked at the whole distro and made sure it worked for even the most common cases.
Case in point: on Red Hat, lynx tries to use xli to display images. Xli is not included in Red Hat (as far as I can tell) so this causes an error message until you fix it. This is not lynx's fault - it's Red Hat's fault.
I love using Linux, but I hate configuring it. Those are completely wasted hours I could have used to develop something new and interesting.
There's the problem. They should have operated it in an off-again, on-again fashion. Then when it finally got stuck it would be on.
But was it a rectangular thing daubed with Rastafarian symbols?
I see your point about exceptions. But as for the C++ spec not requiring thread-safeness - what level of thread-safeness does java provide? Does it automagically lock container classes? If so, couldn't this be an unnecessary performance impediment in some cases - for example when an operation is defined to affect two containers, or the same container twice, and the C++ programmer might use one lock over the entire operation?
Exceptions - C++ has them.
Threading - why should it be built into the language? Why not build in TCP/IP and graphics while you're at it? (Yes I know - to meet its cross-platform goals, java has to abstract threading. But I don't see this as a datapoint blowing C++ out of the water.)
Sgrena claimed the soldiers fired 300-400 bullets into the car. Here are pictures of the car.
Chances are the military is doing exactly the right thing given the circumstances.
Not to me. Are you basing that impression on our media's tendency to highlight every screw-up?
I think the only area where the military needs advice from slashdotters is IT. For example, knowing what a PDF is and isn't.
Just give him a few chickens and he'll be happy.
That a democratically elected government, when picking delegates to an international body, would pick those that represent the voters' views.
It's a slippery slope! Next thing you know, they'll be sending some pro-America guy to the U.N.!
Anyone who infringes a patent is liable, even if he bought an infringing item or method from a vendor. The vendor is liable for contributory infringement. In practice, the inventor usually prefers to sue the vendor rather than pursue a huge number of customers whose liability will be quite small.
For an example of the patent holder pursuing the end user, look up the Solaia case.
Why? Why do you think the VHDL (FPGA description) should be patentable, but not equivalent C code?
And don't say "pure math". The VHDL is just as much pure math as the C.
By calling these machines "robots", we imply that they're autonomous and mobile. But from the article, they will be placed every kilometer along the border. And there is no mention of autonomy.
These are more like remote-controlled guns with video cameras. Maybe they'll be mounted on telephone poles. No new technology really needed here - it could be built with 1970's tech.
The driver here is not to have a more effective border patrol, but to free up lots of troops so they can spend time training rather than guarding the border.
I agree that we have a problem with obvious patents being granted. But why "small"?
If someone discovers how to make an oil refinery 0.1% more efficient by sticking a brass wire into a certain pipe, should that be unpatentable because it's "small"?
I've read a few software patents, and they don't look anything like pure math. They are usually recipes for accomplishing a specific improvement in a specific kind of software.
I have seen software patents that shouldn't have been issued, in my opinion. But it wasn't because they were "pure math". It was, sometimes, because the claims were too broadly worded, covering prior art. Other times, the invention was obvious.
The "pure math" argument is illogical. Many electronic circuit inventions are essentially data processors, and can be simulated in a computer. If someone spends a year of research, and develops a huge PC board, a machine vision module that takes the output of a video camera and senses whether a human being is present, is that patentable? What if he reduces the board to a single FPGA? Still patentable? What if he turns the method into software?
Smedley, I have scrutinized this radar screen and see absolutely no penguins!
...
...
Umm, yes sir, the thing about this radar screen you're holding
No evasions, Smedley! You told me Linux was the next big thing, but my radar shows no trace of it!
But sir, the thing is, this radar screen is actually an etch-a-sketch. Remember, we gave it to you after you went through three notebooks in a week. If you want Linux to show up, just twiddle those knobs
Big picture, Smedley! No implementation details! Linux is NOT showing up on the radar!
John Kerry didn't get his wealth from inheritance. Well, not directly.
Yes, but how can we PWRTFA unless advised of TFA?
So cautious that they're all but useless. If they blacklisted your colo facility, it was probably after a very long period of fruitless negotiations. Odds are the facility should have been blacklisted years ago.
It's interesting that you don't choose to tell us the affected netblock. Is it by any chance a notorious hellhole vomiting spam into everyone's inbox?
You complain that MAPS wasn't around on the weekend waiting for your call. Tell me, are the spammers in your netblock manning the phones every weekend for complaints?
It would be amusing to see a few heads exploding around here as people see Linus working for a "pure-IP" company. Of course there's no real contradiction - Linus believes in IP.
I think a lot of slashdotters haven't faced up to the fact that IP makes the tech industry possible.
So it's OK with you if they develop new technology; they just can't own it? In which case, how will they make money? Or are you saying they must sell off their IP as fast as they generate it?