What bugs me is the stuff they didn't fix. In Empire, when Han is about to be frozen in carbonite, his jacket keeps vanishing and reappearing--a neat trick given that he's handcuffed. I find it the most distracting continuity error in all of film history. Why couldn't they fix that?
I know I am in the minority of slashdoters here but I think that HP is being ethical and responsible in their efforts to protect currency from unauthorized duplication.
I suspect HP isn't doing this out of any sense of right and wrong. I'd wager that they're afraid if they don't do something to curb counterfeiting voluntary, then the government may require them to do something less to their liking. It's the same idea with movie and TV ratings. The industry would rather self-regulate and retain some control than bow down to the will of the lawmakers.
Crooks can walk into any computer store and buy a box of blank checks and print out whatever they want on the checks including whatever routing number and account number they want.... Why hasn't there been a [huge outcry] over this?
An excellent question. The answer: because the government cares about counterfeiting currency, but only banks care about counterfeiting checks. This is consistent with HP acting out of a fear of government regulation as opposed to good corporate citizenship.
Having read code from many other programmers and having interviewed scores more, I've found that the best programmers are those who learned simpler, higher-level languages before others.
If Jessica learned Pascal, then C, then assembly, I'd bet she's pretty good at all three. If Samantha learned the same languages in the opposite order, chances are she's not as good at any of them.
You may disagree, but it has certainly proven true in my experience. My impression is that the folks who learn low-level first get bogged down in the details and have a harder time finding the elegant (read "simple") solutions. Once the problem is decomposed, experience with lower level languages translates into tools for building a clean implementation of the solution.
On the 3rd of February 2004, this page (or rather the page that was here) was swamped by requests and the server subsequentially failed.
This seems to imply that the server crashed. Maybe I'm misreading it in this case, but I've certainly heard other claims that a slashdotting or googleblatting or DoS attack crashed servers.
When a server is saturated with requests, it should reply to as many of them as it can and allow the rest to time out. If the server crashes as a result of the load, that's a bug.
Perhaps it's just a semantic thing, but it seems misleading to blame the crash on the load. The crash is due to a bug that happens to be aggravated by high load. Blaming the load seems to be a way to deflect responsibility from the engineers who built the server.
Furthermore, a glance at an analog speedometer gives you a feel for rate of acceleration (of deceleration) that's really hard to get from a changing numeric display.
I got a cool Logitech gizmo for Christmas that I still haven't been able to use, because it requires the.NET runtime which simply won't install on my PC. I just get a generic "Set up failed" message. Microsoft Customer Support doesn't reply. Logitech Support--after six requests for help--finally offered a few things to try, none of which worked.
Relying on Microsoft to install a new technology on older machines puts Logitech at a real disadvantage. They should have built a traditional app until.NET is more stable.
If you're handing your license to an officer you're way beond anonymity. Your best hope at that point is to keep a low profile.
True, but other people want access to the data on the mag stripe as well. Often, when I write a check at a store, the clerk asks to swipe my license. I always refuse. I tell them that they're welcome to copy my name and ID number from the license, but I'll not make it easy for them to grab my address and other info for snail-mail spamming, data reporting, etc.
It would be less confrontational if I could just have a "defective" mag stripe.
Furthermore, when California first put the stripes on the licenses, there was a bit of a controversy. Two of the tracks are available to standard readers and contain nothing more than what's printed on the front of your license. But there's a third track, which is recorded at a much higher density and in a format that was incompatible with commonly available readers. The DMV refused to disclose what information they were recording in this third track.
My mistake. The laws I heard about are only for larger companies that have affirmative action plans (compulsory or voluntary) and/or H1-B visa applicants.
Take for example the infamous Section 215 that civil libertarians claim allows law enforcement to search your library records. Except this power requires the consent of a federal judge, no library records have ever been searched, and such provisions have already been used in other criminal cases. Library records were searched in the hunt for Andrew Cunanan, the man who shot fashion designer Gianni Versace in 1997, and to hunt down the Zodiac killer in New York in 1990. Yet no one raised a fuss about these searches. It is clear that there is a direct double standard at play, fueled by ignorance of the law.
Let me say up front that I'm not a lawyer. But the biggest problem I see here is that there is very little if any oversight. Traditional search warrants are (or become) public record, making it possible for people to check for abuse. For example, in California, after a wiretap is completed, law enforcement must contact every party that was heard on the line to let them know they had been recorded. With the gag rules in the PATRIOT ACT, there's no after-the-fact oversight to make sure the judge who granted the request was doing the right thing and that the enforcement agencies aren't routinely asking for wide-reaching powers. You say that, "no library records have ever been searched," but you don't know that because of the gag rules.
I hope Joel is using the comments about deleting and shredding resumes figuratively. There are laws that require employers to keep resumes and applications on file for some period of time after they are received.
A friend of mine in college in the late 80's did an internship with Disney Imagineering. At the time, they were redrawing plans for several attractions that were to be copied almost exactly from California and Florida version for use in Euro Disney. There was great concern that the Imperial to metric round-off would be a problem. The contractor's union in France mandated that they only use metric units in the blueprints, which is reasonable. But if you're trying to duplicate a ride that was originally designed in Imperial units, you need to keep in mind quite a few significant digits when you're measuring. For example, a section of track in a given ride might be 10 feet in the US. In metric that's 3.048 m. Would the French contractors really measure to that precision? Or would they round off to 3.05 or event 3.0? They were concerned that roundoff might be systematic causing the errors to accumulate in one direction. This was a big concern, and there were debates over whether some rides needed to be redesigned in metric from the start and possibly give up economies of standardized parts.
That this happened in Space Mountain is also interesting, because Space Mountain was the first rollercoaster to have ATIS (automated track inspection system). Since it was a tightly wound coaster in a confined space it was difficult to do visual inspections. ATIS uses two techniques to detect problems with the track. The rails are actually tubes and they're pressurized in sections. When small cracks start to develop, the pressure drop is detected. Sensors also time cars through different sections of the track. If there is a trend of cars slowing through a section over time, it indicates that the ties between the rails are starting to give. ATIS is so much better than visual inspections at detecting problems early that it's used on most modern roller coasters.
Re: Number 1 reason against software patents
on
Perens on Patents
·
· Score: 1
Copyright is the right way to protect software, not patents.
Copyright does not protect inventions, merely expressions of inventions.
My brother is co-inventor on several patents for signal processing algorithms. These are true inventions. They make it possible to give an anethesiolgist very accurate data in real time. It also makes the device safe to use on premature babies. It took years (and therefore lots of money) to develop these algorithms.
Copyright does not protect the algorithms. Only patents do. Of course, these algorithms happen to be implemented as software running on a DSP.
The only way these efforts could have been funded (in our capitalist economy) is by the potential to make money off the results. Being a small company, they make money by licensing their technologies to the manufacturers. FDA approval requires disclosure of the algorithms and imlementation. Without patent protection, the manufacturers could re-implement the algorithms (on a different processor or in a different language to avoid copyright problems).
I'm not saying that software patents are a great thing. There are a lot of problems with the system. I'm just pointing out that the parent comment really shouldn't be modded as "insightful," when it misses the key distinction between copyright and patent protection and therefore concludes that we only need copyright.
IBM has more patents every year than any other company..., but I haven't seen a high-profile case where IBM went after Joe's Software Shack for IP infringement....
In my experience, IBM does attack medium-sized developers with frivolous patent threats. Fortunately, the company I worked for when I encountered this refused to back down. In one case, we spent two years explaining that our code was not infringing on their patent (scaling fonts for print-preview). When they finally accepted that, they hit us with a different one. It was arguably obvious and unoriginal (showing print-preview and the source document at the same time). Rather than fight it, though, we tweaked our product so that you couldn't see the other windows while doing a preview.
I suspect IBM tried this on lots of other companies as well, because I started seeing more and more programs doing the same thing we did, including ones that came from smaller labels. (I guess we should have patented our technique for avoiding IBM's patent.)
And all
that means is the Recording Industry is attempting to achieve compliance through fear.
I'd bet that's only part of it. I think the real goal here is for their lobbyists say to Congress, "We spent $x million trying to stop copyright infringement, and we didn't make a dent, so you need to do something about it."
I find most salespeople treat everyone like idiots. The women polled simply have the perception that it's directed toward them.
When my wife and I shop together, we get treated better, weather it's for electronic gear or furniture. Couples must agree on the big ticket items, so if you're there alone, the salesdrones probably assume you're just browsing.
I don't mean to imply that there's no sexism out there. I'm just saying it may not be as bad as the survey indicates.
Why cater to (define NOT_FLAMEBATE)lazy people who don't read the articles?
The point of the summary is to help people decide if it's a topic they are interested in reading more about. A few extra words could help people more efficiently decide if they want to RTFA and/or the comments. What's the point of slashdotting the article just to find out it's a topic I'm not interested in?
No, [hardware] production costs aren't getting so low as software.
Actually, for a couple hundred bucks, you can design and implement your own processors with a PAL programmer. There's even an open cores movement with VHDL code for processors, VGA controllers, etc. Sure, they're not as fast as there custom counterparts, but the field is moving into the realm of hobbyist, open-sourcer, and individual innovation.
Even in small quantities, printed circuit boards can be farmed out for dollars a board. Sure, that's more expensive than farming out duplication of CDs, but not by much, and not for that much longer.
I think chips and boards will become analogous to binaries. It's just a way to distribute the implementation of hardware without giving away the source.
re-read the article properly again - I did not see any mention in the article about being denied the right to create commercial closed software.
I read the article quite closely. RMS clearly argues that all software should be open. In fact, the whole point of the essay seems to be that it's not enough to have hybrid free/closed systems. For example, he says it's unacceptable to have proprietary drivers on an otherwise "free" system. He says that businesses that make their commercial software run on Linux aren't helping the movement. He's also always quick to point out that open source software is not necessarily the same as free software.
He's not saying I can't write closed software, but there's certainly no room for it in his vision. If he's successful in pushing for the complete incarnation of his vision, then I'll have lost a freedom.
please remember that no programmer is forced to write open source software
Please re-read my post. I never said that we can't write closed software and profit from it. I said that's a freedom we stand to lose if RMS ever gets his way. I'm all for open and free source alternatives, as long as they are part of the spectrum of choice.
(I have been in the telecoms IT industry now for over 20 years and I have *NEVER* known anyone to buy a legitimate copy of MS Office or Windows yet everyone I know seems to have both!)
Allow me to introduce myself. My name's Adrian, and I've (grudingly) purchased several versions of Windows and Office. I've never used illegally copied software at home. And when I've been expected to use it at work, I've raised a stink until the company paid up or switched to a free (as in beer) alternative.
But copyright and license violation (and the belief that MS is evil) are irrelevant to the topic. There's lots of legal software out there, that's used legally, and that's not made by evil companies. To say RMS-style free source is the only option because it's what the market will bear is specious at best. Like I said, I'm totally behind the idea of open and free alternatives, but I'm against the FSF-only model.
Thirdly, how can you call someone that creates a proprietary stock market model a "mathematician"?
www.webster.com: "mathematician: a specialist or expert in mathematics"
My point is these friends are developing new mathematical theories (not just statistical models), and they're not sharing them with the world at large. They have that right. (They've also been making buckets of money with their model for more than a decade now, but only because their model is closed and because they're small enough not to influence the market.)
Fourthly, software is "intangible" - anyone with a computer and a knowledge of programming can create software. Hardware requires access to expensive manufactuting tools and therefore cannot be open to all by it's very nature.
There's a hardware/software convergence going on. Processors run microcode. Chips are designed as software. Manufacturing of both is getting quite cheap. There are lots of individuals designing processors now, and prototyping with inexpesive PALs. Manufacturing of printed circuit boards and even VLSI chips can be farmed out analogously to farming out duplication of software media. It's a cost of distribution, just like bandwidth or burning CDs is a small cost to distribute this otherwise intangible idea of software.
There's an open cores movement--processor and chip implementations in software that follow the same principles as the open source movement. If RMS opposes closed drivers on a free source machine, shouldn't he also oppose running free source on closed hardware?
Today, I can choose to write free software or closed software. It's my choice, and I like it that way. I have nothing against the free software movement, but I disagree with RMS when he suggests that I shouldn't have the freedom to develop software and try to make it a commercial product of it. Why should I only be allowed to market services like installation and support?
Software developers should be like academians? OK. Not all mathematicians share their advances. I know some who develop proprietary models of the stock market for an investment company. It's not for everyone, but shouldn't they have the freedom to choose such a pursuit.
And what makes software so special? Shouldn't hardware be open? Aren't chips mostly designed with source code now? Aren't production costs getting so low that they are essentially commodities like software?
Slashdot Mirror
What bugs me is the stuff they didn't fix. In Empire, when Han is about to be frozen in carbonite, his jacket keeps vanishing and reappearing--a neat trick given that he's handcuffed. I find it the most distracting continuity error in all of film history. Why couldn't they fix that?
I'll stick with my 12 CAV laser disc edition.
I suspect HP isn't doing this out of any sense of right and wrong. I'd wager that they're afraid if they don't do something to curb counterfeiting voluntary, then the government may require them to do something less to their liking. It's the same idea with movie and TV ratings. The industry would rather self-regulate and retain some control than bow down to the will of the lawmakers.
An excellent question. The answer: because the government cares about counterfeiting currency, but only banks care about counterfeiting checks. This is consistent with HP acting out of a fear of government regulation as opposed to good corporate citizenship.
A real Trekkie would know that there have been five live-action Star Trek television series and one animated one. :-)
Having read code from many other programmers and having interviewed scores more, I've found that the best programmers are those who learned simpler, higher-level languages before others.
If Jessica learned Pascal, then C, then assembly, I'd bet she's pretty good at all three. If Samantha learned the same languages in the opposite order, chances are she's not as good at any of them.
You may disagree, but it has certainly proven true in my experience. My impression is that the folks who learn low-level first get bogged down in the details and have a harder time finding the elegant (read "simple") solutions. Once the problem is decomposed, experience with lower level languages translates into tools for building a clean implementation of the solution.
This seems to imply that the server crashed. Maybe I'm misreading it in this case, but I've certainly heard other claims that a slashdotting or googleblatting or DoS attack crashed servers.
When a server is saturated with requests, it should reply to as many of them as it can and allow the rest to time out. If the server crashes as a result of the load, that's a bug.
Perhaps it's just a semantic thing, but it seems misleading to blame the crash on the load. The crash is due to a bug that happens to be aggravated by high load. Blaming the load seems to be a way to deflect responsibility from the engineers who built the server.
Furthermore, a glance at an analog speedometer gives you a feel for rate of acceleration (of deceleration) that's really hard to get from a changing numeric display.
I got a cool Logitech gizmo for Christmas that I still haven't been able to use, because it requires the .NET runtime which simply won't install on my PC. I just get a generic "Set up failed" message. Microsoft Customer Support doesn't reply. Logitech Support--after six requests for help--finally offered a few things to try, none of which worked.
Relying on Microsoft to install a new technology on older machines puts Logitech at a real disadvantage. They should have built a traditional app until .NET is more stable.
True, but other people want access to the data on the mag stripe as well. Often, when I write a check at a store, the clerk asks to swipe my license. I always refuse. I tell them that they're welcome to copy my name and ID number from the license, but I'll not make it easy for them to grab my address and other info for snail-mail spamming, data reporting, etc.
It would be less confrontational if I could just have a "defective" mag stripe.
Furthermore, when California first put the stripes on the licenses, there was a bit of a controversy. Two of the tracks are available to standard readers and contain nothing more than what's printed on the front of your license. But there's a third track, which is recorded at a much higher density and in a format that was incompatible with commonly available readers. The DMV refused to disclose what information they were recording in this third track.
Careful! Studying the virus code to figure out what it does might be a DMCA violation. ;-)
My mistake. The laws I heard about are only for larger companies that have affirmative action plans (compulsory or voluntary) and/or H1-B visa applicants.
Let me say up front that I'm not a lawyer. But the biggest problem I see here is that there is very little if any oversight. Traditional search warrants are (or become) public record, making it possible for people to check for abuse. For example, in California, after a wiretap is completed, law enforcement must contact every party that was heard on the line to let them know they had been recorded. With the gag rules in the PATRIOT ACT, there's no after-the-fact oversight to make sure the judge who granted the request was doing the right thing and that the enforcement agencies aren't routinely asking for wide-reaching powers. You say that, "no library records have ever been searched," but you don't know that because of the gag rules.
I hope Joel is using the comments about deleting and shredding resumes figuratively. There are laws that require employers to keep resumes and applications on file for some period of time after they are received.
A friend of mine in college in the late 80's did an internship with Disney Imagineering. At the time, they were redrawing plans for several attractions that were to be copied almost exactly from California and Florida version for use in Euro Disney. There was great concern that the Imperial to metric round-off would be a problem. The contractor's union in France mandated that they only use metric units in the blueprints, which is reasonable. But if you're trying to duplicate a ride that was originally designed in Imperial units, you need to keep in mind quite a few significant digits when you're measuring. For example, a section of track in a given ride might be 10 feet in the US. In metric that's 3.048 m. Would the French contractors really measure to that precision? Or would they round off to 3.05 or event 3.0? They were concerned that roundoff might be systematic causing the errors to accumulate in one direction. This was a big concern, and there were debates over whether some rides needed to be redesigned in metric from the start and possibly give up economies of standardized parts.
That this happened in Space Mountain is also interesting, because Space Mountain was the first rollercoaster to have ATIS (automated track inspection system). Since it was a tightly wound coaster in a confined space it was difficult to do visual inspections. ATIS uses two techniques to detect problems with the track. The rails are actually tubes and they're pressurized in sections. When small cracks start to develop, the pressure drop is detected. Sensors also time cars through different sections of the track. If there is a trend of cars slowing through a section over time, it indicates that the ties between the rails are starting to give. ATIS is so much better than visual inspections at detecting problems early that it's used on most modern roller coasters.
Copyright does not protect inventions, merely expressions of inventions.
My brother is co-inventor on several patents for signal processing algorithms. These are true inventions. They make it possible to give an anethesiolgist very accurate data in real time. It also makes the device safe to use on premature babies. It took years (and therefore lots of money) to develop these algorithms.
Copyright does not protect the algorithms. Only patents do. Of course, these algorithms happen to be implemented as software running on a DSP.
The only way these efforts could have been funded (in our capitalist economy) is by the potential to make money off the results. Being a small company, they make money by licensing their technologies to the manufacturers. FDA approval requires disclosure of the algorithms and imlementation. Without patent protection, the manufacturers could re-implement the algorithms (on a different processor or in a different language to avoid copyright problems).
I'm not saying that software patents are a great thing. There are a lot of problems with the system. I'm just pointing out that the parent comment really shouldn't be modded as "insightful," when it misses the key distinction between copyright and patent protection and therefore concludes that we only need copyright.
In my experience, IBM does attack medium-sized developers with frivolous patent threats. Fortunately, the company I worked for when I encountered this refused to back down. In one case, we spent two years explaining that our code was not infringing on their patent (scaling fonts for print-preview). When they finally accepted that, they hit us with a different one. It was arguably obvious and unoriginal (showing print-preview and the source document at the same time). Rather than fight it, though, we tweaked our product so that you couldn't see the other windows while doing a preview.
I suspect IBM tried this on lots of other companies as well, because I started seeing more and more programs doing the same thing we did, including ones that came from smaller labels. (I guess we should have patented our technique for avoiding IBM's patent.)
I'd bet that's only part of it. I think the real goal here is for their lobbyists say to Congress, "We spent $x million trying to stop copyright infringement, and we didn't make a dent, so you need to do something about it."
I find most salespeople treat everyone like idiots. The women polled simply have the perception that it's directed toward them.
When my wife and I shop together, we get treated better, weather it's for electronic gear or furniture. Couples must agree on the big ticket items, so if you're there alone, the salesdrones probably assume you're just browsing.
I don't mean to imply that there's no sexism out there. I'm just saying it may not be as bad as the survey indicates.
It's not the server. They're hand drawing images for each visitor.
Time cards in any form do not measure how long you worked. A cheat-proof time card system measures how long you were present.
Just to be anal:
Now can someone explain the horizontal axis?
The point of the summary is to help people decide if it's a topic they are interested in reading more about. A few extra words could help people more efficiently decide if they want to RTFA and/or the comments. What's the point of slashdotting the article just to find out it's a topic I'm not interested in?
Actually, for a couple hundred bucks, you can design and implement your own processors with a PAL programmer. There's even an open cores movement with VHDL code for processors, VGA controllers, etc. Sure, they're not as fast as there custom counterparts, but the field is moving into the realm of hobbyist, open-sourcer, and individual innovation.
Even in small quantities, printed circuit boards can be farmed out for dollars a board. Sure, that's more expensive than farming out duplication of CDs, but not by much, and not for that much longer.
I think chips and boards will become analogous to binaries. It's just a way to distribute the implementation of hardware without giving away the source.
I read the article quite closely. RMS clearly argues that all software should be open. In fact, the whole point of the essay seems to be that it's not enough to have hybrid free/closed systems. For example, he says it's unacceptable to have proprietary drivers on an otherwise "free" system. He says that businesses that make their commercial software run on Linux aren't helping the movement. He's also always quick to point out that open source software is not necessarily the same as free software.
He's not saying I can't write closed software, but there's certainly no room for it in his vision. If he's successful in pushing for the complete incarnation of his vision, then I'll have lost a freedom.
Please re-read my post. I never said that we can't write closed software and profit from it. I said that's a freedom we stand to lose if RMS ever gets his way. I'm all for open and free source alternatives, as long as they are part of the spectrum of choice.
Allow me to introduce myself. My name's Adrian, and I've (grudingly) purchased several versions of Windows and Office. I've never used illegally copied software at home. And when I've been expected to use it at work, I've raised a stink until the company paid up or switched to a free (as in beer) alternative.
But copyright and license violation (and the belief that MS is evil) are irrelevant to the topic. There's lots of legal software out there, that's used legally, and that's not made by evil companies. To say RMS-style free source is the only option because it's what the market will bear is specious at best. Like I said, I'm totally behind the idea of open and free alternatives, but I'm against the FSF-only model.
www.webster.com: "mathematician: a specialist or expert in mathematics"
My point is these friends are developing new mathematical theories (not just statistical models), and they're not sharing them with the world at large. They have that right. (They've also been making buckets of money with their model for more than a decade now, but only because their model is closed and because they're small enough not to influence the market.)
There's a hardware/software convergence going on. Processors run microcode. Chips are designed as software. Manufacturing of both is getting quite cheap. There are lots of individuals designing processors now, and prototyping with inexpesive PALs. Manufacturing of printed circuit boards and even VLSI chips can be farmed out analogously to farming out duplication of software media. It's a cost of distribution, just like bandwidth or burning CDs is a small cost to distribute this otherwise intangible idea of software.
There's an open cores movement--processor and chip implementations in software that follow the same principles as the open source movement. If RMS opposes closed drivers on a free source machine, shouldn't he also oppose running free source on closed hardware?
Today, I can choose to write free software or closed software. It's my choice, and I like it that way. I have nothing against the free software movement, but I disagree with RMS when he suggests that I shouldn't have the freedom to develop software and try to make it a commercial product of it. Why should I only be allowed to market services like installation and support?
Software developers should be like academians? OK. Not all mathematicians share their advances. I know some who develop proprietary models of the stock market for an investment company. It's not for everyone, but shouldn't they have the freedom to choose such a pursuit.
And what makes software so special? Shouldn't hardware be open? Aren't chips mostly designed with source code now? Aren't production costs getting so low that they are essentially commodities like software?