I can see both sides of this one. And I agree that American unionism, as currently practiced, is quite incompatible with the 'geek cowboy' culture that keeps the computers running. But what if a union was more like a guild? Among other things, it could promise employers that when you hire an (apprentice|journeyman|master) you get a certain level of knowledge. This is similar to the idea of 'certification', except I don't think certification should be controlled by a company like Red Hat, but rather by a body of professionals. Ah, that's the concept I'm groping for. I don't want sysadmins etc. to become blue-collar union members, but I'd like us to become professionals, and currently we're not. And given the nature of our work and personalities, we don't need some age-encrusted authority 'certifying' us for knowing Cobol/CICS. We need a new organizational structure, properly adapted to the internet age. Why aren't we professionals now? Well, professionals are people who took an oath in order to assume their current status. Medical doctors take the hippocratic oath, and it's at least controversial when a doctor administers a lethal injection. When a sysadmin is asked by management to violate a user's privacy, he can't say "I took an oath not to do that. I'd be expelled from the worldwide league of sysadmins." But I kind of wish he could.
But it seems like lots of publicly funded companies pursued too much growth too fast. What is it about the market that makes it impossible for a company to focus on providing quality service, and let growth come when appropriate? It seems like corporations are trying to be malignant tumors - irrational rapid growth followed by death. I wonder how much of Covad's problems came from a) Consumers' inability to perceive how little the ILEC's were offering in their 'cheap DSL' packages and b) The ILECs' deliberate non-cooperation in setting up circuits.
Both the article and this discussion are full of hero-worship for this guy. I don't get it. I think he is utterly talentless and unfunny. I remember seeing 'Bloom County' in the college newspaper, and thinking 'why the hell do they publish this?' Of course this will get modded down for swimming against the current - such are the workings of fanboy-ism. I notice that elsewhere in this discussion someone voiced similar dislike of Bloom Country and was told, 'fine - go back to your crappy Garfield' or words to that effect. So just for the record, all 'newspaper comics' are amazingly stupid, unfunny, and untalented. Dilbert is the only exception. But Bloom County is actually worse than the others. In the interview, Breathed talks about how he was influenced by Trudeau. It's funny, I never fully realized that they are two different people until I read that interview - I'd always assumed one was a pseudonym for the other. So maybe my impression was formed during the 'bad period near the end' that all the Bloomatics are referring to. Maybe I was just turned off by the clumsy political satire. Political satire takes great acuity and subtlety - Breathed did not have these gifts. To me, the current-day Berk Breathed is Tom Tomorrow, whose strip appears in 'alternative' newspapers. Tomorrow seems to use clip art as a framework for political mini-rants and sarcasm. If he posted his ideas in an open forum like usenet, they'd be torn to shreds because they are full of factual and logical errors. But from his privileged position as 'artist' he can write without fear of rebuttal.
I kind of hope that in the future, these kinds of things can be assigned a financial values. After all, that's what keeps our civilization working smoothly instead of killing each other over disagreements. So if a company wants my urine, they should make an offer. If it's high enough, I probably don't mind. Same principle applies to employees running Napster at work and sucking up all the bandwidth. The company needs to negotiate a rate at which they sell personal bandwidth to employees. But then I think the whole notion of employment, with it's "job security", paternalism, and tribalism is ridiculous and needs to be phased out.
I did try talking to a CEO about a range of 'geek' matters, including privacy, accessibility, and download times, that affected his business. He understood everything I was saying. His basic response was, "As a web user, I agree with you. But as a CEO I'm more interested in keeping investors happy."
I think you should have a written agreement describing the terms under which you accept the equipment. Either:
Use of the equipment is part of your compensation. Your only responsibility is to return the equipment in good condition when the company requests it.
The company is co-locating the equipment in your dwelling in order to further their business goals. It is to be used only for company business. The company will pay an appropriate colocation fee every month to cover the space and power consumed, cooling cost, and liability. Colocation is not cheap - if you're going to offer it for less than the market, you should have some clear idea of why you can afford this.
Of course, most companies won't go for either of these ideas. They want you to pay the cost of housing the gear, while they get the benefits. So we're back at square one - you're better off saying "NO" to this type of deal.
I agree with you in general, and I've found that the more demanding you are in looking for a job, the happier you'll be. However, I was offered a very similar agreement at one workplace. Fortunately, I had already been hired. Instead of flying into a rage and telling them to go to hell (first impulse) I responded with a list of modifications I wanted made to the agreement. Predictably, the issue fell into a black hole and was never heard from again. This company was a lot of fun to work for - just because some overzealous lawyer came up with this slave agreement didn't say much about the company. Of course, it's different when you deal with a big company that has a HR department and hiring procedures. Currently I contract through a reputable agency, and if the customer offered me anything to sign, I'd explain that my business relationship is with the agency, not them. I'm sure the agency has signed all kinds of NDA's, but they have lawyers to negotiate these things.
Obviously, a commercial site that expected to get every kind of user is going to have to break their back to make sure they support as many browser versions as practicable, while maintaining a sophisticated interface.
Obviously. But what about Ebay? I guess they don't have a sophisticated interface. Does anybody care? Could somebody compete with Ebay by starting 'the auctions site with the sophisticated interface'? Seems to me that the most popular sites on the web work with all browsers because they're simple, not because they have different versions for different browsers. The sites with 'sophisticated interfaces' are just electronic masturbation for web designers. That problem usually solves itself when the company goes bankrupt.
The one 'legitimate' use for javascript is checking form field values before allowing the user to submit a form. Unfortunately, this sucks just as much as the above-mentioned abuses. Why? Because a) Javascript developed for IE doesn't always work the same on Netscape and b) When the limits on the parameter are changed, there are now two places to update them. While in theory, you could auto-generate the javascript to keep up with the current constraints, in practice it's usually broken. On a deeper level, it breaks the truly wonderful things about web programming. Validating a web interface is easy - write a Perl script to try all legal transactions with a range of form field values, both permitted and not. Record results. However, you can't validate javascript checking these easily, which is probably why it's frequently broken on edge cases.
These people seem sincere and well-meaning. The trouble is, I probably don't want the kind of web they're making. Look at it like this: [ Client ] <---> [ WebServer ] <---> [ DataStore ] In most cases, I'm trying to extract data from the data store. I want the web server to be as transparent as possible. However, the web designers want to demonstrate their cleverness by throwing in all sorts of graphics, javascript, etc. In the current regime, I can just barely use lynx on about 80% of sites. People making serious sites don't make javascript mandatory for navigation. This group is asking to change that. ECMAscript, to take only the most offensive part of their platform, is now a 'standard'. So even though I'd like a standard-compliant, less hackish web, I don't really want the web designers having more and more control over the platform I use. I wonder if someone can come up with a 'safe' javascript interpreter for Lynx and LWP. It would make javascript interfaces accessible to Lynx and to scripts, without giving the javascript author any real control over the client platform. I think CSS is a pretty decent idea, though. I can just refuse to download or use the recommended CSS stylesheet. Then I'm left with more structural markup that I can render however I choose. Everyone wins - the web designer gets to design his heart out, and the user never has to look at the 'design'.
No, a degaussing coil probably wouldn't work, even on an inductively coupled RFID (this story is about capacitively coupled RFIDs). The degaussing coil puts out a field at 50/60 Hz. RFID power supplies are designed to resonate at a higher frequency, such as 20 Khz. However, the degausser field might be so strong that it cuts through that filter. If you could send a huge DC pulse through the coil instead of a sine wave, it would be more likely to work, because a step wave contains energy across the whole spectrum. So if you can charge up a bank of capacitors and then rapidly discharged them through a coil, you can probably nuke the RFID that way.
I think the SAT's are quite good at measuring how well you think. Certainly more accurate than grades, which basically measure how well you kiss ass. The best preparation for the SAT's is to read a lot of books. Of course, I knew people who put the cart before the horse and arduously prepared for the SAT's.
He talked about visiting classrooms where 12-year-old's spent hours studying lists of analogies, a central feature of the SAT.
I remember people like that. Somehow I doubt that all that memorization helped them. If you don't think rationally to start with, analogies will seem an arbitrary puzzle disconnected from daily life. There's a paradox here, somehow related to the fact that people who got into programming for the money aren't making the best money. Anyway, I'm sad to see the SAT losing ground, as it's one of the few standard objective measures in education. But clearly the rulers of the education establishment want 'diversity' rather than academic quality.
I agree. I feel sorry for all the people trying to port their Windows-based business model to Linux. They approach Linux in good faith, but they're too removed from the culture and values of this very different world, so they're rejected. However, I don't like any of the Linux magazines. The only good thing they have is ads, which give you an idea of what hardware vendors are catering to the Linux crowd. (Ads cost money. Press releases don't.) Recently I picked up one of these magazines in the bookstore (can't remember which magazine) and saw a 'Tape Drive Roundup'. I'm interested in buying a new tape, so I read it. I was disappointed. The article showed utter confusion about the world of tape drives. The authors offered no explanation of how they selected the particular models they reviewed. They also failed to differentiate between comparing technologies and comparing vendors. So they compared a DLT from Vendor A to a DAT from Vendor B. This would be like a 'car roundup' that compares a Toyota Tercel, a Honda Accord, and a Ford F350 pickup truck. The other irritating thing about the article was that they tested the drives with some closed-source backup software. There are good reasons to use Veritas if you want to drive a tape robot, but I think most sysadmins choosing a mid-range tape drive are going to use tar/cpio/dump + shell scripts. It seemed to me that the authors were Windows-centric and automatically assumed the commercial GUI program was more interesting. Magazines I actually buy: Sysadmin and The Perl Journal. However, I'm sad to see these things die, because while they weren't helping the community directly, they were helping people find the community. Hackers tend to underestimate the importance of visual artifacts. I think that many real-world linux deployments could not have happened if the PHB's hadn't been assured of Linux's validity by glossy publications. Same thing for Tucows BSD: it sure wasn't going to benefit the hardcore BSD hacker, but it might make a curious Windows user aware of BSD.
I doubt it hurts much really. Allchin's FUD is far below Microsoft's usual standard. Now if I were M$, I'd put together a bunchof 'case histories' of companies deploying linux, becoming frustrated by its limitations, and switching to M$. The media loves telling the story of something that was overhyped and is now debunked. If M$ were really smart, they'd work to sell that story to the media. The brutal frontal assault that Allchin is using can only make M$ look frightened.
Good point, and one that needs to be repeated when this silly meme of 'programmers will starve' comes up. I know a reasonable number of programmers (including myself) and not one of them is working on shrinkwrap software. In some cases, our employers (reluctantly) open source the code. In others, the code is buried in the bowels of the corporation. So this guy making generalizations about 'top programmers' is just silly. First of all, it may be that all the top programmers are maintaining legacy apps at big financial companies. In which case, whatever happens to Microsoft or Linux has almost no effect on them. But its more likely that 'top programmers' are distributed evenly across the different kinds of programmers, which would still mean that the majority of them never come anywhere near the commercial software market. To turn this nonsense around, imagine if medical doctors in the 19th century had started patenting their medical procedures and trying to extract royalties. The small minority of doctors who successfully did this would be highly visible. Then if some doctor hit on the bright idea of *not* patenting his discoveries, it would seem like a radical movement that threatens the survival of doctors. Of course, the majority of doctors weren't in that racket anyway!
Government support for opensource could be a huge boon to what we are doing...
More like a huge disaster. Open source does best when it's fueled by individual enthusiasm, not corporate/government mandates. I don't want Open Source software replacing Microsoft in a government agency because of some fiat from above. I want it happening because the the responsible IT managers decide that it delivers the best benefits at the least cost. Likewise, if the government funded tons of Open Source developers, the results would suck. If the developers are dumb, they'll write bad code. If they're smart, they'll build overly complex application frameworks and stuff that nobody wants. We don't need government support, just freedom from government interference. I'd agree that government employees and employees of government contractors should not be prohibited or hindered from writing Open Source software on their own time.
...they'd be working for locally owned businesses...
But that's been even more problematic. When a US VBC contracts production to a small company that contracts production to even smaller ones, the actual factory conditions can be awful, and the VBC generally has no knowledge of those conditions. This was happening with some clothing maker - can't remember the name.
Yes, I like Chesterton and think that much of what he wrote applies to today's world. After all, he wrote the essay 'Grocers as Gods' which is roughly about the rise of Walmart and others. I think he wrote in that essay that the only reason a merchant lines both sides of the street with his enormous shops is to prevent you from buying what you want. And that applies very well to Walmart, who have mastered the art of becoming a profitable bottleneck between producer and consumer. For some reason, Walmart has an evil vibe, and yet Target, which is quite similar, has a good vibe.
This guy isn't some social crusader seeking to remedy evils. He's an MIT student hacking a system. Nike claimed to build a system with this characteristic:
If you feed money and a text string in one end, AND the text string does NOT meet four criteria, a pair of shoes will come out the other end with the text string printed on them.
The student tested this system and found that there's another, unstated criterion in the real system. He tested the bounds of the alleged freedom of expression Nike is selling, and found an interesting data point.
That's a good point. Now, when a work enters the public domain you are free to make derivative works from it. So what happens if you make a compendium of several recently-freed works, complete with commentary, and then Congress extends copyright retroactively and the freed works are no longer public domain? Do you lose the copyright on your derivative work?
But isn't this the 'halting problem'? I mean, you can prove that a piece of code does something nasty, but you can't prove it doesn't do something nasty. Of course, given a small enough program, you can be fairly confident that it has no hidden capabilities. Still can't prove it though. But yeah, this will be a godsend for the antivirus companies. "In other news, a Cleveland woman was killed by the self-morphing encrypted radioactive fab virus when she clicked on a 'warm fuzzy bunny' in her inbox. If you see a warm fuzzy bunny in your inbox, do _not_ click on it. Her remains are so highly radioactive that they were immediately transported to a toxic waste dump by lead-suited workers from the nearby power plant. Experts say the neighborhood may be habitable again in 25 years."
Police gravitate towards low-risk, low-investment arrests. Given a choice between chasing violent, elusive, possibly armed criminals through a city and arresting harmless kids in their homes, they'll naturally choose the latter. And cracking down on the mafia is hard.
Not paying for something that is not yours, and for which you have no right,...
So who gave you permission to use the word "PRINCIPLE"? Who invented that word? Who owns it? What makes you think you have a right to use it? Intellectual property is an absurd idea. It's like a child planting a flag in the ground and claiming to own the earth. All these people protecting their precious intellectual property are ignoring the vastly greater FREE intellectual property without which they wouldn't even be alive. Intellectual property is owned by a tiny aristocracy that has managed to delude many people into empathizing with them. You sound just like some obedient flunky on the eve of the French revolution condemning the disrespectful attitude of the sans-culottes.
The people in this town seem to feel they have the "right" to fish in the stream and take home all the fish they want, without paying a centime. Is this right? Only if the lord of the manor gives his blessing, for surely everything in the town belongs to him.
I have eaten some fish in the past, but my conscience is troubling me. Sure, Monsigneur is fabulously wealthy, but really, that is his job, he has EARNED it, one way or another. Of course one of you ungrateful rebels will shout: "But Monsigneur loses nothing when I take a fish from the stream!" And you know what? That does not matter one little bit. It is the PRINCIPLE of the act. Not paying for something that is not yours, and for which you have no right, goes against all the values upon which our society is based.
''Open source is an intellectual-property destroyer,'' Allchin said.
I don't get it. Say Microsoft makes a piece of intellectual property, for example IIS. And the open source folks make a competing program, Apache. It might be true to say that Apache is destroying IIS's market share. But Apache is not destroying the intellectual property contained in IIS. When the last IIS webserver is retired, Microsoft will still have just as much ownership of IIS as ever.
''I worry if the government encourages open source, and I don't think we've done enough education of policy makers to understand the threat.''
Open Source may be a threat to Microsoft, but I don't see how it's a threat to the government. So why should policy makers care about the threat. Was the automobile a threat to the carriage industry?
Despite Linux's success in some markets, Allchin says he isn't concerned about sales competition from the product.
Now I'm even more confused. If Allchin is not concerned about sales competition from 'the product', then what exactly is this threat he's going on about? Maybe he's saying he doesn't fear Linux, he fears (Apache|FreeBSD|something).
''We can build a better product than Linux,'' he said.
Oh, good. I look forward to that. Jim, you might want to check with Bill. A few years ago he said that Windows NT would be a better Unix than Unix. Could you ask him how that project is coming? Since NT is being replaced by W2k (yes I know it's really the same) is it finally time to admit that NT has not lived up to that boast?
I'll go out on a limb here: The original poster was not saying that the base and modulus should be generated in real time. That would be pointless. Rather, I think he was saying they should be generated per-machine or per-site. That way the benefit of 'studying' these values is minimized. Also, magic numbers are always undesirable in a cryptographic protocol. Even if you can't see any way they could be cooked. That's why Schneier used digits from PI for constants.
Slashdot Mirror
I can see both sides of this one. And I agree that American unionism, as currently practiced, is quite incompatible with the 'geek cowboy' culture that keeps the computers running. But what if a union was more like a guild? Among other things, it could promise employers that when you hire an (apprentice|journeyman|master) you get a certain level of knowledge. This is similar to the idea of 'certification', except I don't think certification should be controlled by a company like Red Hat, but rather by a body of professionals.
Ah, that's the concept I'm groping for. I don't want sysadmins etc. to become blue-collar union members, but I'd like us to become professionals, and currently we're not. And given the nature of our work and personalities, we don't need some age-encrusted authority 'certifying' us for knowing Cobol/CICS. We need a new organizational structure, properly adapted to the internet age.
Why aren't we professionals now? Well, professionals are people who took an oath in order to assume their current status. Medical doctors take the hippocratic oath, and it's at least controversial when a doctor administers a lethal injection. When a sysadmin is asked by management to violate a user's privacy, he can't say "I took an oath not to do that. I'd be expelled from the worldwide league of sysadmins." But I kind of wish he could.
But it seems like lots of publicly funded companies pursued too much growth too fast. What is it about the market that makes it impossible for a company to focus on providing quality service, and let growth come when appropriate? It seems like corporations are trying to be malignant tumors - irrational rapid growth followed by death.
I wonder how much of Covad's problems came from a) Consumers' inability to perceive how little the ILEC's were offering in their 'cheap DSL' packages and b) The ILECs' deliberate non-cooperation in setting up circuits.
Both the article and this discussion are full of hero-worship for this guy. I don't get it. I think he is utterly talentless and unfunny. I remember seeing 'Bloom County' in the college newspaper, and thinking 'why the hell do they publish this?' Of course this will get modded down for swimming against the current - such are the workings of fanboy-ism.
I notice that elsewhere in this discussion someone voiced similar dislike of Bloom Country and was told, 'fine - go back to your crappy Garfield' or words to that effect. So just for the record, all 'newspaper comics' are amazingly stupid, unfunny, and untalented. Dilbert is the only exception. But Bloom County is actually worse than the others.
In the interview, Breathed talks about how he was influenced by Trudeau. It's funny, I never fully realized that they are two different people until I read that interview - I'd always assumed one was a pseudonym for the other.
So maybe my impression was formed during the 'bad period near the end' that all the Bloomatics are referring to. Maybe I was just turned off by the clumsy political satire. Political satire takes great acuity and subtlety - Breathed did not have these gifts.
To me, the current-day Berk Breathed is Tom Tomorrow, whose strip appears in 'alternative' newspapers. Tomorrow seems to use clip art as a framework for political mini-rants and sarcasm. If he posted his ideas in an open forum like usenet, they'd be torn to shreds because they are full of factual and logical errors. But from his privileged position as 'artist' he can write without fear of rebuttal.
I kind of hope that in the future, these kinds of things can be assigned a financial values. After all, that's what keeps our civilization working smoothly instead of killing each other over disagreements. So if a company wants my urine, they should make an offer. If it's high enough, I probably don't mind.
Same principle applies to employees running Napster at work and sucking up all the bandwidth. The company needs to negotiate a rate at which they sell personal bandwidth to employees.
But then I think the whole notion of employment, with it's "job security", paternalism, and tribalism is ridiculous and needs to be phased out.
I did try talking to a CEO about a range of 'geek' matters, including privacy, accessibility, and download times, that affected his business. He understood everything I was saying. His basic response was, "As a web user, I agree with you. But as a CEO I'm more interested in keeping investors happy."
Of course, most companies won't go for either of these ideas. They want you to pay the cost of housing the gear, while they get the benefits. So we're back at square one - you're better off saying "NO" to this type of deal.
I agree with you in general, and I've found that the more demanding you are in looking for a job, the happier you'll be. However, I was offered a very similar agreement at one workplace. Fortunately, I had already been hired. Instead of flying into a rage and telling them to go to hell (first impulse) I responded with a list of modifications I wanted made to the agreement.
Predictably, the issue fell into a black hole and was never heard from again. This company was a lot of fun to work for - just because some overzealous lawyer came up with this slave agreement didn't say much about the company.
Of course, it's different when you deal with a big company that has a HR department and hiring procedures. Currently I contract through a reputable agency, and if the customer offered me anything to sign, I'd explain that my business relationship is with the agency, not them. I'm sure the agency has signed all kinds of NDA's, but they have lawyers to negotiate these things.
Obviously. But what about Ebay? I guess they don't have a sophisticated interface. Does anybody care? Could somebody compete with Ebay by starting 'the auctions site with the sophisticated interface'?
Seems to me that the most popular sites on the web work with all browsers because they're simple, not because they have different versions for different browsers. The sites with 'sophisticated interfaces' are just electronic masturbation for web designers. That problem usually solves itself when the company goes bankrupt.
The one 'legitimate' use for javascript is checking form field values before allowing the user to submit a form. Unfortunately, this sucks just as much as the above-mentioned abuses. Why? Because a) Javascript developed for IE doesn't always work the same on Netscape and b) When the limits on the parameter are changed, there are now two places to update them. While in theory, you could auto-generate the javascript to keep up with the current constraints, in practice it's usually broken.
On a deeper level, it breaks the truly wonderful things about web programming. Validating a web interface is easy - write a Perl script to try all legal transactions with a range of form field values, both permitted and not. Record results. However, you can't validate javascript checking these easily, which is probably why it's frequently broken on edge cases.
These people seem sincere and well-meaning. The trouble is, I probably don't want the kind of web they're making. Look at it like this:
[ Client ] <---> [ WebServer ] <---> [ DataStore ]
In most cases, I'm trying to extract data from the data store. I want the web server to be as transparent as possible. However, the web designers want to demonstrate their cleverness by throwing in all sorts of graphics, javascript, etc. In the current regime, I can just barely use lynx on about 80% of sites. People making serious sites don't make javascript mandatory for navigation.
This group is asking to change that. ECMAscript, to take only the most offensive part of their platform, is now a 'standard'. So even though I'd like a standard-compliant, less hackish web, I don't really want the web designers having more and more control over the platform I use.
I wonder if someone can come up with a 'safe' javascript interpreter for Lynx and LWP. It would make javascript interfaces accessible to Lynx and to scripts, without giving the javascript author any real control over the client platform.
I think CSS is a pretty decent idea, though. I can just refuse to download or use the recommended CSS stylesheet. Then I'm left with more structural markup that I can render however I choose. Everyone wins - the web designer gets to design his heart out, and the user never has to look at the 'design'.
No, a degaussing coil probably wouldn't work, even on an inductively coupled RFID (this story is about capacitively coupled RFIDs).
The degaussing coil puts out a field at 50/60 Hz. RFID power supplies are designed to resonate at a higher frequency, such as 20 Khz. However, the degausser field might be so strong that it cuts through that filter. If you could send a huge DC pulse through the coil instead of a sine wave, it would be more likely to work, because a step wave contains energy across the whole spectrum. So if you can charge up a bank of capacitors and then rapidly discharged them through a coil, you can probably nuke the RFID that way.
Of course, I knew people who put the cart before the horse and arduously prepared for the SAT's.
I remember people like that. Somehow I doubt that all that memorization helped them. If you don't think rationally to start with, analogies will seem an arbitrary puzzle disconnected from daily life.
There's a paradox here, somehow related to the fact that people who got into programming for the money aren't making the best money.
Anyway, I'm sad to see the SAT losing ground, as it's one of the few standard objective measures in education. But clearly the rulers of the education establishment want 'diversity' rather than academic quality.
I agree. I feel sorry for all the people trying to port their Windows-based business model to Linux. They approach Linux in good faith, but they're too removed from the culture and values of this very different world, so they're rejected.
However, I don't like any of the Linux magazines. The only good thing they have is ads, which give you an idea of what hardware vendors are catering to the Linux crowd. (Ads cost money. Press releases don't.)
Recently I picked up one of these magazines in the bookstore (can't remember which magazine) and saw a 'Tape Drive Roundup'. I'm interested in buying a new tape, so I read it.
I was disappointed. The article showed utter confusion about the world of tape drives. The authors offered no explanation of how they selected the particular models they reviewed. They also failed to differentiate between comparing technologies and comparing vendors. So they compared a DLT from Vendor A to a DAT from Vendor B. This would be like a 'car roundup' that compares a Toyota Tercel, a Honda Accord, and a Ford F350 pickup truck.
The other irritating thing about the article was that they tested the drives with some closed-source backup software. There are good reasons to use Veritas if you want to drive a tape robot, but I think most sysadmins choosing a mid-range tape drive are going to use tar/cpio/dump + shell scripts. It seemed to me that the authors were Windows-centric and automatically assumed the commercial GUI program was more interesting.
Magazines I actually buy: Sysadmin and The Perl Journal.
However, I'm sad to see these things die, because while they weren't helping the community directly, they were helping people find the community. Hackers tend to underestimate the importance of visual artifacts. I think that many real-world linux deployments could not have happened if the PHB's hadn't been assured of Linux's validity by glossy publications. Same thing for Tucows BSD: it sure wasn't going to benefit the hardcore BSD hacker, but it might make a curious Windows user aware of BSD.
I doubt it hurts much really. Allchin's FUD is far below Microsoft's usual standard. Now if I were M$, I'd put together a bunchof 'case histories' of companies deploying linux, becoming frustrated by its limitations, and switching to M$.
The media loves telling the story of something that was overhyped and is now debunked. If M$ were really smart, they'd work to sell that story to the media.
The brutal frontal assault that Allchin is using can only make M$ look frightened.
Good point, and one that needs to be repeated when this silly meme of 'programmers will starve' comes up. I know a reasonable number of programmers (including myself) and not one of them is working on shrinkwrap software. In some cases, our employers (reluctantly) open source the code. In others, the code is buried in the bowels of the corporation.
So this guy making generalizations about 'top programmers' is just silly. First of all, it may be that all the top programmers are maintaining legacy apps at big financial companies. In which case, whatever happens to Microsoft or Linux has almost no effect on them. But its more likely that 'top programmers' are distributed evenly across the different kinds of programmers, which would still mean that the majority of them never come anywhere near the commercial software market.
To turn this nonsense around, imagine if medical doctors in the 19th century had started patenting their medical procedures and trying to extract royalties. The small minority of doctors who successfully did this would be highly visible. Then if some doctor hit on the bright idea of *not* patenting his discoveries, it would seem like a radical movement that threatens the survival of doctors. Of course, the majority of doctors weren't in that racket anyway!
More like a huge disaster. Open source does best when it's fueled by individual enthusiasm, not corporate/government mandates. I don't want Open Source software replacing Microsoft in a government agency because of some fiat from above. I want it happening because the the responsible IT managers decide that it delivers the best benefits at the least cost.
Likewise, if the government funded tons of Open Source developers, the results would suck. If the developers are dumb, they'll write bad code. If they're smart, they'll build overly complex application frameworks and stuff that nobody wants.
We don't need government support, just freedom from government interference. I'd agree that government employees and employees of government contractors should not be prohibited or hindered from writing Open Source software on their own time.
But that's been even more problematic. When a US VBC contracts production to a small company that contracts production to even smaller ones, the actual factory conditions can be awful, and the VBC generally has no knowledge of those conditions. This was happening with some clothing maker - can't remember the name.
Yes, I like Chesterton and think that much of what he wrote applies to today's world. After all, he wrote the essay 'Grocers as Gods' which is roughly about the rise of Walmart and others.
I think he wrote in that essay that the only reason a merchant lines both sides of the street with his enormous shops is to prevent you from buying what you want. And that applies very well to Walmart, who have mastered the art of becoming a profitable bottleneck between producer and consumer.
For some reason, Walmart has an evil vibe, and yet Target, which is quite similar, has a good vibe.
The student tested this system and found that there's another, unstated criterion in the real system. He tested the bounds of the alleged freedom of expression Nike is selling, and found an interesting data point.
That's a good point. Now, when a work enters the public domain you are free to make derivative works from it. So what happens if you make a compendium of several recently-freed works, complete with commentary, and then Congress extends copyright retroactively and the freed works are no longer public domain? Do you lose the copyright on your derivative work?
But isn't this the 'halting problem'? I mean, you can prove that a piece of code does something nasty, but you can't prove it doesn't do something nasty. Of course, given a small enough program, you can be fairly confident that it has no hidden capabilities. Still can't prove it though.
But yeah, this will be a godsend for the antivirus companies. "In other news, a Cleveland woman was killed by the self-morphing encrypted radioactive fab virus when she clicked on a 'warm fuzzy bunny' in her inbox. If you see a warm fuzzy bunny in your inbox, do _not_ click on it. Her remains are so highly radioactive that they were immediately transported to a toxic waste dump by lead-suited workers from the nearby power plant. Experts say the neighborhood may be habitable again in 25 years."
Police gravitate towards low-risk, low-investment arrests. Given a choice between chasing violent, elusive, possibly armed criminals through a city and arresting harmless kids in their homes, they'll naturally choose the latter.
And cracking down on the mafia is hard.
So who gave you permission to use the word "PRINCIPLE"? Who invented that word? Who owns it? What makes you think you have a right to use it?
Intellectual property is an absurd idea. It's like a child planting a flag in the ground and claiming to own the earth. All these people protecting their precious intellectual property are ignoring the vastly greater FREE intellectual property without which they wouldn't even be alive.
Intellectual property is owned by a tiny aristocracy that has managed to delude many people into empathizing with them. You sound just like some obedient flunky on the eve of the French revolution condemning the disrespectful attitude of the sans-culottes.
I don't get it. Say Microsoft makes a piece of intellectual property, for example IIS. And the open source folks make a competing program, Apache. It might be true to say that Apache is destroying IIS's market share. But Apache is not destroying the intellectual property contained in IIS. When the last IIS webserver is retired, Microsoft will still have just as much ownership of IIS as ever.
Open Source may be a threat to Microsoft, but I don't see how it's a threat to the government. So why should policy makers care about the threat. Was the automobile a threat to the carriage industry?
Now I'm even more confused. If Allchin is not concerned about sales competition from 'the product', then what exactly is this threat he's going on about? Maybe he's saying he doesn't fear Linux, he fears (Apache|FreeBSD|something).
Oh, good. I look forward to that. Jim, you might want to check with Bill. A few years ago he said that Windows NT would be a better Unix than Unix. Could you ask him how that project is coming? Since NT is being replaced by W2k (yes I know it's really the same) is it finally time to admit that NT has not lived up to that boast?
I'll go out on a limb here:
The original poster was not saying that the base and modulus should be generated in real time. That would be pointless. Rather, I think he was saying they should be generated per-machine or per-site. That way the benefit of 'studying' these values is minimized. Also, magic numbers are always undesirable in a cryptographic protocol. Even if you can't see any way they could be cooked.
That's why Schneier used digits from PI for constants.