Its not like there were any groundbreaking inventions coming out of mexico... besides maybe the double-sided bong?
There's more to creativity than technical achievement.
Mexico, among a few other Latin American countries -- Chile and Argentina spring most immediately to mind -- was the source of some of the best and most innovative literature of the 20th century, arguably far and away better than most of what came out of the English-speaking world. Mexican artists were significant contributors to the surrealist movement as well. The idea that the common cultural heritage of the Mexican people could be stolen by the Mexican government and pimped out for the profit of politicians and their favored private sector contractors is an utter disgrace for the government and a tragedy for both their citizens and ours.
I hope the Mexican people will fight this hijacking of their rich heritage.
What the hell do troop movements in the Middle East have to do with "News for Nerds"?
If I want shoddily reported, unsubstantiated rumors about the war, I can go to CNN. I count on Slashdot to give me shoddily reported, unsubstantiated rumors about technology.
Nicholas, given that the vast, vast majority of developers write software for internal corporate use and B2B applications, what possible significance can your 40% number have for the consumer desktop? Windows dominated the consumer desktop market before it penetrated a corporate market dominated by Unix, VMS, OS/360, etc., not the other way around. No one seriously disputes the rapid growth of Linux in the enterprise, but it seems to me that the corporate server market -- even the corporate desktop market -- has very little influence on what my grandmother or my daughter use on their desktops.
If 40% of developers are developing for Linux, where are the commercial apps?
I assume you mean boxed, shrink-wrapped apps. Almost no one develops consumer applications. If every developer who worked on consumer applications disappeared tomorrow, the total headcount of developers would not change significantly.
On the one hand, this is the reason that Microsoft's old scare stories about Open Source costing jobs were nonsense -- the vast majority of developers write software which is used in-house or sold at the enterprise level.
On the other hand, this is also the reason why Mr. Petreley's observations about developer percentages don't mattter one damn bit. 40% of developers? Who cares? What percentage of retail consumer software developers are developing for Linux? If the number is even as high 1%, I'd be deeply shocked.
Linux is rapidly taking over in the enterprise space. This no more means that Linux is poised to be your grandmother's OS than it means that Linux is poised to replace PS/2 and XBox.
Not that I don't wish that were the case, but world domination has been Real Soon Now since 1996.
There was also Coherent, and there's still QNX. Even Microsoft had a Unix offering at one point, a 16-bit OS called Xenix, which SCO should know, since Microsoft sold it to them.
Sports are important. [...] Organized sports are an unnatural event.
I have to agree, insofar as physical health is a prerequisite for full mental health, to say nothing of avoiding premature death. One firm we contract with consists entirely of severely obese network engineers. It's not their appearance I find shocking -- indeed, it's irrelevant since the quality of their work is good -- it's the fact that having a high probability of dying in their forties or fifties doesn't seem to faze them at all.
When I was in high school, I eschewed team sports and still do. All the crap coaches spew about the valuable skills one learns from team sports -- teamwork, dedication, etc. -- are better learned elsewhere in areas that don't encourage pointless aggression, meathead elitism, and date rape. I learned teamwork on the job out in the real world. I learned dedication in the face of adversity from the wide variety of spontaneous sources of adversity that the world provides.
There are plenty of ways to get exercise and enjoy the not inconsiderable drug-like rush that comes from sustained physical exertion without being a part of jock culture. Solitary sports like cycling and weight lifting were always my favorites. I have friends who speak highly of running, canoeing, climbing, and martial arts. It's a real shame that the abusive culture that goes with the popular team sports turns off so many people, geeks included, to other, more rewarding forms of athletics.
To return to the original topic, though, it's frankly bizarre that geeks have acquired this strange reputation for random violence when participants in team sports, organized religion, and the armed forces are much, much likelier sources of random murder. But then, those are the people who are in charge.
This isn't an issue limited to free software or even to software at all; it's a technology issue and, reduced to its bare essence, the question is whether an inventor is responsible for the use of his inventions.
The question of intellectual property is entirely beside the point. I remember that, when downloading Oracle for Linux, I was required to fill out an HTML form affirming that I would not use the product for the development of nuclear weapons. Somehow, I have a hard time picturing Saddam Hussein or Kim Jong Il saying, "Well, shit. I can't use Oracle for my clandestine nuclear program because of this license. I guess I'll have to use MySQL instead." The same is true of commercial software. Does anyone believe an export license (or for that matter, a licensing fee) would stop anyone from either purchasing a boxed copy in the US or Europe or just downloading a copy from alt.binaries.warez.ibm-pc? If you don't want your code to be misused, don't release it. If you release it, it eventually, inevitably will be misused.
Several years ago, I completed work on a library and set of tools for textual- and communication-traffic analysis. Among the things you could do with the tools was determining authority relationships between people in an organization on the basis of the patterns of their email communications. Another interesting application, which I tested with a full non-binaries Usenet feed, was a surprisingly effective system for determining the political affiliation of posters on the basis of their non-political postings. (For the curious, I used a sample group of 1,000 posters who made consistent ideological posts to political newsgroups as well as non-political posts to non-political newsgroups.) The accuracy rate over a six month period approached 95%.
Concluding that such code could be used by governments to track political dissidents, I was reluctant to release the code. Once John Ashcroft and John Poindexter appeared on the scene, I destroyed the source. This was probably pointless, as the algorithms being used are well-understood -- only the particular combination of algorithms was novel -- and the NSA probably has similar software written by much smarter guys than me, but when my own government, much less foreign tyrants, is arresting people without charges and holding them incommunicado in undisclosed locations, I didn't want to be responsible for contributing to the next round of political arrests.
Contrary to what the above seems to imply, I don't think inventors ought to be held responsible for the misuse of their inventions. I do think that inventors ought to be held responsible for failing to consider the potential consequences of misuse, however. In my case, I decided the potential benefits were outweighed by the potential abuses and decided not to release. Ultimately, that's all you can do if you are concerned about abuses, for the simple reason that the people who are likely to abuse your code are not going to be stopped by legal fictions like the GPL, copyrights, patents, or anything less substantial than the barrel of a gun.
Mandrake Linux has been beaten down by linux experts...
These are obviously "experts" who don't have to support non-technical users, which makes me wonder if their expertise is limited to their bedrooms in their parents' home. My experience has been that Mandrake is a pretty sweet deal if you're the only sysadmin in the department and have better things to do with your time than configure crap for the receptionist.
Under the current state of US law, unauthorized access to a computer system is a federal crime. (I can't speak to EU laws, but I suspect parallels exist.) If Company X says, "You must use Internet Explorer 5.5 to access this site," then you must use IE 5.5. Of course, it would be just plain stupid to do so, but it's their computer system, and they get to decide who is authorized.
To judge from most of the comments here, the fact that it is incredibly stupid to impose such restrictions has obscured what is actually a legally unambiguous situation. Just because it's dumb doesn't mean it's not legal.
That an http server is nominally "public" doesn't mean diddly here. Any number of http servers provide for member- or employee-only access. The brick and mortar parallel would be those signs that say things like, "No shirt, no shoes, no service."
It is surprising that so few people have touched on the reason why companies might object to the distribution of Perl modules designed to harvest data from their sites: bandwidth costs and site performance. It doesn't take too many cron jobs banging on a site every minute -- and being ignored by their users most of the time -- to degrade site performance for "live" users and run up steep bandwidth bills.
Now, there is certainly no legal basis for Company X to demand that CPAN remove the modules, though it is hardly out of line to ask nicely. But there is firm legal grounds to prohibit anyone from actually using those modules.
Legal action is probably the wrong way to handle this, though. Having written fairly complicated web scrapers before, I know how easy it would be to make a site virtually impossible to harvest. Rather than make a big stink about the Perl programmers who contribute to CPAN, Company X would be well-advised to hire a good Perl programmer to thwart automated harvesters.
Rather than releasing new versions, perhaps the distro vendors should eliminate the concept altogether in favor of the sort of seamless, continuous per-package upgrading I suspect most people would like to see. Does anyone really care what version number a distribution has? I suspect more people care what kernel and security patches and application versions they're running.
For the user, this would have the advantage of being able to click a button or insert the latest update CD and upgrade all of the necessary packages. (We presume, of course, that you could elect to forego certain upgrades -- one might wish to continue running Apache 1.3.x instead of a 2.x version.)
For the vendor, this would be an obvious opportunity to sell subscriptions as well as avoid the endless cost of producing shrinkwrapped distributions.
Of course -- of course -- this would require greater effort on the part of vendors to make sure that the upgrade process is robust and seamless so as to avoid the problems M$ customers have with their so-called Service Packs, but it ought to be doable.
Re:I'm more amazed....
on
Baked Apple
·
· Score: 1
I'm more amazed that no one asked her why she did it... Do we look down on non-computer people so much that we don't even bother to ask anymore why they do stupid things?
If it's physical presence that's the case, this should result in an enormous flood of online-only retailers to Oregon, where there is no sales tax. We could certainly use the jobs here.
OTOH, there are already big disincentives to shopping online: shipping -- which is often more than sales tax would be -- and delivery time. Add sales tax to that, and the only cases in which buying online makes sense would be
When the base price is low enough to compensate for shipping and tax...
When the customer can afford to wait for delivery...
When convenience is the main consideration.
I suspect convenience is less of a selling point in the current state of the economy than it would be at other times.
For books and CDs, which are about the only things I buy online regularly, it's probably easier and cheaper for me to call my local Barnes and Noble or Borders and ask them to special-order something and drive five minutes to pick it up when it arrives. Unless I can get a used copy more cheaply online.
The floppy persists because it is the only convenient way of transporting small files without a network using inexpensive reusable media that everyone can read. All of the floppy's competitors fail on one or more of those three requirements. The only medium that comes close these days would be the rewritable CD, but burning a CD is a lot less convenient than copying a file to a floppy, and -- at least in my experience -- CD-RWs are a lot less reliable than 3.5" floppies if they get passed around much. Floppies, after all, don't scratch easily.
It seems to me that not having a floppy drive would make it harder to flash BIOS firmware. I suppose it could be done with an El Torito image, though.
One of my earliest memories is seeing the last of the Apollo launches from the beach in Florida. I watched the first launch of the Columbia with my class in school. I got to see it in person, once, when it was being kept briefly at Ft. Campbell, KY, to avoid some hurricane or other.
My ten-year-old doesn't understand why this is a big deal. Space travel, to her, is like CDs and PCs and microwave ovens -- a routine part of the world as it is. She was born after the cold war, after the glory days of the space program. Maybe when she's older, she'll understand that the space program transcended all the petty factional divisions and murderous religious and political ideologies of this sad world and was for a lot of us a shining example of the very best of the human race and a beacon of hope for a better future.
Growing up in the 70's, astronauts were the only people I ever really thought of as heroes. NASA was the only government agency I could admire, whatever its faults, without a trace of cynicism. That hasn't changed.
I wish I could somehow take my daughter back in time to that day on the beach when I looked southward towards the Cape and saw a Saturn V rise from the horizon on a pillar of flame. Maybe then she could understand why her parents were crying in front of the TV today. Instead, the best I could manage to say was, "They were astronauts. Our dreams went with them."
Godspeed, folks. You were the best of the best. You will not be forgotten.
Trademarks are going to get pretty difficult if simply rhyming with an existing trademark is going to count as an infringement. I don't know about French, but a quick Perl script chewing on CMU's pronouncing dictionary reveals only a couple dozen single-syllable endings to English words, and only a few hundred two-syllable endings.
Sure, the US is open compared to most of the countries in the world, but it's not as open as open source.
I hear this a lot from Americans. (Don't get your nationalist knickers in a wad; by accident of birth, I'm one of you, too.) The problem is that it is only a half-truth. If by "most of the countries in the world" you mean to include Brunei, Madagascar, and the Sudan, well, sure. But compared to the rest of the industrial democracies, it's not that clear-cut.
"Freedom" isn't a monolithic measurement, except to nationalist politicians. There are quite a few things I can do in various western European countries that I can't do in the United States. The converse is also true. For example, what Americans refer to as First and Fourth Amendment rights are considerably more open in some countries, while the American Second Amendment is pretty unusual for countries not ruled by hereditary warlords.
For my tastes, Germany is a much freer place. Someone who likes to own guns or is a Scientologist would probably feel differently. While it would certainly be nice if there were a most free or most open society, the truth is that you must ask "free and open in which ways?"
At my LUG, whenever a newbie asks, "What distro should I install?" A bunch of techies respond, "Debian."
Sad. My first year with Linux was a matter of endless frustration. It was only because I desperately wanted to get off the Microsoft treadmill that I stuck with it. I'm glad I did, because I love Linux, and the process of clawing my way to competence taught me a lot. (And made me a lot of money, since I went from selling furniture to being a sysadmin over the intervening eight years. Not everyone wants or needs to be a sysadmin, however.
I tried installing Debian recently. Frankly, I was appalled at how primitive it was and how many common packages (including some I depend on) were not included because they were not "free" enough. I would recommend Debian to someone who likes tinkering with their OS, just as I would (perhaps more strongly) recommend Slackware or the highly educational Linux From Scratch. I wouldn't recommend any of the above to a newbie unless I hated their guts and wanted them to stick to Windows.
Mandrake is quick and painless for inexperienced users and, in my experience, autoconfigures more hardware than any other distribution. Nor would I say it's just for newbies -- the experienced desktop user shouldn't have to manually configure anything unless the defaults don't suit him or her. Ever. It's just plain asinine to suggest that there is some kind of moral virtue in using unprofessionally packaged software.
Mandrake is also nice for certain server applications. Their Advanced Extranet Server project bundles pretty much every commonly conceivable Apache-related package in a series of modular RPMs. (Yes, I can compile it myself, but I get paid for producing results, not my hard-won understanding of the poorly documented and often poorly designed dependencies between the necessary packages.) Mandrake's install disk functionality means I can do one install and have some newbie intern roll out dozens of machines for the web server farm without a hitch and without working out the networking issues in advance.
Easy is only bad when it comes at the expense of power and flexibility, a la Microsoft. Mandrake delivers the full power of a feature-packed Linux distribution and manages to make it easy to use as well. I hope Mandrake manages to come through their current difficulties for the simple reason that they make good product that actually helps people get real work done, and they are to be commended for doing a much better job than the other commercial distributions which have had much larger resources to draw from.
Still, sustaining 1,500 LOC per day for a year and a half... that's beyond the productivity level of anyone I've ever seen. I personally have managed 4,500 per day for a period of about a week on occasion... but I wasn't sleeping much during that period.
I broke 1,000 LOC per day for about a week while working for an unnamed gigantic CPU monopolist. I was behind schedule, over budget, and had a hard deadline, and the code itself was fairly repetitive and not terribly efficient. Ordinarily, I'd figure I produce closer to 250 LOC per day during a normal coding period.
Provided this story isn't complete hogwash, my guess is that the reporter asked the boy about the writing the program and he answered that it consisted of 780,000 LOC and took him a year and a half to build. He probably neglected to mention that 90% of those lines were in libraries written by other people. He may not have even intended to be deceptive in any way, figuring that any fool would know that was the case, but not realizing that the reporter was a fool.
You could say that it's my responsibility to keep backups of my data.
Well, yes, that's exactly what I'd say. If you ever give critical data to anyone without having a backup, then you're probably going to screw yourself long before a careless tech has a chance to do so. Of course, if you engage in that kind of negligence while working for someone else, you'll probably be finding work in another field, anyway.
Personally, I think certifications for computer repair are meaningless. Given that assembling computers is only slightly more complicated than assembling Legos, and the cost of simply replacing a motherboard (or whatever) is often less than the cost of having a tech spend a couple of hours performing diagnostics, the additional cost imposed by "licensed" technicians would be pointless.
This is for PCs, mind you -- for high end machines like Sun servers or IBM mainframes, the vendor supplies trained technicians and no outside agency would have the expertise necessary to even design a certification program.
Perhaps more important than any of this is the painfully obvious fact that you can easily have crappy work done on your car in garages full of certified mechanics, and you can get excellent repair work done by shade-tree mechanics. Certification programs exist mainly as a marketing tool and a bar to entry for competitors, and utterly fail to address the main problem with auto mechanics, which is endemic fraud. There is no reason to believe that computer repair -- another field where fraud is endemic -- would be any better served by bogus certifications than the auto repair industry.
However, I hope a trend doesn't develop where OSS becomes known as "what happens to software when companies die."
I would. Not exclusively, certainly, but it's gotten to where I am reluctant to buy commercial software -- not for RMS-style ideological reasons -- but because the average niche-product software company seems to live no more than two or three years. This is not such a big deal with the major applications I'm dependent on, as Microsoft and Adobe will no doubt continue to bleed me for many years to come, but minor apps from minor companies die off at an alarming rate. I can fill a CD-R or two with all of the specialty graphics apps that I use whose producers are no longer around.
I'd like to see it become part of commonly accepted ethical business practice to release source code when a product will no longer be upgraded or supported. It's a great way of supporting the customers who supported you when market conditions or business strategy require you to otherwise abandon them.
So here's to TurboPower for their high standards both coming and going!
That is interesting. I had declined to get Pullman's books for my ten-year-old daughter because I had mistaken them for the same kind of thinly-veiled Christian allegory one finds in C.S. Lewis and Madeleine L'Engle. (Both of whom are excellent writers, but I'm too old to be suckered into their self-destructive superstitions.) I will have to stop at the bookstore on the way home and pick them up for her.
There is a certain attitude, as the original poster noted, that everything on the net should be free. But that's not the main problem.
The problem is that for the average person, the vast majority of what's on the web isn't worth paying for. It doesn't matter how easy it is to pay for, or how reasonable the cost is. There's just no demand for it.
Think of the web as the world's largest bookstore. I -- or anyone else -- might spend a couple of hours at the local Barnes and Noble browsing, but I don't buy everything I look at, and generally don't buy anything on the average visit. Now and then, I see something worth shelling out for, and I buy it. Brick and mortar retailers know this and understand that it's part of the game, and they don't sit around at night thinking up schemes for a per-book browsing fee. If they did, hardly anyone would ever come into the store, much less buy anything. For some reason -- perhaps the total lack of business knowledge that has afflicted online ventures from the beginning -- website producers just don't get this.
On the average day, I visit a couple dozen sites, including Slashdot, Freshmeat, CNN, Google, EurekAlert, various King Features and UFS comics, the New Online Books Page, a couple of hometown newspapers, etc. How many of these would I pay for if I had to? None of them. If I knew that the only way for them to stay online was for people like me to pay for them, I still wouldn't pay for them.
It's not that these aren't mostly fine sites, but the calculation being made here isn't their intrinsic value but rather the opportunity cost. If Site X was the only source of entertainment in my life, I'd surely pay a fair (maybe even unfair) price for it, but I have to ask myself -- would I rather get a book, a CD, rent a movie, spend a weekend at the beach, buy a camcorder, buy dinner, fix the car, etc., instead of subscribing to (or buying individual page views from) a website? In a word, no.
It's not just me, either, to judge from the state of the web content business. For the vast majority of people, the main value of the web lies in the fact that the content is free and convenient. Take that away, and very few people will be willing to pay for anything at all, and very few of them will do more than they do with the paper equivalent -- maybe subscribe to a newspaper, and maybe a couple of magazines. The sad and perhaps shocking truth is that the web just isn't very entertaining compared to traditional media.
It seems to me there is a difference between the police, who are guided by local, state and federal laws regarding use of evidence, and reporters, who have pretty much free reign under the US constitution in what they report.
There's another difference, which is why I'd rather have the press rummaging through my garbage than the police -- reporters aren't allowed to kick in my door and blow my head off. Being embarrassed by the press is a lot less dire of a possibility than having my next of kin washing my brains off the wallpaper.
There need to be more restrictions on state agencies authorized to use deadly force than on some random English major with a printing press.
Slashdot Mirror
Its not like there were any groundbreaking inventions coming out of mexico... besides maybe the double-sided bong?
There's more to creativity than technical achievement.
Mexico, among a few other Latin American countries -- Chile and Argentina spring most immediately to mind -- was the source of some of the best and most innovative literature of the 20th century, arguably far and away better than most of what came out of the English-speaking world. Mexican artists were significant contributors to the surrealist movement as well. The idea that the common cultural heritage of the Mexican people could be stolen by the Mexican government and pimped out for the profit of politicians and their favored private sector contractors is an utter disgrace for the government and a tragedy for both their citizens and ours.
I hope the Mexican people will fight this hijacking of their rich heritage.
What the hell do troop movements in the Middle East have to do with "News for Nerds"?
If I want shoddily reported, unsubstantiated rumors about the war, I can go to CNN. I count on Slashdot to give me shoddily reported, unsubstantiated rumors about technology.
Nicholas, given that the vast, vast majority of developers write software for internal corporate use and B2B applications, what possible significance can your 40% number have for the consumer desktop? Windows dominated the consumer desktop market before it penetrated a corporate market dominated by Unix, VMS, OS/360, etc., not the other way around. No one seriously disputes the rapid growth of Linux in the enterprise, but it seems to me that the corporate server market -- even the corporate desktop market -- has very little influence on what my grandmother or my daughter use on their desktops.
If 40% of developers are developing for Linux, where are the commercial apps?
I assume you mean boxed, shrink-wrapped apps. Almost no one develops consumer applications. If every developer who worked on consumer applications disappeared tomorrow, the total headcount of developers would not change significantly.
On the one hand, this is the reason that Microsoft's old scare stories about Open Source costing jobs were nonsense -- the vast majority of developers write software which is used in-house or sold at the enterprise level.
On the other hand, this is also the reason why Mr. Petreley's observations about developer percentages don't mattter one damn bit. 40% of developers? Who cares? What percentage of retail consumer software developers are developing for Linux? If the number is even as high 1%, I'd be deeply shocked.
Linux is rapidly taking over in the enterprise space. This no more means that Linux is poised to be your grandmother's OS than it means that Linux is poised to replace PS/2 and XBox.
Not that I don't wish that were the case, but world domination has been Real Soon Now since 1996.
What about the x86 version of Solaris?
There was also Coherent, and there's still QNX. Even Microsoft had a Unix offering at one point, a 16-bit OS called Xenix, which SCO should know, since Microsoft sold it to them.
Sports are important. [...] Organized sports are an unnatural event.
I have to agree, insofar as physical health is a prerequisite for full mental health, to say nothing of avoiding premature death. One firm we contract with consists entirely of severely obese network engineers. It's not their appearance I find shocking -- indeed, it's irrelevant since the quality of their work is good -- it's the fact that having a high probability of dying in their forties or fifties doesn't seem to faze them at all.
When I was in high school, I eschewed team sports and still do. All the crap coaches spew about the valuable skills one learns from team sports -- teamwork, dedication, etc. -- are better learned elsewhere in areas that don't encourage pointless aggression, meathead elitism, and date rape. I learned teamwork on the job out in the real world. I learned dedication in the face of adversity from the wide variety of spontaneous sources of adversity that the world provides.
There are plenty of ways to get exercise and enjoy the not inconsiderable drug-like rush that comes from sustained physical exertion without being a part of jock culture. Solitary sports like cycling and weight lifting were always my favorites. I have friends who speak highly of running, canoeing, climbing, and martial arts. It's a real shame that the abusive culture that goes with the popular team sports turns off so many people, geeks included, to other, more rewarding forms of athletics.
To return to the original topic, though, it's frankly bizarre that geeks have acquired this strange reputation for random violence when participants in team sports, organized religion, and the armed forces are much, much likelier sources of random murder. But then, those are the people who are in charge.
This isn't an issue limited to free software or even to software at all; it's a technology issue and, reduced to its bare essence, the question is whether an inventor is responsible for the use of his inventions.
The question of intellectual property is entirely beside the point. I remember that, when downloading Oracle for Linux, I was required to fill out an HTML form affirming that I would not use the product for the development of nuclear weapons. Somehow, I have a hard time picturing Saddam Hussein or Kim Jong Il saying, "Well, shit. I can't use Oracle for my clandestine nuclear program because of this license. I guess I'll have to use MySQL instead." The same is true of commercial software. Does anyone believe an export license (or for that matter, a licensing fee) would stop anyone from either purchasing a boxed copy in the US or Europe or just downloading a copy from alt.binaries.warez.ibm-pc? If you don't want your code to be misused, don't release it. If you release it, it eventually, inevitably will be misused.
Several years ago, I completed work on a library and set of tools for textual- and communication-traffic analysis. Among the things you could do with the tools was determining authority relationships between people in an organization on the basis of the patterns of their email communications. Another interesting application, which I tested with a full non-binaries Usenet feed, was a surprisingly effective system for determining the political affiliation of posters on the basis of their non-political postings. (For the curious, I used a sample group of 1,000 posters who made consistent ideological posts to political newsgroups as well as non-political posts to non-political newsgroups.) The accuracy rate over a six month period approached 95%.
Concluding that such code could be used by governments to track political dissidents, I was reluctant to release the code. Once John Ashcroft and John Poindexter appeared on the scene, I destroyed the source. This was probably pointless, as the algorithms being used are well-understood -- only the particular combination of algorithms was novel -- and the NSA probably has similar software written by much smarter guys than me, but when my own government, much less foreign tyrants, is arresting people without charges and holding them incommunicado in undisclosed locations, I didn't want to be responsible for contributing to the next round of political arrests.
Contrary to what the above seems to imply, I don't think inventors ought to be held responsible for the misuse of their inventions. I do think that inventors ought to be held responsible for failing to consider the potential consequences of misuse, however. In my case, I decided the potential benefits were outweighed by the potential abuses and decided not to release. Ultimately, that's all you can do if you are concerned about abuses, for the simple reason that the people who are likely to abuse your code are not going to be stopped by legal fictions like the GPL, copyrights, patents, or anything less substantial than the barrel of a gun.
Mandrake Linux has been beaten down by linux experts...
These are obviously "experts" who don't have to support non-technical users, which makes me wonder if their expertise is limited to their bedrooms in their parents' home. My experience has been that Mandrake is a pretty sweet deal if you're the only sysadmin in the department and have better things to do with your time than configure crap for the receptionist.
Here's the catch: I will not purchase Windows!
So find the model you want and look for it on eBay or some other outlet for used hardware.
Under the current state of US law, unauthorized access to a computer system is a federal crime. (I can't speak to EU laws, but I suspect parallels exist.) If Company X says, "You must use Internet Explorer 5.5 to access this site," then you must use IE 5.5. Of course, it would be just plain stupid to do so, but it's their computer system, and they get to decide who is authorized.
To judge from most of the comments here, the fact that it is incredibly stupid to impose such restrictions has obscured what is actually a legally unambiguous situation. Just because it's dumb doesn't mean it's not legal.
That an http server is nominally "public" doesn't mean diddly here. Any number of http servers provide for member- or employee-only access. The brick and mortar parallel would be those signs that say things like, "No shirt, no shoes, no service."
It is surprising that so few people have touched on the reason why companies might object to the distribution of Perl modules designed to harvest data from their sites: bandwidth costs and site performance. It doesn't take too many cron jobs banging on a site every minute -- and being ignored by their users most of the time -- to degrade site performance for "live" users and run up steep bandwidth bills.
Now, there is certainly no legal basis for Company X to demand that CPAN remove the modules, though it is hardly out of line to ask nicely. But there is firm legal grounds to prohibit anyone from actually using those modules.
Legal action is probably the wrong way to handle this, though. Having written fairly complicated web scrapers before, I know how easy it would be to make a site virtually impossible to harvest. Rather than make a big stink about the Perl programmers who contribute to CPAN, Company X would be well-advised to hire a good Perl programmer to thwart automated harvesters.
Rather than releasing new versions, perhaps the distro vendors should eliminate the concept altogether in favor of the sort of seamless, continuous per-package upgrading I suspect most people would like to see. Does anyone really care what version number a distribution has? I suspect more people care what kernel and security patches and application versions they're running.
For the user, this would have the advantage of being able to click a button or insert the latest update CD and upgrade all of the necessary packages. (We presume, of course, that you could elect to forego certain upgrades -- one might wish to continue running Apache 1.3.x instead of a 2.x version.)
For the vendor, this would be an obvious opportunity to sell subscriptions as well as avoid the endless cost of producing shrinkwrapped distributions.
Of course -- of course -- this would require greater effort on the part of vendors to make sure that the upgrade process is robust and seamless so as to avoid the problems M$ customers have with their so-called Service Packs, but it ought to be doable.
I'm more amazed that no one asked her why she did it... Do we look down on non-computer people so much that we don't even bother to ask anymore why they do stupid things?
Yes.
OTOH, there are already big disincentives to shopping online: shipping -- which is often more than sales tax would be -- and delivery time. Add sales tax to that, and the only cases in which buying online makes sense would be
I suspect convenience is less of a selling point in the current state of the economy than it would be at other times.
For books and CDs, which are about the only things I buy online regularly, it's probably easier and cheaper for me to call my local Barnes and Noble or Borders and ask them to special-order something and drive five minutes to pick it up when it arrives. Unless I can get a used copy more cheaply online.
The floppy persists because it is the only convenient way of transporting small files without a network using inexpensive reusable media that everyone can read. All of the floppy's competitors fail on one or more of those three requirements. The only medium that comes close these days would be the rewritable CD, but burning a CD is a lot less convenient than copying a file to a floppy, and -- at least in my experience -- CD-RWs are a lot less reliable than 3.5" floppies if they get passed around much. Floppies, after all, don't scratch easily.
It seems to me that not having a floppy drive would make it harder to flash BIOS firmware. I suppose it could be done with an El Torito image, though.
One of my earliest memories is seeing the last of the Apollo launches from the beach in Florida. I watched the first launch of the Columbia with my class in school. I got to see it in person, once, when it was being kept briefly at Ft. Campbell, KY, to avoid some hurricane or other.
My ten-year-old doesn't understand why this is a big deal. Space travel, to her, is like CDs and PCs and microwave ovens -- a routine part of the world as it is. She was born after the cold war, after the glory days of the space program. Maybe when she's older, she'll understand that the space program transcended all the petty factional divisions and murderous religious and political ideologies of this sad world and was for a lot of us a shining example of the very best of the human race and a beacon of hope for a better future.
Growing up in the 70's, astronauts were the only people I ever really thought of as heroes. NASA was the only government agency I could admire, whatever its faults, without a trace of cynicism. That hasn't changed.
I wish I could somehow take my daughter back in time to that day on the beach when I looked southward towards the Cape and saw a Saturn V rise from the horizon on a pillar of flame. Maybe then she could understand why her parents were crying in front of the TV today. Instead, the best I could manage to say was, "They were astronauts. Our dreams went with them."
Godspeed, folks. You were the best of the best. You will not be forgotten.
Trademarks are going to get pretty difficult if simply rhyming with an existing trademark is going to count as an infringement. I don't know about French, but a quick Perl script chewing on CMU's pronouncing dictionary reveals only a couple dozen single-syllable endings to English words, and only a few hundred two-syllable endings.
Sure, the US is open compared to most of the countries in the world, but it's not as open as open source.
I hear this a lot from Americans. (Don't get your nationalist knickers in a wad; by accident of birth, I'm one of you, too.) The problem is that it is only a half-truth. If by "most of the countries in the world" you mean to include Brunei, Madagascar, and the Sudan, well, sure. But compared to the rest of the industrial democracies, it's not that clear-cut.
"Freedom" isn't a monolithic measurement, except to nationalist politicians. There are quite a few things I can do in various western European countries that I can't do in the United States. The converse is also true. For example, what Americans refer to as First and Fourth Amendment rights are considerably more open in some countries, while the American Second Amendment is pretty unusual for countries not ruled by hereditary warlords.
For my tastes, Germany is a much freer place. Someone who likes to own guns or is a Scientologist would probably feel differently. While it would certainly be nice if there were a most free or most open society, the truth is that you must ask "free and open in which ways?"
This lowers Finland on my list of vacation spots.
Same here. Previously, Finland was right up there with the Kerguelen Islands on my list of places to relax on the beach.
At my LUG, whenever a newbie asks, "What distro should I install?" A bunch of techies respond, "Debian."
Sad. My first year with Linux was a matter of endless frustration. It was only because I desperately wanted to get off the Microsoft treadmill that I stuck with it. I'm glad I did, because I love Linux, and the process of clawing my way to competence taught me a lot. (And made me a lot of money, since I went from selling furniture to being a sysadmin over the intervening eight years. Not everyone wants or needs to be a sysadmin, however.
I tried installing Debian recently. Frankly, I was appalled at how primitive it was and how many common packages (including some I depend on) were not included because they were not "free" enough. I would recommend Debian to someone who likes tinkering with their OS, just as I would (perhaps more strongly) recommend Slackware or the highly educational Linux From Scratch. I wouldn't recommend any of the above to a newbie unless I hated their guts and wanted them to stick to Windows.
Mandrake is quick and painless for inexperienced users and, in my experience, autoconfigures more hardware than any other distribution. Nor would I say it's just for newbies -- the experienced desktop user shouldn't have to manually configure anything unless the defaults don't suit him or her. Ever. It's just plain asinine to suggest that there is some kind of moral virtue in using unprofessionally packaged software.
Mandrake is also nice for certain server applications. Their Advanced Extranet Server project bundles pretty much every commonly conceivable Apache-related package in a series of modular RPMs. (Yes, I can compile it myself, but I get paid for producing results, not my hard-won understanding of the poorly documented and often poorly designed dependencies between the necessary packages.) Mandrake's install disk functionality means I can do one install and have some newbie intern roll out dozens of machines for the web server farm without a hitch and without working out the networking issues in advance.
Easy is only bad when it comes at the expense of power and flexibility, a la Microsoft. Mandrake delivers the full power of a feature-packed Linux distribution and manages to make it easy to use as well. I hope Mandrake manages to come through their current difficulties for the simple reason that they make good product that actually helps people get real work done, and they are to be commended for doing a much better job than the other commercial distributions which have had much larger resources to draw from.
Still, sustaining 1,500 LOC per day for a year and a half ... that's beyond the productivity level of anyone I've ever seen. I personally have managed 4,500 per day for a period of about a week on occasion ... but I wasn't sleeping much during that period.
I broke 1,000 LOC per day for about a week while working for an unnamed gigantic CPU monopolist. I was behind schedule, over budget, and had a hard deadline, and the code itself was fairly repetitive and not terribly efficient. Ordinarily, I'd figure I produce closer to 250 LOC per day during a normal coding period.
Provided this story isn't complete hogwash, my guess is that the reporter asked the boy about the writing the program and he answered that it consisted of 780,000 LOC and took him a year and a half to build. He probably neglected to mention that 90% of those lines were in libraries written by other people. He may not have even intended to be deceptive in any way, figuring that any fool would know that was the case, but not realizing that the reporter was a fool.
You could say that it's my responsibility to keep backups of my data.
Well, yes, that's exactly what I'd say. If you ever give critical data to anyone without having a backup, then you're probably going to screw yourself long before a careless tech has a chance to do so. Of course, if you engage in that kind of negligence while working for someone else, you'll probably be finding work in another field, anyway.
Personally, I think certifications for computer repair are meaningless. Given that assembling computers is only slightly more complicated than assembling Legos, and the cost of simply replacing a motherboard (or whatever) is often less than the cost of having a tech spend a couple of hours performing diagnostics, the additional cost imposed by "licensed" technicians would be pointless.
This is for PCs, mind you -- for high end machines like Sun servers or IBM mainframes, the vendor supplies trained technicians and no outside agency would have the expertise necessary to even design a certification program.
Perhaps more important than any of this is the painfully obvious fact that you can easily have crappy work done on your car in garages full of certified mechanics, and you can get excellent repair work done by shade-tree mechanics. Certification programs exist mainly as a marketing tool and a bar to entry for competitors, and utterly fail to address the main problem with auto mechanics, which is endemic fraud. There is no reason to believe that computer repair -- another field where fraud is endemic -- would be any better served by bogus certifications than the auto repair industry.
However, I hope a trend doesn't develop where OSS becomes known as "what happens to software when companies die."
I would. Not exclusively, certainly, but it's gotten to where I am reluctant to buy commercial software -- not for RMS-style ideological reasons -- but because the average niche-product software company seems to live no more than two or three years. This is not such a big deal with the major applications I'm dependent on, as Microsoft and Adobe will no doubt continue to bleed me for many years to come, but minor apps from minor companies die off at an alarming rate. I can fill a CD-R or two with all of the specialty graphics apps that I use whose producers are no longer around.
I'd like to see it become part of commonly accepted ethical business practice to release source code when a product will no longer be upgraded or supported. It's a great way of supporting the customers who supported you when market conditions or business strategy require you to otherwise abandon them.
So here's to TurboPower for their high standards both coming and going!
That is interesting. I had declined to get Pullman's books for my ten-year-old daughter because I had mistaken them for the same kind of thinly-veiled Christian allegory one finds in C.S. Lewis and Madeleine L'Engle. (Both of whom are excellent writers, but I'm too old to be suckered into their self-destructive superstitions.) I will have to stop at the bookstore on the way home and pick them up for her.
There is a certain attitude, as the original poster noted, that everything on the net should be free. But that's not the main problem.
The problem is that for the average person, the vast majority of what's on the web isn't worth paying for. It doesn't matter how easy it is to pay for, or how reasonable the cost is. There's just no demand for it.
Think of the web as the world's largest bookstore. I -- or anyone else -- might spend a couple of hours at the local Barnes and Noble browsing, but I don't buy everything I look at, and generally don't buy anything on the average visit. Now and then, I see something worth shelling out for, and I buy it. Brick and mortar retailers know this and understand that it's part of the game, and they don't sit around at night thinking up schemes for a per-book browsing fee. If they did, hardly anyone would ever come into the store, much less buy anything. For some reason -- perhaps the total lack of business knowledge that has afflicted online ventures from the beginning -- website producers just don't get this.
On the average day, I visit a couple dozen sites, including Slashdot, Freshmeat, CNN, Google, EurekAlert, various King Features and UFS comics, the New Online Books Page, a couple of hometown newspapers, etc. How many of these would I pay for if I had to? None of them. If I knew that the only way for them to stay online was for people like me to pay for them, I still wouldn't pay for them.
It's not that these aren't mostly fine sites, but the calculation being made here isn't their intrinsic value but rather the opportunity cost. If Site X was the only source of entertainment in my life, I'd surely pay a fair (maybe even unfair) price for it, but I have to ask myself -- would I rather get a book, a CD, rent a movie, spend a weekend at the beach, buy a camcorder, buy dinner, fix the car, etc., instead of subscribing to (or buying individual page views from) a website? In a word, no.
It's not just me, either, to judge from the state of the web content business. For the vast majority of people, the main value of the web lies in the fact that the content is free and convenient. Take that away, and very few people will be willing to pay for anything at all, and very few of them will do more than they do with the paper equivalent -- maybe subscribe to a newspaper, and maybe a couple of magazines. The sad and perhaps shocking truth is that the web just isn't very entertaining compared to traditional media.
It seems to me there is a difference between the police, who are guided by local, state and federal laws regarding use of evidence, and reporters, who have pretty much free reign under the US constitution in what they report.
There's another difference, which is why I'd rather have the press rummaging through my garbage than the police -- reporters aren't allowed to kick in my door and blow my head off. Being embarrassed by the press is a lot less dire of a possibility than having my next of kin washing my brains off the wallpaper.
There need to be more restrictions on state agencies authorized to use deadly force than on some random English major with a printing press.