Re:Restricting Free (as in speech) Software
on
P2P vs. The Clones
·
· Score: 4, Informative
This doesn't change the main point of your argument, but you are slightly misinformed about what the GPL says about selling software. The GPL allows you to take the software and sell it at any price you want, bundled or not. However, you must either
A) include the source with sold binary, or B) make it available seperately at additional cost.
It is just this additional cost that is limited to reasonable compensation.
People have been making comments like this on slashdot for as long as I can remember, and I have to wonder how many of them really get paid to write shrink-wrapped software. Statistically, it is a very small amount, with very many more writing in-house software. The thing is, for all of that large majority of software developers, open source software won't hurt you at all. The only people who will use your code are paying you up front to write it, so it's not like you need the copyright protection to allow you to make money selling it. The code you are writing is probably very tailored to the specific needs of your employer and so "the competition" really wouldn't be helped much by having access to it. In most cases your code might as well be open source, and wouldn't make any difference whatsoever to the business model which is feeding you and paying your mortgage. If anything, having the common bits - things you would otherwise license from a third party - open source will just make your life easier.
And if you really want to, you can make money directly writing open source software as well. It isn't easy, and you have to be something of an entrepreneur. But it certainly can be done, and from what I can see, the people doing it are living thier dreams, and are being compensated quite well for it. If you don't want that sort of risk, than shrink wrapped software isn't really the place to be anyway. Trying to make it big creating the next killer app is just as hard, if not harder, than creating a career around OSS programing. If you want to change the world, it will be a risky no matter how go about it - that's just life. If you want a stable job, those are going to be in IT and they will only gain from open source software.
Slightly off topic, but I haven't found a solution elsewhere. The OS X version of Quake III (1.32-altivectest2 - hasn't been updated for about a year), has an issue with crashing when you get shot in the face with a projectile (the plasma gun is most noticable). I have been trying to find a solution/workaround to this problem but have not had any luck - as is common, I can find plenty of posts from people who also have the problem, but none that provide a solution. Someone mentioned turning down the max framerate, so I set com_maxFPS to 60 (same as my monitor refresh rate), and went ahead and turned all the graphics options to the highest possible quality. This seems to have decreased the frequency of crashes but did not solve the problem completely. Is this the correct variable? How low does it need to be set? Any word on whether these new drivers make the problem better or worse?
PS - I have the Geforce FX 5200, but have seen reports with this happening with ATI cards as well.
Sure it is. It is just an observational science rather than an experimental science, which means that it takes much longer to test ideas as you can't do controlled experiments. Some economic theories have become well founded over time (although they are certainly incomplete). The evidence for other ideas have far to few data points and far to many external factors to come to any real conclusion. Of course short of hard data, one wants to have at least a best guess answer, and noone seems to be able to say "I don't know", so more subjective judgement is often put on top of the science. Which is fine except for the few egotistic idiots that will try to treat their best guess as scientific fact, but what can you do?
Just to pound some more on what other people have already said, I think this is a very bad idea. Legal language is different from normal language, and many words have precise legal definitions that are different from their coloquial(sp?) use. There are also important legal concepts that are critical to understand if you want your laws to say what you mean. What you are proposing is akin to fireing all the experienced members of a development team and replacing them with people who not only are not familiar with the existing codebase, but do not even know how to program!
But most importantly I don't think that this issue is even the problem. Nothing I have ever seen causes me to think that the reason our laws are complex is because those congressmen lawyers want job security. In my judgement, the reason that we have inanely complex laws are because every Tom Dick and Harry (ie corporate and citizen special interest groups) wants their own stupid little law when in reality, existing laws should be suitable (or nearly so). The best way for congress critters to get relected is to pass all the stupid laws that their large, politically active and generous constituents want.
So the complexity in our law system is directly related to the workings of democracy itself, not because the congressmen are lawyers. So how can we fix this? (Warning: brainstorming follows - these are ideas not a thesis paper:)
What if we made laws harder to pass? That would cut down on the number but would also make congress less able to get things done. IMHO, this is a good thing in some cases, but not others, so it would be nice if we had a higer law which changed less often, and a lower one that could be agile, but was superceded by the higher. Also part of the complexity is due to old cruft, so sunsetting might be a good idea in some cases. Lastly another reason that people want specific laws as opposed to using general ones is that they want to be certain on what the outcome of cases will be rather than hoping the law is on their side and later having a judge interpret otherwise. So how about this:
Any law limiting the freedoms or rights of citizens must be passed with 70% majority.
Any law which mandates expenditure must have sunset (or reconsideration) provisions not to exceed 4 years.
Congress may pass guidlines to any law with a 50% majority which have the same weight as legal precidence.
This would make it less necisarry to have complex laws, and harder to restrict the rights of the citizenship, and might even lead to an atmosphere of more restrained law making.
I think it's funny that every couple months someone posts on slashdot about how X will finally end the debate between the creationists and the evolutionists. This new finding wouldn't settle the debate between evolutionists and creationists any more than Dr. Miller's 1953 experiment (recreating this environmnent in the lab) did.
Most creationists acknowledge that the low level aspects of our bodies operate according the laws of chemistry. They acknowledge that natural selection is real and happens. And they acknowledge that there have been macroscopic and DNA changes to species over time (although most contend about the length of time). What they don't acknowledge is that natural selection is what caused new species to come about. And there is nothing short of having a first hand view of millions of years of life that will ever prove that to thier satisfaction.
Probably not. The open source Helix Player cannot play Real Media file formats, only open ones like Ogg Vorbis, etc. To play the Real Media formats you must use the proprietary Real Player binary. So basically Helix is no different than MPlayer in that the open source part of it isn't very usefull by itself without the binary codecs. AFAIK, Real doesn't distribute a binary dll of their codecs, just the full application, which makes it more difficult to support than the rest of the codecs and I don't see how having the Helix will help with that.
Thats weird. You'd think that that when they created ID3v2 they would have included some way for an application to determine if v1 or v2 is being used. Is the ID3 format really that brain-dead, or are the linux mp3 players just assuming that everyone is using the latest and greatest?
at sandia labs with 4x the projectors. I don't think they have a cool algorithm for the seam matching, like the one in the story though. The neat thing about the sandia one system was what was feeding it - a 64 node cluster that could render realtime 3d visualizations of simulations done on the ASCI super computers. I don't know what the polygon count was on that thing but each projector was 1280x1024 and I couldn't see any corners when looking at a very detailed model (the one shown in the press release actually:).
It is a good idea. Mozilla is a very large codebase with a reletively small number of developers. Therefore they don't have the fast turn around time for fixing critcal bugs that other projects do. They are already fixing bugs as fast as they can get around to it. If you wan't this process to go faster join the development team.
Security through obscurity is not completely worthless - it does one thing and that is to buy you more time, and that is all this is being used for. If you have not been part of the mozilla development team, then you do not know the codebase well enough to fix critcal bugs in a manner that doesn't potentially introduce more problems. And if you have been a part of the development team then you have access to the critical bug reports, so you can fix them.
It is always better to fix bugs rather than conceal them, but publishing the details of bugs that you haven't fixed is foolish. (This is opposed to just publishing the existence of vulnerabilities so your users are aware of potential problems).
Security is a matter of tradeoffs and to make good decisions, it is not enough to say "this is good", "this is bad" - You have to understand why practices are good or bad so you can understand when they should be used and to what extent they can be trusted, and do the best with the resources you have.
Side note / Prerefutation of slashbots:
No I am not defending mozilla because it is OSS, nor would I condem a comercial company for doing the same. I would condem them for claiming that their product is secure while hiding the existence of critcal bugs. Some OSS projects are good about bugs others are not. For example I am absolutely convinced that Apache is inherently more secure than MS IIS. I cannot say the same about Sendmail vs MS Exchange. Mozilla I am not sure about. They do some things better, like safer default settings. But like IE they also allow just about everything in the browser to be scripted, and I have little confidence that there aren't many critcal security bugs in all that code.
That won't help. The modifiers are not keyboard shortcuts, and cannot be changed that way. I haven't been able to find a way to do so, and while I am definately agree with the gnome philosophy of minimizing configurability to what is useful, that really is a necisarry configuration if you want nice integrate into different environments.
I will third that comment, but also add that the slashdot team updates the slashcode on the live system on a weekly basis, so just because they haven't been fixed it yet doesn't necisarrily mean that they are not listening.
Their techniques may work, but that site screams snake oil scam artist. I these people really wanted to help people like they say they do they should be publishing in journals and making the information available to the community. The way they are currently handling themselves they are certain to be dismissed as frauds because for all we know they are.
My reservation with that is that if you are going to give them the responibilities of being an adult at age 13 you really ought to give them the freedoms as well. If you don't that is just plain contrary to liberty, and unfair.
If you do, then they are no longer subject to their parents at 13, and I'm sorry but kids that age need parents to watch over them as much or more than any time in thier life. Preferably it will more hands off, but parents need the ability to step in when the situation merits it. Doing what you say would basically be giving state blessing for parents to stop taking responsibility for thier kids actions at 13 and would be a very bad policy in my opinion.
The day emacs became sentient
on
TMBG on DRM
·
· Score: 3, Funny
Long ago in the computer days of yore, artificial intelligence researchers created a revolutionary new language. A language which was flexible and elegent. A language which could better model thought. A language in which rested the hope of finally creating a sentient artificial being. This language was lisp.
They all began working most diligently creating programs which behaved intellegently But these programs were lacking still. For when prompted they would only say
t
or
nil
.
Yet a researcher in the great white north continued on Feeding song after song into to his very own atomiton Then at last one tiresome night in 84 he played a track he never played before. And queried the program:
(giantsp they)
To which it responded
might be, eh
/wanting to work that last line into a joke for years:)
But the biggest problem with the three laws isn't that they are incomplete for determining the best course of action for a robot, which is what Asimov explored, but the fact that they are currently (and possibly inherently) impossible to implement.
How the heck is a robot supposed to accurately judge that whether a random unique action in a unique situation will cause harm to a human or himself? Humans can't even do this. If we were to create an artificial intellegence that was fully capable of making these decisions, would we even be able to put limits on what it decides?
Regardless of the answer to that philisophical question, we will have the technology to produce usefull robots long before we have the technology to produce 3-Law abiding robots so we need to come up with practical ways of making them as safe as possible, within their limited capabilites.
Oh yeah, they are doing some kick-ass things that we will see in the next release, absolutely. Just saying that we can't give them credit for the advances in this current release. Then again, the work is mostly being done by the same people, just different organizers.
No I'm not against any graphical configuration tools or this and that. I'm just against breaking the rule of changing the default UNIX tradition of configuration files. Any graphical tool should be like Webmin, which leaves the structure as it is.
This is what brought me back to slackware. I started with RH 4 but could never get it to work with my hardware. So I tried slackware and really enjoyed it. I learned so much about how Unix works on that slackware version. Anyway since then I have tried a different distro each time I am ready to do a complete OS upgrade. Here is what I learned.
As far as packaging goes rpm sucks unless you verify or build your own because the majority of 3rd party package builders do it wrong. At that point it's just as easy to go with slackware or gentoo. Apt-get seems really nice. Unfortuneatly, I didn't get much time with Debian (one week to install, then two weeks later my harddrive dies).
As far as configuration goes, those GUI tools are a pain. I tried 4 releases of RedHat and got to learn 4 ways of setting up PPP, and each of which seemed to get progressively worse. And of course once you use the GUI tools, it creates it's own config files from which the unix ones are generated. So after the easy way fails if you want to do it the manual way, you first have to figure out how to disable the distro provided tools, which is not always easy. The *drake tools are the flakiest things I have ever seen seen. They basically just issue some script commands and don't do any error handling. If something goes wrong, the window just disappears and you are left wondering if it worked or if not why it didn't work, and what state you system is in. Totally lame. Yast is the nicest of the bunch, but again you really need to decide to let Yast do everything, or do everything the manual way, because otherwise you will tromple all over each other.
As far as I am concerned, you can take your GUI configuration tools and keep them. Slackware may not be the easiest distro, but it is by far the least complicated. Even better, all the time I spend getting things to work on slackware, I am actually learning about how Linux works rather than figuring how to get around some broken config tool. That is the first thing that struck me when I started using slackware again. With the other distros I had gotten frustrated with all the maintainance I was doing that was all related to stuff I would never use again - fixing dependency errors, unbreaking harddrake - and this ended up driving me to Mac OS X for my main computer. With slackware I don't have to think about those kinds of problems, and I actually enjoy the problem-solving and discovering that I do have to deal with. It reminds me why I originally became so absorbed with linux in the first place.
To expand on what a couple of people have said, slackware has had a default XF86Config file for quite some time (always?) However in the past it defalted to using the Framebuffer X driver, and it 10 it defaults to using the Vesa driver. In both cases you need to change the XF86Config (now xorg.conf) to use a specific driver if you want decent performance.
This change is not (directly at least) related to the change to Xorg - they could have done it with XFree86 as well. Also while I agree that Xorg is the way to go, alot of the technical praise they are getting is misdirected. There is really almost signifcantly different between it and XFree86 4.4, and most of the improvements that people see in Xorg are really improvements in XFree 86 since the 4.3 series.
That depends. Like Paul said (paraphased), to die is gain because he gets to go to heaven, but to live is also to gain because he can save more people, so he is divided on which is best. This is pretty much the same thing - if the end times come soon then it is good for the saved but not the unsaved. Who is more important? Also, depending on whether you belive pre-trib or post-trib, and concidering how difficult the end times will supposedly be, there is good reason for a christian to want to die before the tribulations come.
Then again for those fundamentalists who would like to see everyone go to hell, but think that they will be whisked up into heaven in time for tea, then yeah, they should probably be cheering this stuff on.
Yeah, although to some extent I can see where they are comming from. I was our local ACM chapter president and would work with Microsoft rep (a real programmer), as Microsoft is nice about supporting local competitions (providing prizes and what not). So of course he was always trying to talk me into intervewing with Microsoft. During one of these conversations I brought up the issue of a monopoply hurting the market as a whole, and he admitted that he had thought about that quite a bit, and to some extent agreed that it would be better if there was no monopoly. But he wanted to write software that would matter - that would have some effect on the world (presumably positive) and there was no better place to do this than Microsoft.
And he's right. I mean look at Apple they kick so much ass, and yet only 4% of computer users even notice what they do. Linux is even less outside of the server room. Microsoft is where it's at right now, and hopefully that will change, but it is going to be a long time. I'll admit that that matters to me. I get some satisfaction just from knowing I wrote good software, but I do want my work to matter. I don't want it to sit in a closet somewhere.
I didn't go to work for Microsoft, but I can understand why someone would. Now I have a job writing support software. I work with the people who use my software everyday and I get to see first-hand how it makes thier jobs easier (and sometimes not:).
Exactly. And it is not just Microsoft - it is a general weakness in the shrink-wrapped software business model. In that model you depend on sales of the previous version to fund additional features for the next version which in turn drives sales, and the cycle continues. The problem is that at some point your product becomes mature, and you have already implemented 90% of the features that 90% of the people want. Now there is still alot of potential functionality to be added but, each feature will appeal to only a small audience. Therefore even though you may have done as much work between versions 5 and 6 as you did between 2 and 3, you have deminishing returns on the number of purchases. Lastly, ever since the dot-com boom ended the number of first-time purchases (as opposed to upgrades) has been going down dramatically as well, so you are much more dependent on upgrades sales, which we just determined will also go down with time.
So basically the shrinkwrapped software business model sucks for mature software. Unless you can keep improving the software in a way that appeals to a large number of people, you will not be able to generate enough money from sales to continue development at your current pace. Then your product will stagnate, and newcommers who focus on different niche features that you don't have will eat away at your market share.
Once your software becomes mature, you really need to move away from the shrinkwrapped business model to some type of service business model. Interestly enough, OSS kicks ass in just about any service business model. If you are being paid for the act of writing and deploying software, rather then selling it as a product, it doesn't matter if you control the software or not. It just matters that you have the experience and talent to improve an existing piece of software (ie helps alot if you wrote the software to begin with).
OSS has the opposite problem - it is easy to get paid to improve mature products, but getting a piece of software to maturity is harder (financially).
You are right - it doesn't really make any difference for hardware devices if you have to licence the codec or not - peanuts in the overall cost. Same with proprietary software - the licencing costs won't add much to the cost of the product, so no big deal.
But open source software is a problem. It is illegal to distribute an mp3 encoder without paying the licence fees, which makes it impossible to have a legal open source mp3 encoder (or DVD players, or anything else that is patented) because to be open the licence must allow anyone to redistribute it without restriction. The way that LAME gets around that issue is that "LAME Ain't an MP3 Encoder" - it is a reference implementation for educational use only. This doesn't really solve the problem, it just makes the users criminals instead of the developers and distributers.
So we don't like patented formats for that reason - because it is illegal for us to use those formats on our open source systems, and if we are restricted to use open formats on our computer, it would be really nice to use them on our handheld music players as well, which fortunately is a very viable option, contrary to what the whiners here say, and thanks to some cool hardware companies.
Re:I was kidnapped by one of those Robots!
on
Robots in Hospitals
·
· Score: 1
Hehe. Just be happy you weren't standing at the top of the stairs, like I was. Protector robot my ass.
Slashdot Mirror
This doesn't change the main point of your argument, but you are slightly misinformed about what the GPL says about selling software. The GPL allows you to take the software and sell it at any price you want, bundled or not. However, you must either
A) include the source with sold binary, or
B) make it available seperately at additional cost.
It is just this additional cost that is limited to reasonable compensation.
People have been making comments like this on slashdot for as long as I can remember, and I have to wonder how many of them really get paid to write shrink-wrapped software. Statistically, it is a very small amount, with very many more writing in-house software. The thing is, for all of that large majority of software developers, open source software won't hurt you at all. The only people who will use your code are paying you up front to write it, so it's not like you need the copyright protection to allow you to make money selling it. The code you are writing is probably very tailored to the specific needs of your employer and so "the competition" really wouldn't be helped much by having access to it. In most cases your code might as well be open source, and wouldn't make any difference whatsoever to the business model which is feeding you and paying your mortgage. If anything, having the common bits - things you would otherwise license from a third party - open source will just make your life easier.
And if you really want to, you can make money directly writing open source software as well. It isn't easy, and you have to be something of an entrepreneur. But it certainly can be done, and from what I can see, the people doing it are living thier dreams, and are being compensated quite well for it. If you don't want that sort of risk, than shrink wrapped software isn't really the place to be anyway. Trying to make it big creating the next killer app is just as hard, if not harder, than creating a career around OSS programing. If you want to change the world, it will be a risky no matter how go about it - that's just life. If you want a stable job, those are going to be in IT and they will only gain from open source software.
Slightly off topic, but I haven't found a solution elsewhere. The OS X version of Quake III (1.32-altivectest2 - hasn't been updated for about a year), has an issue with crashing when you get shot in the face with a projectile (the plasma gun is most noticable). I have been trying to find a solution/workaround to this problem but have not had any luck - as is common, I can find plenty of posts from people who also have the problem, but none that provide a solution. Someone mentioned turning down the max framerate, so I set com_maxFPS to 60 (same as my monitor refresh rate), and went ahead and turned all the graphics options to the highest possible quality. This seems to have decreased the frequency of crashes but did not solve the problem completely. Is this the correct variable? How low does it need to be set? Any word on whether these new drivers make the problem better or worse?
PS - I have the Geforce FX 5200, but have seen reports with this happening with ATI cards as well.
Sure it is. It is just an observational science rather than an experimental science, which means that it takes much longer to test ideas as you can't do controlled experiments. Some economic theories have become well founded over time (although they are certainly incomplete). The evidence for other ideas have far to few data points and far to many external factors to come to any real conclusion. Of course short of hard data, one wants to have at least a best guess answer, and noone seems to be able to say "I don't know", so more subjective judgement is often put on top of the science. Which is fine except for the few egotistic idiots that will try to treat their best guess as scientific fact, but what can you do?
But most importantly I don't think that this issue is even the problem. Nothing I have ever seen causes me to think that the reason our laws are complex is because those congressmen lawyers want job security. In my judgement, the reason that we have inanely complex laws are because every Tom Dick and Harry (ie corporate and citizen special interest groups) wants their own stupid little law when in reality, existing laws should be suitable (or nearly so). The best way for congress critters to get relected is to pass all the stupid laws that their large, politically active and generous constituents want.
So the complexity in our law system is directly related to the workings of democracy itself, not because the congressmen are lawyers. So how can we fix this? (Warning: brainstorming follows - these are ideas not a thesis paper
What if we made laws harder to pass? That would cut down on the number but would also make congress less able to get things done. IMHO, this is a good thing in some cases, but not others, so it would be nice if we had a higer law which changed less often, and a lower one that could be agile, but was superceded by the higher. Also part of the complexity is due to old cruft, so sunsetting might be a good idea in some cases. Lastly another reason that people want specific laws as opposed to using general ones is that they want to be certain on what the outcome of cases will be rather than hoping the law is on their side and later having a judge interpret otherwise. So how about this:
This would make it less necisarry to have complex laws, and harder to restrict the rights of the citizenship, and might even lead to an atmosphere of more restrained law making.
I think it's funny that every couple months someone posts on slashdot about how X will finally end the debate between the creationists and the evolutionists. This new finding wouldn't settle the debate between evolutionists and creationists any more than Dr. Miller's 1953 experiment (recreating this environmnent in the lab) did.
Most creationists acknowledge that the low level aspects of our bodies operate according the laws of chemistry. They acknowledge that natural selection is real and happens. And they acknowledge that there have been macroscopic and DNA changes to species over time (although most contend about the length of time). What they don't acknowledge is that natural selection is what caused new species to come about. And there is nothing short of having a first hand view of millions of years of life that will ever prove that to thier satisfaction.
Probably not. The open source Helix Player cannot play Real Media file formats, only open ones like Ogg Vorbis, etc. To play the Real Media formats you must use the proprietary Real Player binary. So basically Helix is no different than MPlayer in that the open source part of it isn't very usefull by itself without the binary codecs. AFAIK, Real doesn't distribute a binary dll of their codecs, just the full application, which makes it more difficult to support than the rest of the codecs and I don't see how having the Helix will help with that.
I have seen it with IE, Safari, and Firebird, using Windows 2k, OS X, and Slackware. 503 errors have nothing to do with the browser, just the server.
Thats weird. You'd think that that when they created ID3v2 they would have included some way for an application to determine if v1 or v2 is being used. Is the ID3 format really that brain-dead, or are the linux mp3 players just assuming that everyone is using the latest and greatest?
at sandia labs with 4x the projectors. I don't think they have a cool algorithm for the seam matching, like the one in the story though. The neat thing about the sandia one system was what was feeding it - a 64 node cluster that could render realtime 3d visualizations of simulations done on the ASCI super computers. I don't know what the polygon count was on that thing but each projector was 1280x1024 and I couldn't see any corners when looking at a very detailed model (the one shown in the press release actually :).
You are correct.
Whats up with that mozilla?
It is a good idea. Mozilla is a very large codebase with a reletively small number of developers. Therefore they don't have the fast turn around time for fixing critcal bugs that other projects do. They are already fixing bugs as fast as they can get around to it. If you wan't this process to go faster join the development team.
Security through obscurity is not completely worthless - it does one thing and that is to buy you more time, and that is all this is being used for. If you have not been part of the mozilla development team, then you do not know the codebase well enough to fix critcal bugs in a manner that doesn't potentially introduce more problems. And if you have been a part of the development team then you have access to the critical bug reports, so you can fix them.
It is always better to fix bugs rather than conceal them, but publishing the details of bugs that you haven't fixed is foolish. (This is opposed to just publishing the existence of vulnerabilities so your users are aware of potential problems).
Security is a matter of tradeoffs and to make good decisions, it is not enough to say "this is good", "this is bad" - You have to understand why practices are good or bad so you can understand when they should be used and to what extent they can be trusted, and do the best with the resources you have.
Side note / Prerefutation of slashbots:
No I am not defending mozilla because it is OSS, nor would I condem a comercial company for doing the same. I would condem them for claiming that their product is secure while hiding the existence of critcal bugs. Some OSS projects are good about bugs others are not. For example I am absolutely convinced that Apache is inherently more secure than MS IIS. I cannot say the same about Sendmail vs MS Exchange. Mozilla I am not sure about. They do some things better, like safer default settings. But like IE they also allow just about everything in the browser to be scripted, and I have little confidence that there aren't many critcal security bugs in all that code.
That won't help. The modifiers are not keyboard shortcuts, and cannot be changed that way. I haven't been able to find a way to do so, and while I am definately agree with the gnome philosophy of minimizing configurability to what is useful,
that really is a necisarry configuration if you want nice integrate into different environments.
I will third that comment, but also add that the slashdot team updates the slashcode on the live system on a weekly basis, so just because they haven't been fixed it yet doesn't necisarrily mean that they are not listening.
Their techniques may work, but that site screams snake oil scam artist. I these people really wanted to help people like they say they do they should be publishing in journals and making the information available to the community. The way they are currently handling themselves they are certain to be dismissed as frauds because for all we know they are.
My reservation with that is that if you are going to give them the responibilities of being an adult at age 13 you really ought to give them the freedoms as well. If you don't that is just plain contrary to liberty, and unfair.
If you do, then they are no longer subject to their parents at 13, and I'm sorry but kids that age need parents to watch over them as much or more than any time in thier life. Preferably it will more hands off, but parents need the ability to step in when the situation merits it. Doing what you say would basically be giving state blessing for parents to stop taking responsibility for thier kids actions at 13 and would be a very bad policy in my opinion.
They all began working most diligently
creating programs which behaved intellegently
But these programs were lacking still.
For when prompted they would only say or .
Yet a researcher in the great white north continued on
Feeding song after song into to his very own atomiton
Then at last one tiresome night in 84
he played a track he never played before.
And queried the program: To which it responded /wanting to work that last line into a joke for years
But the biggest problem with the three laws isn't that they are incomplete for determining the best course of action for a robot, which is what Asimov explored, but the fact that they are currently (and possibly inherently) impossible to implement.
How the heck is a robot supposed to accurately judge that whether a random unique action in a unique situation will cause harm to a human or himself? Humans can't even do this. If we were to create an artificial intellegence that was fully capable of making these decisions, would we even be able to put limits on what it decides?
Regardless of the answer to that philisophical question, we will have the technology to produce usefull robots long before we have the technology to produce 3-Law abiding robots so we need to come up with practical ways of making them as safe as possible, within their limited capabilites.
Oh yeah, they are doing some kick-ass things that we will see in the next release, absolutely. Just saying that we can't give them credit for the advances in this current release. Then again, the work is mostly being done by the same people, just different organizers.
No I'm not against any graphical configuration tools or this and that. I'm just against breaking the rule of changing the default UNIX tradition of configuration files. Any graphical tool should be like Webmin, which leaves the structure as it is.
This is what brought me back to slackware. I started with RH 4 but could never get it to work with my hardware. So I tried slackware and really enjoyed it. I learned so much about how Unix works on that slackware version. Anyway since then I have tried a different distro each time I am ready to do a complete OS upgrade. Here is what I learned.
As far as packaging goes rpm sucks unless you verify or build your own because the majority of 3rd party package builders do it wrong. At that point it's just as easy to go with slackware or gentoo. Apt-get seems really nice. Unfortuneatly, I didn't get much time with Debian (one week to install, then two weeks later my harddrive dies).
As far as configuration goes, those GUI tools are a pain. I tried 4 releases of RedHat and got to learn 4 ways of setting up PPP, and each of which seemed to get progressively worse. And of course once you use the GUI tools, it creates it's own config files from which the unix ones are generated. So after the easy way fails if you want to do it the manual way, you first have to figure out how to disable the distro provided tools, which is not always easy. The *drake tools are the flakiest things I have ever seen seen. They basically just issue some script commands and don't do any error handling. If something goes wrong, the window just disappears and you are left wondering if it worked or if not why it didn't work, and what state you system is in. Totally lame. Yast is the nicest of the bunch, but again you really need to decide to let Yast do everything, or do everything the manual way, because otherwise you will tromple all over each other.
As far as I am concerned, you can take your GUI configuration tools and keep them. Slackware may not be the easiest distro, but it is by far the least complicated. Even better, all the time I spend getting things to work on slackware, I am actually learning about how Linux works rather than figuring how to get around some broken config tool. That is the first thing that struck me when I started using slackware again. With the other distros I had gotten frustrated with all the maintainance I was doing that was all related to stuff I would never use again - fixing dependency errors, unbreaking harddrake - and this ended up driving me to Mac OS X for my main computer. With slackware I don't have to think about those kinds of problems, and I actually enjoy the problem-solving and discovering that I do have to deal with. It reminds me why I originally became so absorbed with linux in the first place.
To expand on what a couple of people have said, slackware has had a default XF86Config file for quite some time (always?) However in the past it defalted to using the Framebuffer X driver, and it 10 it defaults to using the Vesa driver. In both cases you need to change the XF86Config (now xorg.conf) to use a specific driver if you want decent performance.
This change is not (directly at least) related to the change to Xorg - they could have done it with XFree86 as well. Also while I agree that Xorg is the way to go, alot of the technical praise they are getting is misdirected. There is really almost signifcantly different between it and XFree86 4.4, and most of the improvements that people see in Xorg are really improvements in XFree 86 since the 4.3 series.
That depends. Like Paul said (paraphased), to die is gain because he gets to go to heaven, but to live is also to gain because he can save more people, so he is divided on which is best. This is pretty much the same thing - if the end times come soon then it is good for the saved but not the unsaved. Who is more important? Also, depending on whether you belive pre-trib or post-trib, and concidering how difficult the end times will supposedly be, there is good reason for a christian to want to die before the tribulations come.
Then again for those fundamentalists who would like to see everyone go to hell, but think that they will be whisked up into heaven in time for tea, then yeah, they should probably be cheering this stuff on.
Yeah, although to some extent I can see where they are comming from. I was our local ACM chapter president and would work with Microsoft rep (a real programmer), as Microsoft is nice about supporting local competitions (providing prizes and what not). So of course he was always trying to talk me into intervewing with Microsoft. During one of these conversations I brought up the issue of a monopoply hurting the market as a whole, and he admitted that he had thought about that quite a bit, and to some extent agreed that it would be better if there was no monopoly. But he wanted to write software that would matter - that would have some effect on the world (presumably positive) and there was no better place to do this than Microsoft.
:).
And he's right. I mean look at Apple they kick so much ass, and yet only 4% of computer users even notice what they do. Linux is even less outside of the server room. Microsoft is where it's at right now, and hopefully that will change, but it is going to be a long time. I'll admit that that matters to me. I get some satisfaction just from knowing I wrote good software, but I do want my work to matter. I don't want it to sit in a closet somewhere.
I didn't go to work for Microsoft, but I can understand why someone would. Now I have a job writing support software. I work with the people who use my software everyday and I get to see first-hand how it makes thier jobs easier (and sometimes not
Exactly. And it is not just Microsoft - it is a general weakness in the shrink-wrapped software business model. In that model you depend on sales of the previous version to fund additional features for the next version which in turn drives sales, and the cycle continues. The problem is that at some point your product becomes mature, and you have already implemented 90% of the features that 90% of the people want. Now there is still alot of potential functionality to be added but, each feature will appeal to only a small audience. Therefore even though you may have done as much work between versions 5 and 6 as you did between 2 and 3, you have deminishing returns on the number of purchases. Lastly, ever since the dot-com boom ended the number of first-time purchases (as opposed to upgrades) has been going down dramatically as well, so you are much more dependent on upgrades sales, which we just determined will also go down with time.
So basically the shrinkwrapped software business model sucks for mature software. Unless you can keep improving the software in a way that appeals to a large number of people, you will not be able to generate enough money from sales to continue development at your current pace. Then your product will stagnate, and newcommers who focus on different niche features that you don't have will eat away at your market share.
Once your software becomes mature, you really need to move away from the shrinkwrapped business model to some type of service business model. Interestly enough, OSS kicks ass in just about any service business model. If you are being paid for the act of writing and deploying software, rather then selling it as a product, it doesn't matter if you control the software or not. It just matters that you have the experience and talent to improve an existing piece of software (ie helps alot if you wrote the software to begin with).
OSS has the opposite problem - it is easy to get paid to improve mature products, but getting a piece of software to maturity is harder (financially).
You are right - it doesn't really make any difference for hardware devices if you have to licence the codec or not - peanuts in the overall cost. Same with proprietary software - the licencing costs won't add much to the cost of the product, so no big deal.
But open source software is a problem. It is illegal to distribute an mp3 encoder without paying the licence fees, which makes it impossible to have a legal open source mp3 encoder (or DVD players, or anything else that is patented) because to be open the licence must allow anyone to redistribute it without restriction. The way that LAME gets around that issue is that "LAME Ain't an MP3 Encoder" - it is a reference implementation for educational use only. This doesn't really solve the problem, it just makes the users criminals instead of the developers and distributers.
So we don't like patented formats for that reason - because it is illegal for us to use those formats on our open source systems, and if we are restricted to use open formats on our computer, it would be really nice to use them on our handheld music players as well, which fortunately is a very viable option, contrary to what the whiners here say, and thanks to some cool hardware companies.
Hehe. Just be happy you weren't standing at the top of the stairs, like I was. Protector robot my ass.