But, they want a computer user to be able to send to a cell-phone through IM. Or, for a cell-phone user to send to a Yahoo!, MSN or ICQ user. Of course, if AOL got off their butts, you'd be able to send to an AIM user as well, but for right now, they'll be busted... oh well, get a real ISP/IMP.
RedHat [sells] subscription based 'support services' which [comes] with, by the way, free software upgrades during the time of the subscription. I mean, is there a whole lot of differences? I don't think the question is will the business community take to this new idea, it's how have they accepted it so far?
Well, let's see. MS wants to charge you a fee to get a 3-year license on the software, after which time you are not allowed to use it (it will probably stop working).
Red Hat charges nothing for download of their OS, nothing for access to the updates and their OS keeps working as long as you don't use hardware that's newer than the OS, and even then many things will work.
What you are refering to is Red Hat's RHN service which offers a convinient way to update your software through a GUI, an/or automatically update the software. They charge for this service, but they do not restrict access to the software or the raw updates at all. They also are still distributing updates for Red Hat 5.x which is 3 years old at this point.
If MS stops supporting Win98, you're up the creek. No source, no updates, no nothin'.
If you read/. and manage Windows systems, I urge you to send mail to Microsoft on this issue. They need feedback, lots of it.
Something of this form would do:
To Whom It May Concern,
It has come to my attention that Microsoft is planning to add an obsolescence feature to their software for enterprise customers. As a consumer of your software, and of other software which will almost certainly follow Microsoft's lead, I urge you to abandon this tactic as soon as possible.
Please, understand that many companies such as ours will need to re-evaluate our software choices should this come to pass. We use and appreciate Microsoft's software, but cannot tollerate this level of demand from any vendor.
Thank you for your time.
That should get the right mix of reaction without setting off the loony alerts.
If your company doesn't have a better address to send this to, try:
Microsoft Corporation
One Microsoft Way
Redmond, WA 98052-6399
USA
In a discussion of Napster, gnutella comes up. Seems on-topic. I reply clarifying some points, and adding a plee for help in a particular area of the stated topic (e.g. gnutella) and that's off-topic? Someone must have been browsing flat/newest....
I have enough karma to burn on good causes like anti-violence, so I guess I'll keep doing so. I'll be polite and as on-topic as I can, but I think that this is an issue that deserves to be heard.
When you search gnutella for mp3s, you get very little (read no) porn.
When you search for video, you get a very interesting thing: you get the ratios in which people ACTUALLY HAVE AN INTEREST in video. These ratios will change as the gnutella user-base becomes more diverse, of course.
One thing that gets me though, is the violent videos. There are clips of all kinds of atrocities from rape to murder. My cache grabs many of these, and sometimes I can't tell what they are without looking at them.
Ye gods, what horror. I get a knot in my stomach, and want to do something to stop what I'm watching, but all I can do is hit the stop button, I can't help the person who is, almost certainly in most of these cases, already dead.
If anyone wants to put together a foundation for paying people to research stopping such things (or an existing anti-violence group wants to start paying attention to gnutella), I will gladly pay into it.
At the very least (and I grant that this is hardly even a band-aid), people, please label the crap out there. I'd love to configure my cache to ignore files with certain keywords, but nothing is consistant.
If you just sit on your couch like a lump you can never achieve the acclaim of your peers or do something blindingly stupid enough to win the Darwin Award. The line between the two is success.
If this guy does what he says he will do, he will be famous. If he screws up... well, let's just say that he'll not be worrying the rest of us with the proliferation of his DNA;)
I wish him the best, and hope he does it. Amature spaceflight might just be the only way to wake up corporations to the value of private space exploitation.
Even a perfectly copied DVD video would contain watermarks that would prevent a DVD recorder with a watermark detection chip from playing the bootleg copy.
Now, I'm a computer scientist, so perhaps my brain has been twisted by logic, but doesn't this mean that the original disc would not play? Or, are they suggesting that there's something on the DVD that a "perfectly copied DVD" would not have? If so, it isn't very perfect is it?
Sounds like yet another scheme that can only prevent hobbiests from copying discs. Real pirates can always just copy the media bit-for-bit, flaw-for-flaw.
First off, since everyone seems ready to discuss my age, I'm 31.
Second, I've used Napster maybe twice. I'm a Gnutella man, because sometimes you just want video to go with your audio;-)
Third, I'm a buyer of CDs, movies, etc. in what I suspect is average perportion and magnitude.
Gnutella, Napster and the rest just extend the ways that we have to share our music with others. Before these services, I was ripping my music and putting it on my web site. Not because I wanted people to "steal" it, but because I wanted to promote what I thought was good music. I wanted to share with others. I would go to friends and as: have you ever heard The Raven by The Alan Parsons Project? No? Here, let me give you a URL.
I've never offered non-original content up over Gnutella or Napster (except insofar as my gnut client will offer up a "cache"), but that's only because I've been too busy, and/or too lazy.
No, there is no price-point at which "piracy" will stop. If the RIAA came to my house with 100,000 CDs and gave them to me for free, I'd say thank you and then go back to using Gnutella. Why?
Because, I don't *want* to pirate music. I want to listen to things I've never heard before. I want to experience the convinience of thinking of a song, typing in a keyword or two, waiting a few moments and then playing it.
The RIAA and it's members DO NOT SELL THIS. They sell CDs. CDs are cool, and from time to time I buy them (usually based on what I hear through downloaded music).
Here's the key: there's a service that I want, and no one provides it. There is no price point at which CDs become that service.
The RIAA members do not want to provide this service because a) it's difficult to guarantee returns and b) it removes the push-model from the equation. What does this mean? Right now, when Sony wants to sell a new artist, they go through a very well defined process to make it popular. If people were just downloading whatever they wanted based on either random or non-RIAA-member-supplied criteria, they would not be able to control "the next big thing".
This would be tragic for Sony, etc. They would have to provide HIGH QUALITY MUSIC and support their artists much more. If you've ever worked in an industry where channels are more important than content (e.g. the Internet), you will understand why the RIAA is so scared that they would make enemies like their customers, Princeton, the EFF and Stanford....
Translation: Someone said Gnome could do something. I must prove that KDE can do it too. I must defend KDE from the assertion that Gnome is useful. I must explain to others that Gnome is not the one true way.
Yep, that's EXACTLY what I was complaining about. I think KDE and Gnome are both cool, but the zealots on both sides that seem hell-bent on denying the power and usefulness of the other side are starting to make me long for X10 with uwm....
Two days? It took about 2 hours on my 300Mhz x86. What are we calling moderate hardware?
Also, I find it interesting that Ximian is considered some kind of corporate raider. These guys are free software hackers who decided to make it their day-job. I work just down the street from them, and have stopped in their office before. Let me assure you that they are not the evil capitalist pig-dogs trying to take over Gnome....
Before someone goes off the deep end trying to "re-package Gnome" without the offensive pixmaps of doom, I'd rather they spend time hacking on some of the code. There are features that need to be completed before Gnome will represent the definitive MS-killer (though it's most of the way there, IMHO).
Once you download and install Red Carpet, you have full verification of binaries all the way through the process. The go-gnome installer is a bootstrap process. You can download source, compile and begin the install if you want, but this is not grandmother compliant....
1. Factual errors
2. KDE ranting where it doesn't belong.
Miguel formed Ximian (Helix back then) because he thought that it was the right thing to do to keep Gnome growing, and get commercial acceptance. Given HP and Sun's moves, I agree.
Gnome is still just as free as Mozilla (even though, like Mozilla many of the developers work for a commercial entity).
If you don't like where Gnome is going, feel free to fork it. I think you'll have a little trouble just keeping up with the updates, but hey, that shouldn't stop you from trying! Then again, you could contribute....
This was yet another "but, if they're trying to make money it's not free, right?" articles that you see from time to time. It's always done by someone who a) just saw free software for the first time or b) has an axe to grind because they like another project more.
He likes KDE. Cool, let 'im. He don't need to piss on our playground because he's got a pet desktop.
One would almost think that these folks had an agenda;-)
<RANT>
First, the "someone, somewhere" comment about paying for Gnome gets a two-word answer: "Sun, HP".
Next, on KDE. I don't give a rat's left kidney about KDE, and why the heck does every 2-bit reporter with a browser have to compare Gnome and KDE?
Don't get me wrong, I wish the KDE folks a lot of luck, just not my cup of tea. We're almost mature enough to stop mentioning Linux every time we talk about BSD (and visa versa), hopefuly we can drop the Gnome/KDE thing soon.
Now, as for "KDE has actual applications". See my coments about about KDE, but for Gnome, we have:
1. AbiWord (word processing)
2. Gnumeric (spreadsheet)
3. Evolution (groupware; under development)
4. Gnomecal (caldendar)
5. Gnucash (finance)
6. Glade (GUI development)
7. Dia (vector layout)
8. GnomeICU (instant messaging)
9. LOTS more that I don't have time to type.
On the Gtk front (non-gnome, just using the same toolkit) there's Gimp (photo-editing), Mozilla (web browsing, HTML editing etc), and again a good many others.
Can we drop the "there aren't any applications" thing.
1. It is the medium through which discussion of programming techniques can be explored.
2. Code is math+logic, and expression of mathematical techniques and logical progressions is a long-defended practice.
3. Beauty doesn't really enter into it. When computer programmers communicate ideas from humor to philosophy to religeon to politics through source code, it is clearly an expressive medium.
In short: programming languages, terse though they may be, are the medium of modern technical exchange and thought.
In example: EMACS is a program which has existed since the mid '80s. The code contains religeous expression (the Towers of Hanoi, a classic application of mathematical techniques to religeous expression), political protest (Spook mode for confounding the NSA), humor (Yow mode). All of this, and EMACS is primarily designed to edit programs and text!
My answer was 1, which seemed the only reasonable answer. The problems that I had with the answer are:
1. No where in the problem statement is there any mention of it taking 1 day for someone to come to a conclusion (thus "end of the third day" is an ignorable unit of time).
2. Why are we assuming that the people in the village are competent logicians?
Hell, the way it's worded, it's perfectly reasonable to answer: all of them. Every last villager is a sinner, and it took them three days to pack their bags.
I hate stupid logic problems. Give me a simple math problem any time.
When one of the alternatives does catch up, like Gnutella might do, the music industry will just shift its attention to the next.
Too late, they already have. And, they pretty quickly realized just how useless that was. First, everyone they've busted has larger-than-average *purchased* CD collections (do you think any of them will ever buy another CD?) Second, they didn't put a dent in Gnutella, since it's totally de-centralized.
Saying that filters are no use because you can use a proxy is saying that law against selling licquor to minors is no use because you can have a fake ID.
There's a huge difference. When someone "discovers" a good place to buy alcohol, they don't send the address to 50 of their closest friends, who then download it in the span of 5 minutes before the authorities can notice what's going on. Information moves too fast to be tracked by humans, and it (so far) too complex to be unwoven and effectively censored by machines.
I was going to moderate this dicussion, but then I saw that this had no replies, and it REALLY needs one.
The "far more powerful laws" of information are observed beautifully in the Napster situation, in that there are now un-policed alternate implementations. Also, there's Gnutella, Freenet and a host of smaller projects. Each have a different spin (Gnutella is for ANY type of file, Freenet is specifically aimed at maintainin the most "interesting" content forever in order to promote free speach).
However, a much better example is software. Companies with billions of dollars in the bank have been working for decades now to prevent the advent of commercial-quality free software. What's the result? Apache has 60% of the Web server market. Linux is one of the most reliable Internet server platforms on the planet. There are implementations of just about every type of software ever written that are free. Why? Is this because one man has a personal mission to create software? Nope. It's because software is basically information, and information cannot be controled. One of the most interesting aspects of human society is our desire to share information (secret keeping is actually very hard).
Commercial enterprises like Napster can be squashed, but the "laws of information freedom" (which is a bastardization of the original poster's term, not mine) do not constrain themselves to a single company.
Only 5 companies providing high-speed internet access? Ok, I could cope with that. It's not zero competion. As long as they did it right.
But, it'll be AT&T (no dream) and the mini-bells (may all the little gods protect us).
A friend of mine is a Verizon (NET+NYNEX+Bell Atlantic) DSL customer. He has huge flakiness problems with the line, and has been dealing with their customer support drones for months. They even transfered his ticket to a 2nd level support engineer whose response was "oh, check to make sure you haven't done something stupid with your regular phone line. I'll go ahead and close the ticket."
No, there must be stronger competition than there is today. I don't know what the right solution is. Some advocate nationalizing the wires and leasing it back to multiple local providers, but I don't think that would make things any better, just more beaurecratic. Perhaps fedral grants to companies that want to lay their own wire? Or, even a loosening of the wireless provisions....
I fear the future of home Internet access. It's only going to get more flaky and more restrictive. Sigh.
copy on write and signals have nothing to do with each other
In mach, they most certainly do. There are a number of resources that are difficult to copy-on-write because they are too high-level to be managed by the VM. With mach's copy-on-write semantics, you can actually manage the copy in a user-space callback.
message passing ipc is more expensive than just invoking a user space signal handler
Expensive yes, but much richer. Sending a SIGFIZZ to the process is all well and good, but in most cases, the process then needs to go dip into some kernel datastructure or make a system call to get more information. With message passing, you get it all in one pass, and you can define richer message classes than the traditional POSIXish signal mechanism can deal with.
the whole point of a microkernel is to have a different fault domain for each subsystem
This is actually a kind of funny statement. Since microkernels are being used most often because they provide a conviniently small and flexible abstraction layer for monolithic operating systems, I think you have to revise your thinking here. Perhaps when microkernels were envisioned this was the idea, but throughout the late '90s and early '00s, we've certainly seen very little of this sort of microkernel use.
This use of microkernels is not invalid, nor is it rare. Would all monolithic OSes benefit from a microkernel-as-abstraction-layer? Yes I think so in the ideal, however many OSes (e.g. Linux) have seen tremendous benefit from the monolithic model that they started with, and to re-impliment on top of, e.g., mach, would be quite painful. I'm not convinced that the benefits would outweigh the cost.
NT has no relation (other than conceptual) to Mach. MacOS X, NeXT OS and OSF/1 (from which True64 descends) are all Mach with some relatively monolithic layers on top.
It is important to note that those three systems are monolithic systems running on top of a Mach microkernel. Mach has not, in any of those cases, been modified to the extent that it no longer functions as a true microkernel.
Also, the original post made no such distinction. It made a blanket statement that Mach had never been used in a production OS. This is untrue, and there is a fair deal of advantage in having the microkernel abstraction available, even if you run a monolithic kernel layer on top of it.
For example, take the fact that OSF/1 based systems use Mach message passing and VM, so they can handle such fun concepts as reserve-on-write memory allocation with message-callbacks on failure. In UNIX terms, that means that "malloc(1024*1024*1024)" does not reduce the amount of VM available... which is both scary and useful at the same time;-)
The proliferation of copy-on-write semantics in Mach-derived systems is also very useful, and again benefits greatly from the way mach does message-passing instead just having signals to convey information from the kernel to user space.
Linux (or FreeBSD or MacOS X, etc) is UNIX is 1960s technology in the same way that 2000 is NT is VMS is 1960s technology.
Neither one is a very fair statement, since development has progressed dramatically (including at least one major re-implementation along the way) in both cases.
Slashdot Mirror
But, they want a computer user to be able to send to a cell-phone through IM. Or, for a cell-phone user to send to a Yahoo!, MSN or ICQ user. Of course, if AOL got off their butts, you'd be able to send to an AIM user as well, but for right now, they'll be busted... oh well, get a real ISP/IMP.
You can use Jabber to talk to MSN, Yahoo!, ICQ and (when they're not being total dicks) AIM.
So, now you don't have to run a seperate client for each. At worst, you have to run Gaim and a Jabber client (e.g. gabber).
RedHat [sells] subscription based 'support services' which [comes] with, by the way, free software upgrades during the time of the subscription. I mean, is there a whole lot of differences? I don't think the question is will the business community take to this new idea, it's how have they accepted it so far?
Well, let's see. MS wants to charge you a fee to get a 3-year license on the software, after which time you are not allowed to use it (it will probably stop working).
Red Hat charges nothing for download of their OS, nothing for access to the updates and their OS keeps working as long as you don't use hardware that's newer than the OS, and even then many things will work.
What you are refering to is Red Hat's RHN service which offers a convinient way to update your software through a GUI, an/or automatically update the software. They charge for this service, but they do not restrict access to the software or the raw updates at all. They also are still distributing updates for Red Hat 5.x which is 3 years old at this point.
If MS stops supporting Win98, you're up the creek. No source, no updates, no nothin'.
Something of this form would do:
That should get the right mix of reaction without setting off the loony alerts.
If your company doesn't have a better address to send this to, try:
In a discussion of Napster, gnutella comes up. Seems on-topic. I reply clarifying some points, and adding a plee for help in a particular area of the stated topic (e.g. gnutella) and that's off-topic? Someone must have been browsing flat/newest....
I have enough karma to burn on good causes like anti-violence, so I guess I'll keep doing so. I'll be polite and as on-topic as I can, but I think that this is an issue that deserves to be heard.
When you search gnutella for mp3s, you get very little (read no) porn.
When you search for video, you get a very interesting thing: you get the ratios in which people ACTUALLY HAVE AN INTEREST in video. These ratios will change as the gnutella user-base becomes more diverse, of course.
One thing that gets me though, is the violent videos. There are clips of all kinds of atrocities from rape to murder. My cache grabs many of these, and sometimes I can't tell what they are without looking at them.
Ye gods, what horror. I get a knot in my stomach, and want to do something to stop what I'm watching, but all I can do is hit the stop button, I can't help the person who is, almost certainly in most of these cases, already dead.
If anyone wants to put together a foundation for paying people to research stopping such things (or an existing anti-violence group wants to start paying attention to gnutella), I will gladly pay into it.
At the very least (and I grant that this is hardly even a band-aid), people, please label the crap out there. I'd love to configure my cache to ignore files with certain keywords, but nothing is consistant.
Exactly what the Darwin Award is all about.
;)
If you just sit on your couch like a lump you can never achieve the acclaim of your peers or do something blindingly stupid enough to win the Darwin Award. The line between the two is success.
If this guy does what he says he will do, he will be famous. If he screws up... well, let's just say that he'll not be worrying the rest of us with the proliferation of his DNA
I wish him the best, and hope he does it. Amature spaceflight might just be the only way to wake up corporations to the value of private space exploitation.
This is, as you point out, a bad situation. History shows us that prohibiting access to alcohol simply makes the problem worse.
Drugs that cause problems cause more problems when they are illegal. Sad but true.
Sounds like yet another scheme that can only prevent hobbiests from copying discs. Real pirates can always just copy the media bit-for-bit, flaw-for-flaw.
First off, since everyone seems ready to discuss my age, I'm 31.
;-)
Second, I've used Napster maybe twice. I'm a Gnutella man, because sometimes you just want video to go with your audio
Third, I'm a buyer of CDs, movies, etc. in what I suspect is average perportion and magnitude.
Gnutella, Napster and the rest just extend the ways that we have to share our music with others. Before these services, I was ripping my music and putting it on my web site. Not because I wanted people to "steal" it, but because I wanted to promote what I thought was good music. I wanted to share with others. I would go to friends and as: have you ever heard The Raven by The Alan Parsons Project? No? Here, let me give you a URL.
I've never offered non-original content up over Gnutella or Napster (except insofar as my gnut client will offer up a "cache"), but that's only because I've been too busy, and/or too lazy.
No, there is no price-point at which "piracy" will stop. If the RIAA came to my house with 100,000 CDs and gave them to me for free, I'd say thank you and then go back to using Gnutella. Why?
Because, I don't *want* to pirate music. I want to listen to things I've never heard before. I want to experience the convinience of thinking of a song, typing in a keyword or two, waiting a few moments and then playing it.
The RIAA and it's members DO NOT SELL THIS. They sell CDs. CDs are cool, and from time to time I buy them (usually based on what I hear through downloaded music).
Here's the key: there's a service that I want, and no one provides it. There is no price point at which CDs become that service.
The RIAA members do not want to provide this service because a) it's difficult to guarantee returns and b) it removes the push-model from the equation. What does this mean? Right now, when Sony wants to sell a new artist, they go through a very well defined process to make it popular. If people were just downloading whatever they wanted based on either random or non-RIAA-member-supplied criteria, they would not be able to control "the next big thing".
This would be tragic for Sony, etc. They would have to provide HIGH QUALITY MUSIC and support their artists much more. If you've ever worked in an industry where channels are more important than content (e.g. the Internet), you will understand why the RIAA is so scared that they would make enemies like their customers, Princeton, the EFF and Stanford....
Translation: Someone said Gnome could do something. I must prove that KDE can do it too. I must defend KDE from the assertion that Gnome is useful. I must explain to others that Gnome is not the one true way.
Yep, that's EXACTLY what I was complaining about. I think KDE and Gnome are both cool, but the zealots on both sides that seem hell-bent on denying the power and usefulness of the other side are starting to make me long for X10 with uwm....
Two days? It took about 2 hours on my 300Mhz x86. What are we calling moderate hardware?
Also, I find it interesting that Ximian is considered some kind of corporate raider. These guys are free software hackers who decided to make it their day-job. I work just down the street from them, and have stopped in their office before. Let me assure you that they are not the evil capitalist pig-dogs trying to take over Gnome....
Before someone goes off the deep end trying to "re-package Gnome" without the offensive pixmaps of doom, I'd rather they spend time hacking on some of the code. There are features that need to be completed before Gnome will represent the definitive MS-killer (though it's most of the way there, IMHO).
Learn a little bit more before you post.
Once you download and install Red Carpet, you have full verification of binaries all the way through the process. The go-gnome installer is a bootstrap process. You can download source, compile and begin the install if you want, but this is not grandmother compliant....
1. Factual errors
2. KDE ranting where it doesn't belong.
Miguel formed Ximian (Helix back then) because he thought that it was the right thing to do to keep Gnome growing, and get commercial acceptance. Given HP and Sun's moves, I agree.
Gnome is still just as free as Mozilla (even though, like Mozilla many of the developers work for a commercial entity).
If you don't like where Gnome is going, feel free to fork it. I think you'll have a little trouble just keeping up with the updates, but hey, that shouldn't stop you from trying! Then again, you could contribute....
This was yet another "but, if they're trying to make money it's not free, right?" articles that you see from time to time. It's always done by someone who a) just saw free software for the first time or b) has an axe to grind because they like another project more.
He likes KDE. Cool, let 'im. He don't need to piss on our playground because he's got a pet desktop.
One would almost think that these folks had an agenda ;-)
<RANT>
First, the "someone, somewhere" comment about paying for Gnome gets a two-word answer: "Sun, HP".
Next, on KDE. I don't give a rat's left kidney about KDE, and why the heck does every 2-bit reporter with a browser have to compare Gnome and KDE?
Don't get me wrong, I wish the KDE folks a lot of luck, just not my cup of tea. We're almost mature enough to stop mentioning Linux every time we talk about BSD (and visa versa), hopefuly we can drop the Gnome/KDE thing soon.
Now, as for "KDE has actual applications". See my coments about about KDE, but for Gnome, we have:
1. AbiWord (word processing)
2. Gnumeric (spreadsheet)
3. Evolution (groupware; under development)
4. Gnomecal (caldendar)
5. Gnucash (finance)
6. Glade (GUI development)
7. Dia (vector layout)
8. GnomeICU (instant messaging)
9. LOTS more that I don't have time to type.
On the Gtk front (non-gnome, just using the same toolkit) there's Gimp (photo-editing), Mozilla (web browsing, HTML editing etc), and again a good many others.
Can we drop the "there aren't any applications" thing.
<RANT>
lynx -source http://go-gnome.com/ | sh
Done.
Here are the reasons that code is expression:
1. It is the medium through which discussion of programming techniques can be explored.
2. Code is math+logic, and expression of mathematical techniques and logical progressions is a long-defended practice.
3. Beauty doesn't really enter into it. When computer programmers communicate ideas from humor to philosophy to religeon to politics through source code, it is clearly an expressive medium.
In short: programming languages, terse though they may be, are the medium of modern technical exchange and thought.
In example: EMACS is a program which has existed since the mid '80s. The code contains religeous expression (the Towers of Hanoi, a classic application of mathematical techniques to religeous expression), political protest (Spook mode for confounding the NSA), humor (Yow mode). All of this, and EMACS is primarily designed to edit programs and text!
My answer was 1, which seemed the only reasonable answer. The problems that I had with the answer are:
1. No where in the problem statement is there any mention of it taking 1 day for someone to come to a conclusion (thus "end of the third day" is an ignorable unit of time).
2. Why are we assuming that the people in the village are competent logicians?
Hell, the way it's worded, it's perfectly reasonable to answer: all of them. Every last villager is a sinner, and it took them three days to pack their bags.
I hate stupid logic problems. Give me a simple math problem any time.
When one of the alternatives does catch up, like Gnutella might do, the music industry will just shift its attention to the next.
Too late, they already have. And, they pretty quickly realized just how useless that was. First, everyone they've busted has larger-than-average *purchased* CD collections (do you think any of them will ever buy another CD?) Second, they didn't put a dent in Gnutella, since it's totally de-centralized.
Saying that filters are no use because you can use a proxy is saying that law against selling licquor to minors is no use because you can have a fake ID.
There's a huge difference. When someone "discovers" a good place to buy alcohol, they don't send the address to 50 of their closest friends, who then download it in the span of 5 minutes before the authorities can notice what's going on. Information moves too fast to be tracked by humans, and it (so far) too complex to be unwoven and effectively censored by machines.
I was going to moderate this dicussion, but then I saw that this had no replies, and it REALLY needs one.
The "far more powerful laws" of information are observed beautifully in the Napster situation, in that there are now un-policed alternate implementations. Also, there's Gnutella, Freenet and a host of smaller projects. Each have a different spin (Gnutella is for ANY type of file, Freenet is specifically aimed at maintainin the most "interesting" content forever in order to promote free speach).
However, a much better example is software. Companies with billions of dollars in the bank have been working for decades now to prevent the advent of commercial-quality free software. What's the result? Apache has 60% of the Web server market. Linux is one of the most reliable Internet server platforms on the planet. There are implementations of just about every type of software ever written that are free. Why? Is this because one man has a personal mission to create software? Nope. It's because software is basically information, and information cannot be controled. One of the most interesting aspects of human society is our desire to share information (secret keeping is actually very hard).
Commercial enterprises like Napster can be squashed, but the "laws of information freedom" (which is a bastardization of the original poster's term, not mine) do not constrain themselves to a single company.
Only 5 companies providing high-speed internet access? Ok, I could cope with that. It's not zero competion. As long as they did it right.
But, it'll be AT&T (no dream) and the mini-bells (may all the little gods protect us).
A friend of mine is a Verizon (NET+NYNEX+Bell Atlantic) DSL customer. He has huge flakiness problems with the line, and has been dealing with their customer support drones for months. They even transfered his ticket to a 2nd level support engineer whose response was "oh, check to make sure you haven't done something stupid with your regular phone line. I'll go ahead and close the ticket."
No, there must be stronger competition than there is today. I don't know what the right solution is. Some advocate nationalizing the wires and leasing it back to multiple local providers, but I don't think that would make things any better, just more beaurecratic. Perhaps fedral grants to companies that want to lay their own wire? Or, even a loosening of the wireless provisions....
I fear the future of home Internet access. It's only going to get more flaky and more restrictive. Sigh.
copy on write and signals have nothing to do with each other
In mach, they most certainly do. There are a number of resources that are difficult to copy-on-write because they are too high-level to be managed by the VM. With mach's copy-on-write semantics, you can actually manage the copy in a user-space callback.
message passing ipc is more expensive than just invoking a user space signal handler
Expensive yes, but much richer. Sending a SIGFIZZ to the process is all well and good, but in most cases, the process then needs to go dip into some kernel datastructure or make a system call to get more information. With message passing, you get it all in one pass, and you can define richer message classes than the traditional POSIXish signal mechanism can deal with.
the whole point of a microkernel is to have a different fault domain for each subsystem
This is actually a kind of funny statement. Since microkernels are being used most often because they provide a conviniently small and flexible abstraction layer for monolithic operating systems, I think you have to revise your thinking here. Perhaps when microkernels were envisioned this was the idea, but throughout the late '90s and early '00s, we've certainly seen very little of this sort of microkernel use.
This use of microkernels is not invalid, nor is it rare. Would all monolithic OSes benefit from a microkernel-as-abstraction-layer? Yes I think so in the ideal, however many OSes (e.g. Linux) have seen tremendous benefit from the monolithic model that they started with, and to re-impliment on top of, e.g., mach, would be quite painful. I'm not convinced that the benefits would outweigh the cost.
NT has no relation (other than conceptual) to Mach. MacOS X, NeXT OS and OSF/1 (from which True64 descends) are all Mach with some relatively monolithic layers on top.
;-)
It is important to note that those three systems are monolithic systems running on top of a Mach microkernel. Mach has not, in any of those cases, been modified to the extent that it no longer functions as a true microkernel.
Also, the original post made no such distinction. It made a blanket statement that Mach had never been used in a production OS. This is untrue, and there is a fair deal of advantage in having the microkernel abstraction available, even if you run a monolithic kernel layer on top of it.
For example, take the fact that OSF/1 based systems use Mach message passing and VM, so they can handle such fun concepts as reserve-on-write memory allocation with message-callbacks on failure. In UNIX terms, that means that "malloc(1024*1024*1024)" does not reduce the amount of VM available... which is both scary and useful at the same time
The proliferation of copy-on-write semantics in Mach-derived systems is also very useful, and again benefits greatly from the way mach does message-passing instead just having signals to convey information from the kernel to user space.
Linux (or FreeBSD or MacOS X, etc) is UNIX is 1960s technology in the same way that 2000 is NT is VMS is 1960s technology.
;-)
Neither one is a very fair statement, since development has progressed dramatically (including at least one major re-implementation along the way) in both cases.
Oh well, trolls can't be right all the time