There is much to admire about the Debian project, but it's a mass of people and procedures. Fedora is better off if it can be faster, with a coherent design coming from a small number of architects. It also serves as testing ground for new Red Hat Enterprise features. Debian's 11-port setup is a real handicap to moving fast because many of those targets have really slow CPUs and get no testing from upstream.
Nothing prevents a Red Hat Enterprise customer from making a copy of the software and giving it to someone else. However, the contract you must sign before Red Hat will give you the binary distribution says that you pay for support for N machines. If you want to run it on N+1 machines, you owe Red Hat more money. It further says that Red Hat can audit you, just like a proprietary software company. The only way to get support from Red Hat is to sign the contract, which means that you either pay per machine, or you somehow get the code and do without support.
Red Hat actually makes the SRPMs for their Enterprise system available to all.
That's almost exactly what we said about egcs versus gcc when egcs started, to keep the FSF from flipping out. However, the result was that egcs ended up replacing GCC (what was originally planned as egcs 1.2 became gcc 2.95). This is good strategy for those who wish to avoid a fork: arrange that the fork can eventually become the main branch.
Whether xouvert will replace or take over Xfree86 depends on whether the majority of developers abandon the xfree86 ship and work on the xouvert branch.
For their first couple of years of operation, their database apparently had El Camino Real through Silicon Valley marked as a freeway. The driving directions would often tell you to stay on El Camino for about six miles, which considering there is an unsynchronized traffic light every three blocks, would get old fast. They appear to have fixed that problem some time ago, though.
Linus wrote: In other words, this should calm things down so that by the end of October we can look at the state of 2.6.0 without having a lot of noise from 'not strictly necessary' stuff."
That is, at the end of October he will "look at the state of 2.6.0". That's quite different from shipping it.
The good folks at Red Hat have come up with a cool way to avoid some of the problems of monoculture in GNU/Linux: position independent executables. Addresses of code segments can be randomized at load time by the dynamic linker. The result is that common techniques for writing buffer overflow exploits no longer work, because every executable on every server is different. You can no longer insert code into a buffer whose length is not checked and then override the return address to point to it, because you don't know what return address to use. Worms can't spread if this technique is used.
While this technique still doesn't stop people from exploiting cross-site scripting bugs, it's
progress.
RMS does not claim to have invented the concept of a multi-user timesharing system with security features; he just says that GNU and Linux software was designed that way from the beginning (taking ideas from existing systems). Since Microsoft started with the concept of one user and no security, then when they added multiple users and security they had to make tons of compromises to keep the legacy software running.
The result is that Windows XP has a very nice security model which is turned off by default, because if you turn it on, you can't use any apps that were originally designed for older Windows flavors.
No, you're wrong: the article says that the entire kernel was compared with the SCO kernel. What's confusing you is that SGI says the point was to look for anything that might implicate SGI, but in the process they compared the whole thing.
No, nothing I said was factually incorrect. Under the DMCA, the complainant sends a complaint, and if the defendant doesn't produce an affidavit (swearing under penalty of perjury that there is no copyright violation), the ISP must take down the site, or else the ISP becomes liable for copyright infringement if the complaint holds up. That's what a takedown notice is.
Microsoft is sending a demand from its lawyers to take down the site, but they aren't using a procedure that will cause the ISP to take action.
"Takedown notice" usually refers to a procedure defined in the DMCA that allows someone alleging copyright infringement to demand that an ISP take a site down. This is not what is happening in this case. Microsoft has no means of forcing Lindows to take the site down without a lengthy legal process.
I fear that the people who are locked into contracts are screwed, but the good news is that, if we can develop economic models that let musicians do better in the Free World than in RIAA-world, new musicians won't join, and established musicians won't renew contracts once their commitments are up. More and more musicians are going to start to realize that detaching yourself from your label and going independent can get you more money on 1/10 the sales, because you get to keep it.
But if anyone is an unsigned band: don't sign with an RIAA-connected label, no matter what they offer, because they're going to screw you and because they're going down.
But why would anyone think that Amazon could be trusted for e-commerce searches? If someone is selling a product for a lower price than Amazon is, do you really think that their search engine will point me there?
I think that Google could be beaten, but not by Amazon or Microsoft. The problem is that a search engine has to be perceived as neutral.
A little guy with much better technology could make headway.
Also, the likelihood of Google screwing up will increase greatly once they go public. Investors will demand more return, and management might eventually do something that blows the company reputation.
It's the problem with being a public corporation. The investors demand exponential growth, forever. This means that management feels pressured to keep doing new things, or try to squeeze blood from a stone. The problem is, they go beyond their domain of competence and soon hose everything.
Oregon is going to have to raise more money somehow. Already they've had to resort to closing schools three weeks early, and I suspect that the amount of dererred maintainance of critical infrastructure has reached dangerous levels.
Also, Oregon has relatively high property and income taxes, which any companies moving there will have to pay. Of course, there's the trick of living in Washington State (no state income tax) and shopping in Oregon, thus freeloading on the other citizens of two states.
Careful. There are a whole series of Creative Commons licenses. You need to read the terms in detail to figure out if you can use a Creative Commons-licensed work as part of a game.
Another problem is people applying the GPL to a sound file without understanding the implications.
If that sound file was produced from a sound synthesis program, shipping only the output would be like shipping only the binary of a GPLed program. The GPL says that you have to provide "source code" and defines the source as the preferred form for doing modifications to the work. So, if you made an image with the GIMP, the XCF file (with the image separated into layers) would be the "source".
No, the vulnerabilities are due to new code in 3.7; the Red Hat and Debian people who backported only the security fixes to older OpenSSH versions are safe. They are not old vulnerabilities that were discovered by an increase in code vetting.
You dropped a zero: it's $500,000 in taxes should you ship 500k units. And that's just for the MP3 royalties: if you do video, you've got to pay for that too.
If apt is used, conflicts are not necessarily a problem, provided that the conflicts are correctly described in the apt database. If you try to install a package that conflicts with some other package, you are given the option to proceed (and remove the conflicting package) or not, and with either choice your system stays consistent.
Slashdot Mirror
There is much to admire about the Debian project, but it's a mass of people and procedures. Fedora is better off if it can be faster, with a coherent design coming from a small number of architects. It also serves as testing ground for new Red Hat Enterprise features. Debian's 11-port setup is a real handicap to moving fast because many of those targets have really slow CPUs and get no testing from upstream.
Nothing prevents a Red Hat Enterprise customer from making a copy of the software and giving it to someone else. However, the contract you must sign before Red Hat will give you the binary distribution says that you pay for support for N machines. If you want to run it on N+1 machines, you owe Red Hat more money. It further says that Red Hat can audit you, just like a proprietary software company. The only way to get support from Red Hat is to sign the contract, which means that you either pay per machine, or you somehow get the code and do without support.
Red Hat actually makes the SRPMs for their Enterprise system available to all.
I wonder how bad beer will start to taste once many varieties of genetically modified yeast escape into the environment.
That's almost exactly what we said about egcs versus gcc when egcs started, to keep the FSF from flipping out. However, the result was that egcs ended up replacing GCC (what was originally planned as egcs 1.2 became gcc 2.95). This is good strategy for those who wish to avoid a fork: arrange that the fork can eventually become the main branch.
Whether xouvert will replace or take over Xfree86 depends on whether the majority of developers abandon the xfree86 ship and work on the xouvert branch.
Since Mapquest went online in 1996, he can be at most 14, and that's if we define "most" as "barely more than half".
For their first couple of years of operation, their database apparently had El Camino Real through Silicon Valley marked as a freeway. The driving directions would often tell you to stay on El Camino for about six miles, which considering there is an unsynchronized traffic light every three blocks, would get old fast. They appear to have fixed that problem some time ago, though.
Slashdot can now expect a DMCA lawsuit to turn over your name, so they can sue you next.
Oops, wait: Microsoft provided tweakui, so it is Microsoft that is tracking in circumvention devices.
Linus wrote: In other words, this should calm things down so that by the end of October we can look at the state of 2.6.0 without having a lot of noise from 'not strictly necessary' stuff."
That is, at the end of October he will "look at the state of 2.6.0". That's quite different from shipping it.
"Third-world South Korea"? You're talking about one of the most advanced countries on the planet.
The good folks at Red Hat have come up with a cool way to avoid some of the problems of monoculture in GNU/Linux: position independent executables. Addresses of code segments can be randomized at load time by the dynamic linker. The result is that common techniques for writing buffer overflow exploits no longer work, because every executable on every server is different. You can no longer insert code into a buffer whose length is not checked and then override the return address to point to it, because you don't know what return address to use. Worms can't spread if this technique is used.
While this technique still doesn't stop people from exploiting cross-site scripting bugs, it's progress.
RMS does not claim to have invented the concept of a multi-user timesharing system with security features; he just says that GNU and Linux software was designed that way from the beginning (taking ideas from existing systems). Since Microsoft started with the concept of one user and no security, then when they added multiple users and security they had to make tons of compromises to keep the legacy software running.
The result is that Windows XP has a very nice security model which is turned off by default, because if you turn it on, you can't use any apps that were originally designed for older Windows flavors.
No, you're wrong: the article says that the entire kernel was compared with the SCO kernel. What's confusing you is that SGI says the point was to look for anything that might implicate SGI, but in the process they compared the whole thing.
SGI used ESR's Comparator, plus other tools.
No, nothing I said was factually incorrect. Under the DMCA, the complainant sends a complaint, and if the defendant doesn't produce an affidavit (swearing under penalty of perjury that there is no copyright violation), the ISP must take down the site, or else the ISP becomes liable for copyright infringement if the complaint holds up. That's what a takedown notice is.
Microsoft is sending a demand from its lawyers to take down the site, but they aren't using a procedure that will cause the ISP to take action.
"Takedown notice" usually refers to a procedure defined in the DMCA that allows someone alleging copyright infringement to demand that an ISP take a site down. This is not what is happening in this case. Microsoft has no means of forcing Lindows to take the site down without a lengthy legal process.
If, under this financial model, people only pay 1/10 as much, musicians will make more money, because they aren't getting 5% of the money now.
I fear that the people who are locked into contracts are screwed, but the good news is that, if we can develop economic models that let musicians do better in the Free World than in RIAA-world, new musicians won't join, and established musicians won't renew contracts once their commitments are up. More and more musicians are going to start to realize that detaching yourself from your label and going independent can get you more money on 1/10 the sales, because you get to keep it.
But if anyone is an unsigned band: don't sign with an RIAA-connected label, no matter what they offer, because they're going to screw you and because they're going down.
But why would anyone think that Amazon could be trusted for e-commerce searches? If someone is selling a product for a lower price than Amazon is, do you really think that their search engine will point me there?
I think that Google could be beaten, but not by Amazon or Microsoft. The problem is that a search engine has to be perceived as neutral.
A little guy with much better technology could make headway.
Also, the likelihood of Google screwing up will increase greatly once they go public. Investors will demand more return, and management might eventually do something that blows the company reputation.
It's the problem with being a public corporation. The investors demand exponential growth, forever. This means that management feels pressured to keep doing new things, or try to squeeze blood from a stone. The problem is, they go beyond their domain of competence and soon hose everything.
Oregon is going to have to raise more money somehow. Already they've had to resort to closing schools three weeks early, and I suspect that the amount of dererred maintainance of critical infrastructure has reached dangerous levels.
Also, Oregon has relatively high property and income taxes, which any companies moving there will have to pay. Of course, there's the trick of living in Washington State (no state income tax) and shopping in Oregon, thus freeloading on the other citizens of two states.
Careful. There are a whole series of Creative Commons licenses. You need to read the terms in detail to figure out if you can use a Creative Commons-licensed work as part of a game.
Another problem is people applying the GPL to a sound file without understanding the implications. If that sound file was produced from a sound synthesis program, shipping only the output would be like shipping only the binary of a GPLed program. The GPL says that you have to provide "source code" and defines the source as the preferred form for doing modifications to the work. So, if you made an image with the GIMP, the XCF file (with the image separated into layers) would be the "source".
No, the vulnerabilities are due to new code in 3.7; the Red Hat and Debian people who backported only the security fixes to older OpenSSH versions are safe. They are not old vulnerabilities that were discovered by an increase in code vetting.
And with what money is ogg going to advertise?
Ogg will grow, but slowly, as manufacturers who need to cut costs learn about it.
You dropped a zero: it's $500,000 in taxes should you ship 500k units. And that's just for the MP3 royalties: if you do video, you've got to pay for that too.
If apt is used, conflicts are not necessarily a problem, provided that the conflicts are correctly described in the apt database. If you try to install a package that conflicts with some other package, you are given the option to proceed (and remove the conflicting package) or not, and with either choice your system stays consistent.