But the whining about GPL somehow makes me think that some people believe that they have the right to do whatever they want with any piece of software
You mean, as if it were "free?"
This demonstrates the deceptive tactics of the FSF. Not only does it fraudulently call itself the "Free" Software Foundation, but it also claims that GPLed software is "free." It isn't -- not only according to Debian's definition but simply according to common sense.
You speak from a heart full of anger and hatred about this subject.
No; it's your postings that show great anger. As do RMS's actions, which were undertaken out of anger. (For a history of how the FSF came to be, see Steven Levy's Hackers: Heroes of the Comptuer Revolution.) The GPL is an expression and a tool of hatred. We should therefore reject it.
There is no field of endeavor or group of people prohibited by the GPL.
Not prohibited; discriminated against. Which the GPL is absolutely intended to do. The only people who are impacted by the GPL are developers who are attempting to earn a living from their work. Stallman specifically says that he would like to prevent programmers from earning a good livelihood in his essay The GNU Manifesto. And, in a much more recent posting, he says, "Discrimination is the principle and the purpose." QED.
Free Software is about FREE SOFTWARE, not your freedom...
You've just shown, very clearly, the fraud that is inherent in the rhetoric of Richard Stallman and the FSF.
In the GPL's preamble and elsewhere, Stallman starts out talking about people's freedom. Then, once the reader's confidence is gained, he suddenly engages in a technique called "reframing," in which the point of view shifts subtly. Suddenly, he's defending the software's imaginary rights, and saying that people's rights must be abridged to protect them. GPL zealots, who either have read the GPL uncritically or have been taken in by the ruse, quickly follow, stating that software matters... and people do not.
Dead wrong, Joe. The GPL does not meet and has never met Debian's guidelines for "free" software. It violates at least two points: The prohibition against discrimination against a field of endeavor, and the prohibition against discrimination against a group of people. Even Stallman states, explicitly, that the GPL is intended to and does discriminate in this way. So, there you have it, right from the, er, gnu's mouth.
The GPL is based on copyright. This means that you can read, use and study code covered by it, and use and techniques and knowledge you gained by
doing so in your own work.
Not true. If you look at GPLed code, and then write something similar, you can be accused of having created a derivative work -- in which case you will lose the rights to your work. (Worse yet, you'll be forced to license it under the GPL, which could destroy your livelihood and/or your business.)
A lawyer who in fact works for the Open Source Initiative has written an essay explaining the dangers of perusing source code which is licensed in a way that is hostile to your interests. (In his case, he's looking at Microsoft's "shared source" license.) See
It's fine that he wants to include the rants, but to make then not only immutable in form but inviolate in terms of inclusion is where people have an issue. I think some people balk at the notion that this is fully in the same spirit as the freedoms they have with the software itself
It's worse than that. The rants are intentionally deceptive, and so he attempts to oblige those who pass the software on to propagate the fraud.
So what we have here is Debian claiming that the FSF is not radical enough in its promotion of Free Software?
No. It's pointing out that what the FSF is advocating is not actually freedom. The FSF advocates specific licensing terms for software which are not truly free. Yes, they involve giving some things away at no cost, but their true purpose is to harm specific groups of people and fields of endeavor which Richard Stallman does not like. Stallman is zealously pursuing his "enemies," whereas Debian seeks logical consistency and ethical behavior. This is worth respecting. It's sad that Debian has so much invested in GPLed software, because the GPL does not meet Debian's own definition of "free" (as explained earlier).
It also says "Activities other than copying, distribution and modification are not covered by this License" and "The act of running the Program is not restricted".
Bradley Kuhn recently announced that the FSF plans to modify the GPL so that it restricts the running of the program, or derivatives of it, for profit.
In fact, sendmail is in the unfree subdir as well.
Which is as it should be. Most people don't know it, but Sendmail has a viral license which gives Sendmail, Inc. special privileges that no other contributor has.
Ironically, despite the fact that it is promoted by an entity which calls itself the "Free" Software Foundation, the GPL is not "free" according to the Debian definition (and is also not "Open Source" under the Open Source Definition). Why? because it discriminates against a group of people (programmers who produce commercial software) and against a field of endeavor (the creation of commercial software). This discrimination is intentional. As Richard Stallman himself has written publicly:
In the GNU Project, discrimination against proprietary software is not
just a policy--it's the principle and the purpose.
Most proponents of open source and freely distributable software of all kinds recognize that commercial software also has an important place in the world. But the FSF does not. Its licenses are designed to discriminate, and therefore do not conform either to the Debian defninition or the OSI definition. This isn't a matter of ideology; it's a matter of fact. Stallman himself says so.
The problem with using QoS to throttle Windows Update is that they're served by Akamai. The updates can come in from many different places. Oh, and you can't use "Software Update Services" unless you're in control of the end user's machine and can configure it to use your server.
During the last 24 hours (shortly after this article was posted to Slashdot) our cache suddenly began getting hits on Windows Update content. Not all of it, mind you, but some.... Which is a good thing, because Windows Update traffic shot up to a whopping 70.1% percent of the last day's throughput.
I suspect that someone at Microsoft has been reading this discussion, which is good.
Most of the stuff that became cacheable, though, was for Windows XP. Windows 98 and Me updates (and we have a lot of users running 98 and Me) are still dragging the system down. I know, I know, you guys at Microsoft are trying to persuade Windows users to upgrade. But could you please not attempt to do this by making updates slower? We ISPs would appreciate it.
I've lost count of the times I've run into problems with transparent caches feeding me stale data; the last place I want to see stale data is when fetching security updates.
Doesn't happen. If there's an update to the update, it's done as a separate update.
Windows Downloads requires you to know what you're looking for. Windows Update automatically checks the machine and grabs all updates that haven't been applied yet. Needless to say, 99% of all users use Windows Update.
Actually, porn is down at 1-2%. Banner advertising (especially Netscape, Yahoo, Doubleclick, eBay, and AOL) would be next after Windows Update if it weren't cacheable, but since the cache handles most of it we only waste about 10% on it.
Just checked the stats for the past 24 hours (from a Squid cache). This time, *.windowsupdate.com generated 56.11% of the traffic, with a hit rate of only 2.37%. In short, Microsoft is eating (and expending!) huge amounts of bandwidth, and almost none of what is being transmitted can be cached. What a waste.
Microsoft's Software Update Services require you to modify all of the clients. Those that aren't modified still try to access Microsoft's Windows Update site.
So, since ISPs can't administer their users' systems, this really isn't an answer. Caching is a much better solution.
It's possible to set this up within an institution, but not at an ISP. You'll have hundreds of users, each running a slightly different version and configuration of Windows. And you, yourself, aren't the administrator of those machines nor do you have an institutional license for Windows. (If you're a smart ISP, you're not running Windows on many machines, if at all.)
Squid can't cache the updates. We've tested this. The URLs for the transfers (files come in as small GETs about 1 KB in size) aren't set up to allow caching.
I'll never buy a ThinkPad again. The one I'm about to retire (with great relief) had an MWave DSP that ran out of gas trying to serve as both the sound card and a modem.... It has never quite worked properly and can lock up the machine at random. I added a modem in the PCMCIA slot, and quickly discovered that if you populate the slots fully, or run CardBus cards, the machine runs out of IRQs. Service is also a nightmare; they take far longer than claimed and always try to find something to charge you for and/or blame the problem on you. Take my advice: Stay far, far away from IBM laptops.
Slashdot Mirror
...is the penguin skeleton in the cover art. Subtle.
You mean, as if it were "free?"
This demonstrates the deceptive tactics of the FSF. Not only does it fraudulently call itself the "Free" Software Foundation, but it also claims that GPLed software is "free." It isn't -- not only according to Debian's definition but simply according to common sense.
No; it's your postings that show great anger. As do RMS's actions, which were undertaken out of anger. (For a history of how the FSF came to be, see Steven Levy's Hackers: Heroes of the Comptuer Revolution.) The GPL is an expression and a tool of hatred. We should therefore reject it.
Not prohibited; discriminated against. Which the GPL is absolutely intended to do. The only people who are impacted by the GPL are developers who are attempting to earn a living from their work. Stallman specifically says that he would like to prevent programmers from earning a good livelihood in his essay The GNU Manifesto. And, in a much more recent posting, he says, "Discrimination is the principle and the purpose." QED.
You've just shown, very clearly, the fraud that is inherent in the rhetoric of Richard Stallman and the FSF.
In the GPL's preamble and elsewhere, Stallman starts out talking about people's freedom. Then, once the reader's confidence is gained, he suddenly engages in a technique called "reframing," in which the point of view shifts subtly. Suddenly, he's defending the software's imaginary rights, and saying that people's rights must be abridged to protect them. GPL zealots, who either have read the GPL uncritically or have been taken in by the ruse, quickly follow, stating that software matters... and people do not.
In short, you've been "had."
--Brett Glass
--Brett Glass
Not true. If you look at GPLed code, and then write something similar, you can be accused of having created a derivative work -- in which case you will lose the rights to your work. (Worse yet, you'll be forced to license it under the GPL, which could destroy your livelihood and/or your business.)
A lawyer who in fact works for the Open Source Initiative has written an essay explaining the dangers of perusing source code which is licensed in a way that is hostile to your interests. (In his case, he's looking at Microsoft's "shared source" license.) See
http://www.rosenlaw.com/html/GL8.pdf
--Brett Glass
Don't click on the above. (I certainly did NOT post it.)
It's worse than that. The rants are intentionally deceptive, and so he attempts to oblige those who pass the software on to propagate the fraud.
No. It's pointing out that what the FSF is advocating is not actually freedom. The FSF advocates specific licensing terms for software which are not truly free. Yes, they involve giving some things away at no cost, but their true purpose is to harm specific groups of people and fields of endeavor which Richard Stallman does not like. Stallman is zealously pursuing his "enemies," whereas Debian seeks logical consistency and ethical behavior. This is worth respecting. It's sad that Debian has so much invested in GPLed software, because the GPL does not meet Debian's own definition of "free" (as explained earlier).
Bradley Kuhn recently announced that the FSF plans to modify the GPL so that it restricts the running of the program, or derivatives of it, for profit.
Which is as it should be. Most people don't know it, but Sendmail has a viral license which gives Sendmail, Inc. special privileges that no other contributor has.
They can't "take" it. It was already freely given.
makes an incompatible copy of it, and bundles it with Windows to millions of people,
So? That's freedom.
leaving you to compete with a non-free version of your own work.
No; it's leaving them to try and compete with something that's available at no cost. Which puts them over a barrel, not you.
In the GNU Project, discrimination against proprietary software is not just a policy--it's the principle and the purpose.
Most proponents of open source and freely distributable software of all kinds recognize that commercial software also has an important place in the world. But the FSF does not. Its licenses are designed to discriminate, and therefore do not conform either to the Debian defninition or the OSI definition. This isn't a matter of ideology; it's a matter of fact. Stallman himself says so.
The problem with using QoS to throttle Windows Update is that they're served by Akamai. The updates can come in from many different places. Oh, and you can't use "Software Update Services" unless you're in control of the end user's machine and can configure it to use your server.
I suspect that someone at Microsoft has been reading this discussion, which is good.
Most of the stuff that became cacheable, though, was for Windows XP. Windows 98 and Me updates (and we have a lot of users running 98 and Me) are still dragging the system down. I know, I know, you guys at Microsoft are trying to persuade Windows users to upgrade. But could you please not attempt to do this by making updates slower? We ISPs would appreciate it.
Doesn't happen. If there's an update to the update, it's done as a separate update.
Windows Downloads requires you to know what you're looking for. Windows Update automatically checks the machine and grabs all updates that haven't been applied yet. Needless to say, 99% of all users use Windows Update.
Actually, porn is down at 1-2%. Banner advertising (especially Netscape, Yahoo, Doubleclick, eBay, and AOL) would be next after Windows Update if it weren't cacheable, but since the cache handles most of it we only waste about 10% on it.
Ahem.... Red Hat updates are cacheable. But the percentage of Linux users on our network is in the single digits. Most users run Windows.
Just checked the stats for the past 24 hours (from a Squid cache). This time, *.windowsupdate.com generated 56.11% of the traffic, with a hit rate of only 2.37%. In short, Microsoft is eating (and expending!) huge amounts of bandwidth, and almost none of what is being transmitted can be cached. What a waste.
So, since ISPs can't administer their users' systems, this really isn't an answer. Caching is a much better solution.
It's possible to set this up within an institution, but not at an ISP. You'll have hundreds of users, each running a slightly different version and configuration of Windows. And you, yourself, aren't the administrator of those machines nor do you have an institutional license for Windows. (If you're a smart ISP, you're not running Windows on many machines, if at all.)
Squid can't cache the updates. We've tested this. The URLs for the transfers (files come in as small GETs about 1 KB in size) aren't set up to allow caching.
I'll never buy a ThinkPad again. The one I'm about to retire (with great relief) had an MWave DSP that ran out of gas trying to serve as both the sound card and a modem.... It has never quite worked properly and can lock up the machine at random. I added a modem in the PCMCIA slot, and quickly discovered that if you populate the slots fully, or run CardBus cards, the machine runs out of IRQs. Service is also a nightmare; they take far longer than claimed and always try to find something to charge you for and/or blame the problem on you. Take my advice: Stay far, far away from IBM laptops.