So, I'm biased: I work for Canonical. But I'm guessing many would want to give Canonical this sort of data, so long as it's innocuously used (i.e., for the reasons stated in the original article). I don't want Canonical tracking my every move (that's Google's job;-), but I do feel fine letting it know that I'm an Ubuntu user, so that Canonical can more effectively count users and make informed decisions based on that information.
Reading the comments above, it seems I'm not alone. I actually went out and installed canonical-census so that Canonical has data on use (i.e., I've added myself to the total Ubuntu user count), as the package is otherwise only installed on OEM installations. But how many of you others (who installed Ubuntu yourselves rather than buying it through Dell or someone else preinstalled) would like an easy, opt-in mechanism for providing this information?
I know there will be plenty with privacy concerns, and I respect that. But I'm guessing many others would be happy to provide this sort of information. (Yes, you can use Synaptic to do this, but as Ubuntu becomes more and more mainstream there will be plenty of people who don't want to get into Synaptic or a command line.)
I'm not suggesting that Canonical has plans to broaden the use of this package. So far as I know, we don't. I'm merely asking whether you'd support making it more easily available and, if so, under what conditions. (Is there some value we could be giving users in exchange for that opt-in, for example?)
That's the point (read the full article). We keep expecting open source to topple old hegemonies, but the reality is that it's simply helping to create them (Google) and keep them in check (everyone, including Google). That's a very important role, but it's not the BigCo Destroyer role we too often assign to open source.
This is absolutely true, but isn't that same protection against competitors more efficiently realized through proprietary add-ons to the open, Apache-licensed core?
I'm surprised you had any trouble finding a non-partner to provide support. I see dozens on a regular basis. There are hundreds....
We would never ask our partners to do work for free. Why do you think it's odd that the partners would prefer that you use Alfresco for a fee?
It's not misinformation. See my comment below. I asked Eben about this in a public forum during his OSBC2007 keynote. He indicated that the FSF had compromised in order to get GPLv3 passed. When was the last time the FSF capitulated like that?
Having said that, I totally grok why the FSF might think it's legal case would be weak, but it's even weaker if it tries to bring a suit on a license it didn't write, and which is nowhere near as clean as the original GPLv3 was on network distribution.
I agree that the Apple stand-in is silly and annoying. But I'm particularly annoyed by the FSF taking a hard line on the music I listen to (not that important in the grand scheme of things), and wholly overlooking SaaS/the cloud. Despite Slashdot softening my original post (which is its right), all Rob/Timothy did is point to a summit or two that the FSF staged. Guess what? Summits mean nothing. The FSF had its chance with GPLv3 and completely let the industry down, only providing a Band-Aid after the fact with AGPL, which was not its invention and which Richard, Eben, and others all publicly stated was a minor (and kludgy) salve.
I asked Eben point-blank at OSBC2007 about why the FSF had gone soft on network services (which was originally part of the definition of "distribute" in the original GPLv3) in GPLv3. His answer? They had to compromise in order to get parties like Google to the table. When was the last time you heard the FSF capitulate like that, especially to a business interest?
As more and more software moves to the web, we're still without a good license to protect freedom. Instead of worrying about one company's device, the FSF should be worried about the cloud services that huge swaths of people are using...including the FSF.
And a Java-based application (like a content management system) behaves even more differently, yet a Red Hat-esque model has worked for Alfresco. It's possible that it will stall over time but that's where building in additional value around the core product becomes important. Happy experimenting!
Not a product. That is my point! I am arguing that MySQL should have gone that route rather than closing off add-ons to its database except to the extent that those add-ons are separate services.
Not actually true. RHN has always been proprietary (until they announced a change to this, though I think they have yet to open source it).
I was actually arguing (if you followed the link) that MySQL should follow Red Hat's policy: Make its code open but only (initially) available to paid subscribers. Once someone pays, they can redistribute as it's GPL. But few exercise that right.
So, we're in agreement. Btw, I talked with MySQL about this and it's not even remotely clear that they are making these add-ons proprietary in the way you envision. It sounds to me like they're doing what Red Hat does.
MySQL may simply be doing what Red Hat did before it: Keeping its core 100% open while offering new services (e.g., Red Hat Network) as commercial, add-on services. I don't think MySQL has done a good job of telling this story, but I do think that's the story (or should be).
Many didn't like Red Hat's split between Fedora and RHEL, but ultimately it has arguably been good for both Fedora and RHEL. MySQL just needs to better define what it's doing.
Connect the dots. It's very likely that Linux was dropped from retail stores because it was successful, not because it was a failure. Wal-Mart was one of the few companies to go on the record in favor of Microsoft's patent pact with Novell. Wal-Mart never goes on the record for IT purchases. There's clearly a very chummy relationship between Wal-Mart and Microsoft at the most senior levels. I'd say it was news of Linux's success that prompted Microsoft to pressure Wal-Mart on this issue, not Linux's failure to sell.
McAfee knows what is expected of it. It just doesn't want to comply. It's the same old story. "We want the software - it's great! But we don't want to have to change our 20th-century business model to get it." That's McAfee's problem, not the software's. It's not a GPL thing. It's a matter of unsated covetousness.
There is no positive way to see this for the industry. Long term, it's even bad for Novell. Novell has become a vassal to Microsoft - it can claim to be independent, but when 1/3 of Novell's revenue comes from its top competitor, there is no way it can make independent decisions. It depends on Microsoft for all of its profits, I'd be willing to bet.
Sadly, the minute that Microsoft views Novell as a threat, it will simply turn off the spigot and Novell will die. Novell needs to find independent means to grow its business beyond Microsoft.
It released its open-source Flickr application (Thalia) and it's actively working on open sourcing other projects that it's working on. Now if we could just get the corporate world to be as liberal/free with its code as the academics....
Give me a break. This is open source in the same way that Red Hat Enterprise Linux, JBoss, or any of these others are. To get initial access to the ENTERPRISE code, you have to pay, because we're not obligated to release GPL code to anyone who asks. Only to those to whom we distribute the software.
Our subscription contract gives you absolute rights to view, modify, etc. I'm not sure how much more open-source open source you wanted/were expecting, but we're open source enough to appease an open-source purist like me.
Slashdot Mirror
So, I'm biased: I work for Canonical. But I'm guessing many would want to give Canonical this sort of data, so long as it's innocuously used (i.e., for the reasons stated in the original article). I don't want Canonical tracking my every move (that's Google's job ;-), but I do feel fine letting it know that I'm an Ubuntu user, so that Canonical can more effectively count users and make informed decisions based on that information.
Reading the comments above, it seems I'm not alone. I actually went out and installed canonical-census so that Canonical has data on use (i.e., I've added myself to the total Ubuntu user count), as the package is otherwise only installed on OEM installations. But how many of you others (who installed Ubuntu yourselves rather than buying it through Dell or someone else preinstalled) would like an easy, opt-in mechanism for providing this information?
I know there will be plenty with privacy concerns, and I respect that. But I'm guessing many others would be happy to provide this sort of information. (Yes, you can use Synaptic to do this, but as Ubuntu becomes more and more mainstream there will be plenty of people who don't want to get into Synaptic or a command line.)
I'm not suggesting that Canonical has plans to broaden the use of this package. So far as I know, we don't. I'm merely asking whether you'd support making it more easily available and, if so, under what conditions. (Is there some value we could be giving users in exchange for that opt-in, for example?)
That's the point (read the full article). We keep expecting open source to topple old hegemonies, but the reality is that it's simply helping to create them (Google) and keep them in check (everyone, including Google). That's a very important role, but it's not the BigCo Destroyer role we too often assign to open source.
This is absolutely true, but isn't that same protection against competitors more efficiently realized through proprietary add-ons to the open, Apache-licensed core?
I'm surprised you had any trouble finding a non-partner to provide support. I see dozens on a regular basis. There are hundreds.... We would never ask our partners to do work for free. Why do you think it's odd that the partners would prefer that you use Alfresco for a fee?
Actually, there is some truth to this (though I know you were mocking the misspelling). I don't respond to trolls.
It's not misinformation. See my comment below. I asked Eben about this in a public forum during his OSBC2007 keynote. He indicated that the FSF had compromised in order to get GPLv3 passed. When was the last time the FSF capitulated like that?
Having said that, I totally grok why the FSF might think it's legal case would be weak, but it's even weaker if it tries to bring a suit on a license it didn't write, and which is nowhere near as clean as the original GPLv3 was on network distribution.
I agree that the Apple stand-in is silly and annoying. But I'm particularly annoyed by the FSF taking a hard line on the music I listen to (not that important in the grand scheme of things), and wholly overlooking SaaS/the cloud. Despite Slashdot softening my original post (which is its right), all Rob/Timothy did is point to a summit or two that the FSF staged. Guess what? Summits mean nothing. The FSF had its chance with GPLv3 and completely let the industry down, only providing a Band-Aid after the fact with AGPL, which was not its invention and which Richard, Eben, and others all publicly stated was a minor (and kludgy) salve. I asked Eben point-blank at OSBC2007 about why the FSF had gone soft on network services (which was originally part of the definition of "distribute" in the original GPLv3) in GPLv3. His answer? They had to compromise in order to get parties like Google to the table. When was the last time you heard the FSF capitulate like that, especially to a business interest? As more and more software moves to the web, we're still without a good license to protect freedom. Instead of worrying about one company's device, the FSF should be worried about the cloud services that huge swaths of people are using...including the FSF.
And a Java-based application (like a content management system) behaves even more differently, yet a Red Hat-esque model has worked for Alfresco. It's possible that it will stall over time but that's where building in additional value around the core product becomes important. Happy experimenting!
Not a product. That is my point! I am arguing that MySQL should have gone that route rather than closing off add-ons to its database except to the extent that those add-ons are separate services.
Not actually true. RHN has always been proprietary (until they announced a change to this, though I think they have yet to open source it). I was actually arguing (if you followed the link) that MySQL should follow Red Hat's policy: Make its code open but only (initially) available to paid subscribers. Once someone pays, they can redistribute as it's GPL. But few exercise that right. So, we're in agreement. Btw, I talked with MySQL about this and it's not even remotely clear that they are making these add-ons proprietary in the way you envision. It sounds to me like they're doing what Red Hat does.
MySQL may simply be doing what Red Hat did before it: Keeping its core 100% open while offering new services (e.g., Red Hat Network) as commercial, add-on services. I don't think MySQL has done a good job of telling this story, but I do think that's the story (or should be). Many didn't like Red Hat's split between Fedora and RHEL, but ultimately it has arguably been good for both Fedora and RHEL. MySQL just needs to better define what it's doing.
Connect the dots. It's very likely that Linux was dropped from retail stores because it was successful, not because it was a failure. Wal-Mart was one of the few companies to go on the record in favor of Microsoft's patent pact with Novell. Wal-Mart never goes on the record for IT purchases. There's clearly a very chummy relationship between Wal-Mart and Microsoft at the most senior levels. I'd say it was news of Linux's success that prompted Microsoft to pressure Wal-Mart on this issue, not Linux's failure to sell.
McAfee knows what is expected of it. It just doesn't want to comply. It's the same old story. "We want the software - it's great! But we don't want to have to change our 20th-century business model to get it." That's McAfee's problem, not the software's. It's not a GPL thing. It's a matter of unsated covetousness.
There is no positive way to see this for the industry. Long term, it's even bad for Novell. Novell has become a vassal to Microsoft - it can claim to be independent, but when 1/3 of Novell's revenue comes from its top competitor, there is no way it can make independent decisions. It depends on Microsoft for all of its profits, I'd be willing to bet. Sadly, the minute that Microsoft views Novell as a threat, it will simply turn off the spigot and Novell will die. Novell needs to find independent means to grow its business beyond Microsoft.
It released its open-source Flickr application (Thalia) and it's actively working on open sourcing other projects that it's working on. Now if we could just get the corporate world to be as liberal/free with its code as the academics....
Give me a break. This is open source in the same way that Red Hat Enterprise Linux, JBoss, or any of these others are. To get initial access to the ENTERPRISE code, you have to pay, because we're not obligated to release GPL code to anyone who asks. Only to those to whom we distribute the software. Our subscription contract gives you absolute rights to view, modify, etc. I'm not sure how much more open-source open source you wanted/were expecting, but we're open source enough to appease an open-source purist like me.