A follow up to this post, Linus recently commented further on the GPLv3 issue. From the article:
At the same time, he suggests that his opposition may have been distorted or exaggerated. “GPLv3 is not 'evil,'” he says. “It just doesn't stand up to the great licenses out there, like the GPLv2.”
The current version of the GPL (GPLv2), Torvalds says, is “something where the open source people can meet with the free software people in perfect harmony. People from all over, regardless of their background, belief systems, or whether they are rabid about it or not, can happily agree about the GPLv2, and that's one of the reasons it's been so successful.”
By contrast, Torvalds says, “I think the GPLv3 is expressly designed to not allow that meeting. Exactly because the FSF considers us open source people 'heretics.'”
Once the GPLv3 goes from draft mode to live, it will be interesting to see if other GPL licensed projects simply move to the new version (remember the stock v2 GPL has the “or (at your option) any later version” clause, or do something similar to the Linux kernel and specify v2 only. I'd guess it depends heavily on what the final v3 ends up looking like. Also in the article was a comment from Linus about the supposed increased compatibility that GPLv3 has with other Open Source license, such as the Apache license:
In fact, Torvalds worries that one of the goals of GPL3 is to absorb part of the open source communities. For example, he notes that “one of the stated goals of the FSF with the GPLv3 was to expressly design the new license to be compatible with the Apache license. That sounds like a great thing, doesn't it? It sounds nice. 'Compatible' is such a nice word. Let's just all sing songs about it around the camp-fire.
“But if you actually look behind all the nice words, it's just a polite way of saying, 'We want to hijack the code of those projects that use the Apache license, too, and turn that code into GPLv3. Because the definition of 'compatible with the GPLv3' is strictly one-way compatibility. You can convert Apache-licensed projects into the GPLv3, but not the other way. Doesn't sound quite as much as a “Kumbaya” moment any more when you put it that way, now, does it?”
Speaking for himself and the Linux kernel, Torvalds says, “I don't need to try to hijack somebody else's project. I did my own. It stays GPLv2.”
Fairly harsh words. While there is certainly no love lost between Linus and RMS, hopefully this won't turn into a situation that further fractures the Open Source community. We already have enough of those situations.
–jeremy
Open Source, Linus, GPL, GPLv3, FSF, RMS, Linux
Archive for September, 2006
Quite a battle is currently raging on the Debian Bugzilla about the status of the Firefox package. The dispute is over the use of the trademarked Firefox logo. From the article:
When most people think about the Mozilla Firefox browser, they think of it as being open source and free.
The truth is, while Mozilla Firefox is open source, it is not entirely free, and it may not even be legally compatible with Debian GNU/Linux, one of the most popular community Linux distribution bases.
The Firefox logo is trademarked, so Debian doesn't consider it to be Free and will not include it as part of its distribution. Mozilla claims that using the Firefox name without the official branding is a trademark violation.
Furthermore, Mozilla claims that if Debian runs any patches to the version of Firefox included with Debian distros, it has to run them by Mozilla first for approval.
The end result of all this appears to be that the branding and name are going to be completely removed before Etch is released. It's not yet clear whether derivative distributions such as Ubuntu will follow suit. Mozilla has certainly stepped up it's enforcement policy recently, but the reality is that they have to. Not defending your trademark means losing it. They have a specific idea of what Firefox is, and if you want to use the name you need to abide by that idea. If you don't, you are still 100% free to use the code, just not the name. I'd say calling it “not open” is a bit disingenuous, but it's easy to see both sides here. It puts Firefox in a tough place when anyone can add any code to the product and still call it Firefox. After all, if it crashes the user will not care (or in many cases even know) that a custom patch caused the problem. All they know is that FF crashed. The flip side here is that having to run every patch though Mozilla does add an extra burden. In the end, it's a burden they need to bare if they want to use the name. Since it's fairly clear it's something they are unwilling (and due to the social contact probably unable) to do, out goes the name. In the end, the sad part is that there are 3 losers and no winners. “Firefox” gets less market share, Debian doesn't ship (by name) with a browser people expect and the users get nothing but confusion.
–jeremy
Debian, Etch, Firefox, Mozilla, Open Source, Linux
I.B.M., the nation’s largest patent holder, will publish its patent filings on the Web for public review as part of a new policy that the company hopes will be a model for others. From the NY Times article:
The policy, being announced today, includes standards like clearly identifying the corporate ownership of patents, to avoid filings that cloak authorship under the name of an individual or dummy company. It also asserts that so-called business methods alone — broad descriptions of ideas, without technical specifics — should not be patentable.
and
I.B.M. is not the only institution interested in using Internet collaboration to help improve the patent system. Last month, the patent office agreed to try a pilot project of soliciting outside comments on patent filings, including claims of prior art and originality.
I.B.M. is one of several companies that have agreed to submit some patent applications for open peer review as part of the project, beginning early next year. The others include Microsoft, General Electric, Hewlett-Packard, Oracle, Intel and Red Hat.
Great to see IBM stepping up and attempting to address what is a growing problem. They cite the delay in the legal system, specifically mentioning Congress, as one reason for moving ahead on this. I thought it was interesting to see Microsoft on the list of names of Technology companies participating in the more open USPTO pilot project. While I doubt we'll see then end of software patents any time soon, getting rid of business methods patents is a good first step. That fact that large patent holders such as IBM are admitting there is a problem with the system and actively doing something about it may give Congress the push they need to move forward on the issue. Let's just hope they get things right this time.
–jeremy
IBM, patents, Open Source
A follow up on a story we haven't heard much about in a while. It looks like Munich is finally moving forward with its plan to migrate most of their desktop computers to Linux. A couple quotes from the article:
Munich has begun its migration to Linux on the desktop, a year later than planned and nearly three years since the city announced its move to open source software.
“There have been some delays along the way but we’re now moving steadily ahead,” Florian Schiessl, manager of the Limux project for the city of Munich, said Thursday by telephone.
By the end of this year, the city of Munich plans to have migrated 200 computers to the open source desktop environment. “Most of these computers are used for relatively simple office communications,” he said.
The configuration is based on Linux Distribution Debian GNU/Linux 3.1, the KDE 3.5 user interface and OpenOffice 2.
It's great to see this highly publicized migration finally come to fruition. While they are certainly behind schedule, a lot of that had to do with some unneeded patent issues and the reality is that most projects of this scope end up being way behind schedule if they're completed at all. The fact that they added one year to the pilot project and still decided to move forward makes it clear to me that they are happy with both their original analysis and their decision to move to Linux. The plan now is to have 80% of the desktops converted over by the end of 2008. With about 14,000 desktops total and some fairly complex processes used by some of the larger departments, this seems like an aggressive but realistic goal. We'll certainly keep watching.
–jeremy
Linux, Open Source, Munich, Debian, KDE
It's a well know issue that Linus is not a fan of the new GPLv3 (previous GPLv3 coverage here, here, here and here). A group of kernel maintainers have now released a document, entitled The Dangers and Problems with GPLv3, that explains their position. Here is the document preamble:
This document is a position statement on the GNU General Public License version 3 (in its current Draft 2 form) and its surrounding process issued by some of the Maintainers of the Linux Kernel speaking purely in their role as kernel maintainers. In no regard should any opinion expressed herein be construed to represent the views of any entities employing or being associated with any of the authors.
The document is a fairly in depth overview of the current position of many high profile kernel contributors and contains a mix of history, goals and current perceived problems with the GPLv3. They bring up specific issue with “DRM Clauses”, “Additional Restrictions Clause” and “Patents Provisions”. Overall, if this is a topic that interests you I'd encourage you to head over to LWN and read the entire article. As I've stated before there seems to still be some significant resistance to the GPLv3 and we could be in real danger of a situation where some GPL software is incompatible with other GPL software. That's a situation that isn't good for anyone. One hopes these issues can be worked out amicably and things can move forward smoothly. Given how adamant both sides seem in their views, especially with regard to DRM, I'm not sure this is going to be the case. As always, this is a topic I'll be posting updates to as it unfolds. In a related story, a poll of kernel developers was recently posted with the following results (full poll details available here):
Total Votes Cast 29
Average Vote -2.0 +/- 0.7
Lowest Vote -3.0
Highest Vote 0.0
Median Vote -2.0
–jeremy
Open Source, GPL, GPLv3, kernel, Linux, DRM, FSF
I just booked my flight so it's official – LQ is LWE UK bound. If you'll be in or around London on October 25 or 26 make sure to stop in the .Org Village and say hello. From the looks of things we are going to have the largest mod turnout ever. On a side note, US Air has got to be the worst airlines ever. I've told myself I'd never fly US Air again, but I had a credit so I felt obligated to use it. Believe or not after the fees and difference between online and phone pricing the credit shrunk to about 12.5% of its original value. That and the fact that in my experience they are on time about 20% of the time means this will be my last US Air flight. I'll happily pay more or just not travel to avoid these jokers.
–jeremy
LinuxWorld, Linux, Open Source, US Air, USAir
It looks like Google may be experimenting with OpenSolaris, according to a couple sources. From the article:
Google runs a stripped-down version of Red Hat Linux specially modified by its engineers. But another source, a Solaris systems administrator who recently interviewed for a job at Google, said he was told the company plans to create and test its own modified version of OpenSolaris.
“I am 100% certain that there are literally dozens of people horsing around with OpenSolaris inside Google,” said Stephen Arnold, a technology consultant and author of The Google Legacy. Moving to OpenSolaris, he said, would be a natural move for Google, with its large number of former Sun employees and its never-ending drive to push the performance of its data centers to the hilt. But Arnold said he doubts that Google, which finished rolling out its highly-secret data centers in 2004, is deploying OpenSolaris widely yet. “Will it quickly replace Linux anytime soon? No,” he said.
Now it's clear that Linux is fairly entrenched at Google and literally dozens of people at Google are always “horsing around” with all kinds of projects that will never come to fruition. That being said, what would the implications of Google dumping Linux for OpenSolaris be for the Linux community? I think the loss would be two fold. First, Google engineers put a lot of time and work into various aspects of Linux that end up moving upstream. Losing that would be bad, but in the end other companies would likely take up most of the slack. The other element is the hit to the reputation Linux has. “If it's good enough for Google..” is something you'll here quite a bit at Linux shops. While it wouldn't be a ding Linux couldn't recover from and it wouldn't actually impact the quality of Linux in any way, perception means a lot; especially in the corporate world. “Why did they make the migration” would be the subject of a massive amount of scrutiny, and rightfully so. Google is seen as a technically savvy company with talented engineers that really know their stuff. For them to move would be a huge seal of approval for OpenSolaris and something I think Sun would absolutely love to see happen. All in all, this is just another rumor at the moment, but it's definitely one the Linux world will be keeping a close eye on.
–jeremy
Sun, OpenSolaris, Linux, Open Source, Google
The announcement about seeds lead me to this LWN article. I often talk about the advantages and strengths of Open Source here, so it's only fair that I also talk about where it is weak and can use improvement. Now, we know that Open Source produces better code. But if you look at most of the Open Source projects that have gained mass adoption (think Firefox, Apache, the kernel, etc.) you will notice a few similarities. They have a somewhat closed aspects of them outside of the actual code, the have a somewhat formal organizational structure and they have a set of defined processes and procedures. I think it's clear that as projects get to a certain size and popularity, they need these things. A while ago, Gentoo moved to a more democratic system. From what I'm reading (note: I am not an active Gentoo user nor do I regularly keep up with the Gentoo community in depth. I've spent a decent amount of time over the last day or so reading as many different sources of info as I could, and am drawing my conclusions from that) that transition has caused some problems. To me, when Danny left, Gentoo lost more than its creator – they lost their leader. They're now learning the hard way that consensus is the lack of leadership and the result seems to be analysis paralysis. This isn't something that is specific to Gentoo though, it's an easy trap for any project (especially one that gains wide exposure) can fall into. Losing your leader isn't a death knell, but it certainly exacerbates an already tenuous situation. I think the Gentoo situation is underscored by the fact that even the announcement that Danny is returning was meet with resistance. The issue wasn't at all that he was coming back (quite the contrary), but with the process (or better stated, the lack thereof). There are a couple lessons to be learned here and it's an issue that I pay close attention to, as the founder and leader of LQ. This is a topic I'll be thinking about more and posting about in the near future. In the mean time, I welcome your comments on the subject. It's a topic that I think is vital to the continued healthy growth of the Open Source ecosystem.
–jeremy
Gentoo, Linux, Open Source
The Dunc-Tank is an experiment to see how targeted fund raising can improve Debian. From the article:
The brainchild of Debian project leader, Brisbane, Australia-based Anthony Towns, Dunc-Tank is an experimental project that seeks to find ways of funding Debian development. That money will then go into paying people to sit down and do “useful” Debian work rather than relying on a core of dedicated volunteers to contribute code outside of their regular day jobs.
“Most developers work on Debian in their spare time, and the developers who do the most work on Debian devote most of their spare time to it. When that spare time disappears, due to work commitments, family commitments, holidays or sickness, so does the work on Debian,” Towns said.
To achieve its goal, Dunc-Tank is funding release managers Steve Langasek and Andi Barth to work full time on etch's release over the months of October and November respectively. All things going smoothly, etch will be released on December 4, 2006.
It should be noted that there is no formal association between Dunc-Tank and Debian, and in fact the project was meet with some resistance on debian-private. One of the main objections seemed to be the potential that it could ruin the volunteer nature of the Debian project. Done properly I don't think that has to be the case, but it is something they need to be cognizant of. Overall I think the idea has merit and this could set a precedent that results in greater improvements in Open Source projects, especially in areas that aren't “interesting” to the average OSS developer. To be honest, I'm always surprised how little bounty and payment type systems are used in Open Source. I suspect we just haven't found the right model yet. Dunc-Tank isn't quite ready to accept donations yet (they hope to be ready within a week), but when they are LQ will be chipping into the pot.
–jeremy
Debian, Linux, Open Source, Dunc-Tank, etch
There are some posts you simply aren't prepared for and think you'll never have to make. This is one. One of our mods passed away far too young. David, mcleodnine on LQ, was a long time member and one of the very early mods (he was also a panelist on the inaugural episode of the LQ Radio Show). He was the quintessential computer enthusiast and turned his hobby into a business that recently went into the black. He was a guy you would run into at all kinds of odd hours on IRC and have a great discussion out of now where. He'll be missed more than I'm sure he knew. LQ is missing more than a mod…it's missing a friend. Out of respect, LQ will be closed for a short time later today – a moment of silence.
–jeremy
Subscribe