Looking for truth…
The BSD and the GPL ‘camps’ can happily live in harmony and they often do indeed. There is a certain hostility, however, between the OpenBSD folks and the FSF/GNU in particular. This raises questions that will be addressed here; but first — a little background.
Much of this recent tension began with this incident from last year.
Developers of OpenBSD took code from their brethren at Linux, violating the code’s licence, the GPL. To the horror of the Linux folk, the OpenBSD licence allows proprietary use.
Then came another incident.
Discussion continues on the Linux Kernel mailing list about the legality and morality of re-licensing BSD/GPL dual-licensed code under only the GPL.
More recently we saw the SFLC stepping in to address those issues which went the other way around (mind this case of reciprocity).
Let me therefore point out one last time that if the threats of litigation and bluster about crime and malpractice–none of which has the slightest basis in fact or law–were withdrawn, we would be able to resume detailed communication with everyone who has a stake in the outcome.
More on this debate here.
To save you from all that tiresome drama, here is a recent thread which is filled with hostility.
GNU Project and Free Software Foundation founder Richard Stallman posted a message on the OpenBSD -misc mailing list titled, “real men don’t attack straw men”, suggesting that some comments he had made were being misrepresented.
So that’s where things stand. All in all, there are a few angry people. Theo de Raadt and Richard Stallman do not get along, to say the very least. There is more to this story and we will come to this in a moment.
What’s the Big Deal Anyway?
Last week we spotted an article (blog) which discusses the impact of the BSD and GPL licences. It explains what they mean to Free software.
Corporations like Microsoft love the BSD-style licenses. Case in point is the Microsoft NT TCP/IP stack, which is basically a binary copy of the BSD TCP/IP stack. With a BSD-Style license, while code can not be stolen, rights of ownership cannot be enforced either.
As pointed out a week ago when Nokia’s latest acquisition got announced, a BSD licence for Trolltech’s Qt would be good news to Microsoft.
The founder of JBoss addressed a very similar question just a fortnight ago. He talked about dual licences too.
Quickies #5: GPL or BSD?
At the end of the day the argument is also one of philosophical taste. I say the GPL is great because it enables individual developers to grow businesses fast on dual licensing. Others may prefer the more permissive approach of letting anyone do whatever they please with their code.
In the middle of December, another blog post on this topic earned a fair bit of attention. It described the GPL/BSD dilemma as a question of freedom’s sustainability.
So what can we conclude from all this? Both license models make software free, but only GPL software is sustainably free. The BSD gives greater freedom, the GPL gives more freedom. Choose which one you value more.
Why the hostility then?
Looking Beneath the Surface
Based on the above, developers truly have the freedom to choose licences for their free software. There is little or no reason to fight here, so the question to ask is: might someone be fueling unnecessary wars? As far as drivers are concerned, reconciliation appears to have already been reached.
“With BSDs, there is the expectation that many sides will work as a group to achieve the same goals.”So, we have begun studying to see if the BSDs are somehow being used to combat GNU and RMS. The mailing lists may contain evidence, based on something we were told. Explaining what appears like hostility or apologists is hard. Both sides of this debate have communications that are not visible. Transparency is still limited, especially because of E-mail.
Where Microsoft or Apple are concerned, the main difference is that there’s no money involved in at least one side, so motives are not easy to interpret. Sometimes we still perceive this as a struggle between commercial interests of one company and ‘everybody else’, including the innocent consumer. With BSDs, there is the expectation that many sides will work as a group to achieve the same goals. It certainly works for the variety of different GNU/Linux distributions. Moreover, it’s worth remembering that similar components like the desktop environments are shared.
Example of Civil Wars
A reader contacted us regarding this older post. The reader states (and yes, permission was granted to publish this):
Ok. Bruce Byfield lost that round but don’t let him drag you down. And _don’t_ let the MSFTers play you two off against each other. (*)
The head-on style works well, but keep in mind that name calling is a distraction intended to change the subject and keep you from going after the real target. If you go head-on after the distraction, he’s got you.
Think of it as the red cape to keep the bull off the matador. Or of the birds that squawk and limp as you get close to their nest: the more they squawk, the closer you are. Or dogs that bark when they are frightened. Let ‘em bark. That means you’re winning.
In addition to the ‘hater’ label, you’ll also see MSFTers and their fifth-columnists claim ‘conspiracy’ to try to discredit criticism. So rather than letting them change the subject to a debate about ‘hate’ or ‘conspiracy’ pursue instead the points which caused the name calling.
Also it’s interesting to see them try to play bsd vs gpl(*), two license which have very much in common. These are simply tools with overlapping, but somewhat different, goals. At least for the foreseeable future, one should avoid that distraction as well and when others get dragged in, steer the debate back to the strengths and similarities.
Again, rather than going after the label, go after what the label is trying to hide. That label is used to pull people off topic or to discredit discussion of quality, legality, ethics, economics of software.
“On the Internet, nobody knows you’re a dog …
… until you start barking.”
Why Attack Stallman?
Further to this, since we are aware of abusive messages getting posted and used against RMS, we wanted to get hold of the headers of some abusive messages that had been sent. Someone whom we contacted said:
I’ve looked briefly at the official archives, and they had only the web interface which has sanitized the headers. I delete list messages when I’m through with them. On many of these, I was curious enough to check the headers first and lookup who owns the domain. However, any and all of that can easily be forged.
But if you can get me a tarball of the mailing list archives for OpenBSD Misc, for example, I can extract them for you and do a whois lookup.
I suspect more that Theo is easily manipulated and that he has egregious human relations skills. In a lot of cases, his first move is to call a lot of names. He joined most of the threads very late, if at all. I would not worry about Theo, he’s like that, and it was his occasional antics that got him kicked out of NetBSD and caused him to start OpenBSD. (They waited till he went to sleep and removed his access.)
The lists get trolled. Theo and others unwittingly participate, sometimes. Othertimes, the trolls feed themselves until it becomes a problem. RMS handled the whole thing well. Though I hate to see him softening on MS Windows.
The FreeBSD lists are worse, having been infested with MSFTers more thoroughly and for longer. They feed each other a lot of disinformation about GPL and RMS. Sometimes that takes the form of trolls against RMS and Torvalds.
One aspect of that amounts to a denial of service attack against RMS. If they cannot burn him out, they can drown him out. I notice that many magazines will not touch on any of his topics, but instead go with jibes about his appearance, etc.
Slightly out of context, but more or less the same thing:
There will be no trail of payments and receipts leading back to Redmond. We currently have the following data to work with, assuming acquisition of the message archives:
+ A) frequency of words, phrases and themes
+ B) type of argument / fallacy
+ C) date, time and (purported) hostname of sender
+ D) clusters of A + C
The attacks on Stallman are no isolated incident. It appears as though many people are subjected to similar treatment, me included. I know for a fact that Pamela Jones gets intolerant messages too.
In USENET, I have my name shoved in the subject lines every day now. All kinds of abusive messages are published. Anonymous people try to poison the Web with names, accompanied by obscene words and negative connotations. A lot of it is slanderous and libelous. The purpose:
- Incite people against me
- Rob a person’s credibility
- Make the person emotional to evoke angry writings
Military propaganda had some similar themes, with examples such as false rumours about Napoleon having small penis and Hitler having just one testicle. Connotations and rumours work wonders! Speaking to the reader on this subject he adds:
Take it as a sign you are on the right topics.
However, there is the concrete problem with crapflooding Google and other search engines. Hence my complaints about Google News. Google News itself may be unimportant but letting MSFTers outside or inside Google bury topics is not appropriate.
I’m sure you’re busy but would you be able to track down a downloadable tarball of messages against you or RMS? For the latter, I’d suggest Misc OpenBSD Yes, if you can get the raw messages. However, about the only part of the header that has a high probability of accuracy is the date-time stamp. That and everything else are determined by whatever forwards the message.
Perl: the duct tape of the Internet.
CPAN: the bailing wire of the Internet.
Nothing on this has been done since the conversation, but it would be an interesting thing to experiment with shall time permit. Lastly, the reader adds in reference to an article that I published last week:
You mentioned Apple and BSD, which reminds me that I was recently reminded of these two items:
Of the BSD’s, ironically, FreeBSD strikes me as least free due to tolerance of proprietary binary objects. That opens the door to a whole world of security problems and vendor- and platform-lock-in. That gripe aside, though, this an excellent move towards opening up. For the most part, it give enormous opportunity and flexibility to the customers.
Also, QNX recently tooks some big steps in the same direction: 2000 (no source code, but free of charge and source code for drivers and apps):
2007: (even the kernel’s source code, though not yet an open source license)
There you have 2007 as a major tipping point even if the media refuse to cover anything except one vendor and that vendor’s interests.
If one turns focus to just the BSDs for a bit, the three main ones have kind of a three-way development engine going:
There are no real answers here, but stay tuned as we are likely to publish more about this in the future. █
“A couple of years ago this guy called Ken Brown wrote a book saying that Linus stole Linux from me… It later came out that Microsoft had paid him to do this…”
–Andrew S Tanenbaum, father on MINIX