~bma / journal / tags / free software

I said I’d write about it, so I suppose I’d better. A few weeks ago I heard that Richard Stallman was to give a lecture at Manchester University; looking into it, though, it seemed to be prohibitively expensive, mostly because it would require staying somewhere overnight. A little later, though, I discovered he was also giving one at Cambridge University – and earlier in the afternoon, so it would be perfectly possible to get there and back in one day. Tickets were booked.

The morning of the 30th comes; I get up at 0600 (okay, okay, 0630), get ready, and leave the house. It’s not until I’m halfway across the park that I realise I’ve left behind the maps of Cambridge I’d printed off. Jog back across the park, down the hill, up the stairs, into the office. One page printed, out of paper error. No paper around. Stick that one back in to print on the other side. It prints on the same side instead, but luckily both parts are mostly readable. Back out of the house, to the station, onto the train just in time. I find Gem already on board and playing with her EeePC, and spend the next few hours alternating between sleep and The Selfish Gene.

Skippy joins us at Reading, and spends the last stage of the journey to London arguing about the economics of cars versus public transport with a random stranger. We arrive at Paddington, I get something to eat, then we head across London to King’s Cross, then onwards north to Cambridge.

We arrive at Cambridge with no further mishaps, and get the bus into the centre of town. As we get off, we ask the driver where to get the bus to Madingley Road Park & Ride, and he tells us it’s just around the corner. After a few minutes waiting, we realise that although the bus that stops here is the right number, it’s going the wrong way. We hunt around for the real bus stop, only to see our bus go right past us – although this does mean we know where it stops now, and Cambridge has public transport that doesn’t suck, so we only have to wait a few minutes for the next one.

Madingley Road is quite long; it’s several kilometres from the city centre to the Computer Science lab. We spot what we think might be the road before it, and sure enough, the next road has a sign for the Cavendish Laboratory, which we know is on the same road. Skippy presses the bell. The bus goes right past the bus stop. Skippy presses the bell some more. Eventually, the bus pulls into the Park & Ride, about five minutes walk up the road, in the rain. Yay. We find our way to the William Gates Building (oh, the irony), and I get a picture of Hector by the sign before we go in.

The lecture itself is pretty interesting; rms explains that it’s not about free software, but rather an answer to a question people often ask when he talks about free software – can the same rules be applied to things that aren’t software? He starts off by explaining a little of the history of free software, and then the history of copyright – why it came about in the first place, how it has evolved, and how, although the laws we have now were perfectly reasonable a century ago, or even half a century, they’re not so useful now. After that, he proposes some possible changes to copyright law, to make it fairer both to authors/artists/musicians and to end-users. He recommends that functional works – software, recipes, textbooks, reference books, technical manuals, and so on – be required to be Free-with-a-big-F; "testimonal works" (i.e. works which explain an author’s opinions, beliefs, experiences, etc.) should be free to share and copy, but not free to modify, since that just leads to misrepresenting an author’s views. Finally, artistic works need not be free to copy, either. He also advocates reducing the length of copyright; ten years, he says, might be more reasonable. He goes on to talk about ways that musicians, specifically, could earn money without the overbearing presence of the record labels, like having a "Donate" button on music players that would automatically pay the musician a small amount of money.

The talk ends, and rms answers questions from people who, it seems, don’t quite get where he’s coming from. After that, Gem and I purchase some stickers and whatnot to support the FSF, and rms signs my copy of Free Software, Free Society – but refuses to sign the O’Reilly Emacs book that Gem had brought, because it’s not free. Hmm. We head out to the bus stop, and debate the hackability of networked bus stop signs that tell you when the next bus is due – though this one, for all it’s right next to the computer science department of one of the top universities in the world, insists that there’s no bus due when, in fact, there is. Oh well. Back to the train, grabbing a cup of coffee on the way. On the train, Skippy and I debate the value of free software for reading data from the onboard computer in modern cars. Back in London, Skippy takes us to CCK (very nice coffee and cake, even if the artwork is, um, "interesting"), and Gem discusses the website with the manager. This is followed by a mad dash across London as we try to get back to Paddington on time for the last train before the sleeper service – we make it, but barely.

Skippy abandons us again at Reading, and we’re left to sit around while the train waits for engineering works. We finally get back to Plymouth just before 0100.

From http://www.zdnetasia.com/news/software/0,39044164,62035655,00.htm:

Chizen characterized the open-source community as taking its inspiration from commercial companies. "The open-source community takes a lot of the practices and some of the ideas from commercial companies and enhances them," he said. "If we didn’t exist, there would be less to enhance."

So, basically, it’s okay for companies to exploit copyright and patent law, because if they didn’t the free software world wouldn’t know what to do with itself. Or something. What a lot of arse.

On the OpenBSD mailing list recently, there’s been an ongoing flamewar between Richard Stallman on one side, and apparently everyone else on the list on the other. Now, I happen to think that RMS is either mistaken, or that his beliefs are unreasonably extreme in this case. However, he’s earned my respect for remaining calm, polite, and on-topic. Various other participants in the discussion, on the OpenBSD side, have basically resorted to ad-hominem attacks; one guy continuously posts pages and pages of ranting about RMS’ hypocrisy, and Theo de Raadt can apparently come up with little better in the way of argument than "you’re a slimy hypocrite, when was the last time you wrote any code?"; not exactly showing themselves in a good light.

There’s been discussion some time back (September or so) about the use of BSD-licenced code (wireless drivers from OpenBSD) in the Linux kernel; specifically, that changes to the drivers were only released under the GPL and therefore unusable by OpenBSD (which won’t accept any licence more restrictive than the original Berkeley licence).

Most of the people appear to be missing the point. Of course, the Linux developers are perfectly entitled to do so–the OpenBSD licence doesn’t require anyone to release any modifications at all, and BSD code is used by, for example, Sun, Apple, and Microsoft for this very reason.

However, that doesn’t mean they should. Just because you don’t need to release your changes doesn’t mean you shouldn’t do so anyway; Apple especially have contributed a fair bit back to FreeBSD despite not being required to. And from free software developers, a refusal to release code under a licence that’s acceptable to the author is a little hypocritical.

It shouldn’t be a case of "do it because you have to". It should be, as I’ve mentioned_before, something you do because it’s the right thing to do. The Linux guys know it’s right, yet still they avoid doing it.

A while back, after all the uproar with Ubuntu including proprietary, binary drivers, Mark Shuttleworth announced that there’d be a version of Ubuntu with nothing but free software. The new version of Ubuntu has been released, along with Gobuntu, a version with “no firmware, drivers, applications, or content included in Gobuntu that does not include the full source or whose license does not provide the right to use, study, modify, and redistribute the body of work”.

Unfortunately, they’ve failed already, since Gobuntu includes the non-free Firefox logo, whose licence does not permit modification or redistribution.

Now, I don’t care whether you think modifying artwork is important–it’s a stated goal of Gobuntu to have no non-free content. The logo is not non-free content, so Ubuntu have failed in their goal.

Likewise, I don’t care if you think that the licence doesn’t matter because the logo is a trademark anyway–as Mark Pilgrim points out, that’s two separate restrictions instead of one, and therefore even harder to justify.

Rich will probably make some comment about Debian developers being idiots for caring too much about freedom. He’s never managed to justify that in the past, so I suggest he doesn’t bother this time. Plenty of free software projects trademark their name and/or logo; nobody but Mozilla feel the need to get arsey about it. If someone chose to create a spyware-filled or trojaned version of Firefox, calling it Firefox and putting the Firefox logo on it, Mozilla could still stop them under trademark law and it’d still be a crime under various Misuse of Computers-type acts. There’s no need for a non-free licence.

If you want a distribution that actually cares about freedom, try Debian, which I’ve always found easier to install and use, and more memory- and space-efficient, than Ubuntu.

Some time ago, Eric Raymond (pioneer of the open source movement, as opposed to the free software movement) wrote a paper in which he, basically, instructed Richard Stallman (pioneer of the free software movement) to "shut up and show them the code".

Well, for a start, this is rather rude, especially since rms has probably written more free software than almost anyone else (he’s spent the last 25 years writing GNU software, plus at least a decade before that working at MIT), and coming from esr, who–well, he hasn’t, put it that way–it’s especially unfounded. You can say a lot of things about Richard Stallman, but accusing him of not writing enough free software is not one of them.

Secondly, esr apparently just doesn’t get it. He claims that rms’ focus on freedom, on the rights of software users, is "bad tactics", when in fact, it’s not a tactic, not a means to an end–it’s the end itself. The goal is not, has never been, to write an operating system for which all the source code is available, as the term "open source" implies–it has always been to write a system which allows its users the freedom to use, study, share, and improve it–of which two of those freedoms require access to the source code. If anything, open-source is the means and freedom is the end, not the other way around as esr would have us believe.

For the majority of computer users, the term "open source" is meaningless; not (necessarily) because they’ve not heard of "open-source software", but because they have no idea what "source" is; a substantial portion of the rest simply don’t care, because they have no interest in seeing the source. Most people care about freedom, though, and would like to be able to download a program and use it without having to worry about how long before the free trial runs out, or whether the program will expect them to fill in all their details to be sent off to the vendor before using it, or be pestered for a licence key every time they start up, or be treated like a criminal until they "prove" that they’re not (we’ve had habeas corpus since 1215, you know? [1]). Why promote open-source, something only of interest to programmers? What about the rest of the world, users who couldn’t care less about modifying a program but still want the freedom to use and distribute it?

I’m sick of people using the terms “FOSS”, “FLOSS”, etc, for “free/open-source software” or “free/libre and open-source software”. It sounds ridiculous, especially because Free Software and Open Source software aren’t the same thing despite what the Open Source Foundation might like you to believe. Though most of what’s referred to as Open-Source is also Free, there’s a lot more leeway with what you can call open-source.

As I mentioned in the earlier post (linked above), the term open-source tends to be use by people who care more about ways they can make every penny possible out of any given situation; to them, it’s a marketing gimmick. Free Software is free because it’s beneficial to users - all users, equally, which is why the Junta of Extremadura in Spain and the City of Munich save themselves thousands in licencing by using Debian, as do hundreds of companies, schools, universities, and governments around the world, without having to worry about what it’ll cost to upgrade to the next version.

Yes, it has technical merits too: frankly, you’ve have to be smoking something that’s sold in little plastic bags to run a mailserver on Windows, which is why even the University of Plymouth, noted for its wide and unapologetic use of Windows, has front-line MXes that run Exim (though the loss of the Sun lab is mourned by Plymouth geeks).

Using more than one term for purportedly the same thing just confuses people; having to differentiate between Free Software, Open-Source Software, OSS, FLOSS, and FOSS (not to mention “Shared Source”) would confuse a new user a lot more than having to explain that Free Software doesn’t necessarly have to cost nothing.

Right. Firstly, let me make it clear that although I avoid proprietary software where possible myself, I have no objections to other people choosing to use it.

On the other hand, I do have a problem with Mark Shuttleworth singing the praises of Ubuntu for “…protecting the rights of people to get it free of charge, free to modify…” when in fact they do nothing of the sort - Ubuntu is, as I’ve mentioned previously, quite happy to distribute software of dubious freedom (binary-only firmware and device drivers) if it’ll make the system more appealing to the average non-technical user.

There’s nothing wrong with taking either stance, but you can’t claim to do one thing and then go and do the other. That’s dishonest, and is part of the reason I don’t use Ubuntu.

You also probably shouldn’t claim to do so in a blog post that you cross-post to the OpenSUSE dev list. That’s just plain silly.

Via Josselin Mouette’s blog, among others.

For an operating system that puts freedom ahead of popularity, try Debian.

As you may have noticed if you’ve used ?Debian (or Ubuntu), Firefox in those distributions doesn’t use the “standard” fox icon. This is because the fox icon is trademarked and can only be used with the permission of the Mozilla Foundation. Now, although they’re willing to give permission to Debian, they’re not willing to grant permission to everyone; Debian for its part won’t accept special exceptions granted to it like this - if Debian is allowed but nobody else is, it’s not free.

This wasn’t a problem. The icon was replaced, all was well.

However, now the Mozilla Foundation are saying the same thing about the Firefox name (the Mozilla part of the name went away at the same time as the icon). Debian will have to rename its version of the browser or face legal unpleasantness.

Is this really ?free software? Yes, it’s undeniably open-source, but that’s not the same thing - you can release the source code for something without it being free. I understand why the Mozilla Foundation is doing these things, but that’s not the point. Free software is more that just access to the source code - I couldn’t care less if I can modify most of the free software I use, though it’s certainly nice to be able to do so. It’s about freedom:

From http://www.gnu.org/philosophy/free-sw.html:

• The freedom to run the program, for any purpose (freedom 0).
• The freedom to study how the program works, and adapt it to your needs (freedom 1). Access to the source code is a precondition for this.
• The freedom to redistribute copies so you can help your neighbor (freedom 2).
• The freedom to improve the program, and release your improvements to the public, so that the whole community benefits (freedom 3). Access to the source code is a precondition for this.

In the case of Firefox, you’re limited in terms of what changes you make - if you make any major changes, you have to change the name, the icons, etc., or get permission from the Mozilla Foundation. That’s not freedom.

Update: After rereading the very long, rather dull and repetitive thread regarding the issue, I just want to clarify the situation to myself:

• The Firefox name can’t be used without the official logos.
• The logos can’t be used, since they’re not free (as mentioned above, each change needs Mozilla Foundation approval before the logos can be used).
• The logos are only non-free because the copyright licence includes those restrictions.
• The restrictions do exactly the same as the trademark on the logos does: if the copyright licence was free, the logos would still be protected by trademark law. Debian have no problem distributing trademarked material assuming they have permission.
• Mozilla don’t want to remove the restrictions in the copyright licence because they want to “protect their brand”, despite their brand being protected by trademark law anyway.

Crazy…

In the last couple of days, I’ve had a couple of chances to explain Free Software to people: once to a friend who was trying to install Microsoft Office and had no licence key, and once to another friend who just wondered what the hell that red swirly thing on my t-shirt was.

The first friend, when I told her about OpenOffice.org, asked what the catch was, and thought it all sounded too good to be true. I don’t know whether she tried it out anyway or not.

The second friend (a complete non-techie) thought that the idea was fantastic, and apparently has been explaining it to her (completely bewildered, presumably) friends. :) Unfortunately, the first reaction is all too common.