Friday, December 30, 2005

Why Rights Are Not Always Important

Consider this article. Excerpt: FBI is checking mosques with Geiger-Mueller counters to check whether they contain radioactive material. And of course, some liberals are very upset by this: "All Americans should be concerned about the apparent trend toward a two-tiered system of justice, with full rights for most citizens, and another diminished set of rights for Muslims."

The problem with such statements is that they are completely missing the target.

Let's go back to the roots: the ultimate goal of the values of modern society, democracy, human rights, civil rights etc. is trying to reduce human suffering as much as possible. Everything else is but a mean to achieve this end, the ultimate end is only the reduction of suffering and nothing else, never, ever. It is only the end that must not be changed, the means, the tools could should always change to reflect the changes in the circumstances and in the challenges.

Throughout the history, the majority of human suffering has been caused by other humans, and within this subset of human-caused suffering the majority has been caused by state oppression. This is why a few years, decades or centuries ago - it depends on which country you live in - people who wanted to reduce human suffering decided that because of this, let's protect people both from the state and from each other. Let's figure out concepts that assure that the state - or other humans - won't make people suffer. Let's define what neither the state nor others cannot do to you. Let's call these concepts "rights". Human rights, civil rights - you can call them whatever you want. These rights are the means they developed to attack the historically most prominent causes of human suffering.

These concepts, these rights are actually working quite well whenever they are upheld. It would be completely stupid to deny this: the whole fact that I can and may publish my thoughts here is due to these rights. They are important tools. I would never deny that.

Nevertheless, however important they are, they are nothing but tools. They are just tools, just means to reduce human suffering. They only have use value, they do not have an absolute value. Whenever their use value is in doubt, it is not a sacrilege to break them. They important thing is only the ultimate end - the reduction of human suffering. The tools, the rights are less important.

We should always look at the end and compare possible harm with possible benefit. When a mosque is searched with Geiger-Mueller counter, it is only a minor inconvenience for the believers. Actually, if my own home would be searched this way, it would also be a only a minor inconvenience for me too. It wouldn't exactly make me happy but I'd forget it in two days or so. It might hurt my ego but that would only mean I need to do more meditation. So, the possible harm is about zero. The possible benefit is preventing a terrorist from blowing an A-bomb and killing millions. (Of course, it is not an effective way to prevent them doing it: after the first mosque was searched, if any terrorists had some nuke in a mosque anywhere, I am sure they have quickly removed it and put it elsewhere.)

This is a situation where rights are failing: the possible benefit is a billionfold bigger than the possible harm, so the search should definitely be done. It means in this case the rights are a wrong tool to reduce human suffering. And whenever the tool is wrong, you should not use it. There are definite situations where rights are not important, because they are failing to reduce human suffering. I think we should look at things from the viewpoint of what (possibly) reduces suffering and what (possibly) increases it and always compare harms with benefits and try to find an optimal solution. If rights comform to that solution, it's nice.

If not, let's suspend some of the rights. There is nothing wrong with that. You don't have to stick to your tools when they are not appropriate.

No harmless breach of some abstract rights are as bad as killing millions with an A-bomb.

Wednesday, December 28, 2005

Linux: where is the killer application?

OK, let's start with a question: what do these Open Source applications have in common: Firefox,, Apache, MySQL, PostgreSQL, GIMP, Perl/Python/Ruby interpreters, Rails, Eclipse, SPE, GTK with Glade, EMACS, and bash?

Answer 1: they are probably the most popular Open Source applications - generally, if you are a Linux user, you spend 90% of your logged-in time using these software.

Answer 2: all of them have a Windows version too.

Answer 3: usually the Windows version is easier to install and configure, and often packs more features.

Compare, for example, installing and configuring MySQL + Apache + PHP on Linux with the one-click graphical installer of WAMP.

Compare that on my Windows box at work offers to create new documents when I right-click the desktop, but not on my home box running Ubuntu Breezy.

Compare that it took me hours to install and configure Ruby + Ruby on Rails + MySQL + RADRails on my Ubuntu Breezy box until I gave up, googled around and found that you have to install some MySQL dev packages. Should Claudio not write this very useful article, I'd never have Rails at home as I was really close to giving up. Now, just compare it with the one-click "Instant Rails" installer on Windows.

Don't misunderstand me: this is not an anti-Linux rant. I am using Linux for more than four years now, started with SuSE 7.3 and then went for UHU 1.1, Gentoo 2005.1 and now I have Ubuntu Breezy. I like the Linux culture because I value idealism over profits, freedom over restrictive licences and gift cultures over quid pro quo. I am willing to accept my share of suffering Linux often gives me in exchange for having this nice warm feeling of being part of an idealistic and humane culture. (Although I often feel like a damned masochist as on my office box running Windows everything works click-click-click while at home I always have to dig-dig-dig to get things done on my Linux boxes, but hey, idealism alwas demanded a price. No free lunch.)

I also think that based on the amazing level of usability development Linux applications went through in 2004-2005 (consider GAIM, for example), Linux will become at least as usable as Windows by 2007-2008.

But the killer application is still missing: all the important Open Source applications are also there for Windows. If you are a user and spend you time browsing the web in Firefox and working with spreadsheets in OpenOffice and drawing with GTK, or you are a programmer and hack Perl code for Apache in EMACS, are you really interested what operating system you are running on? Is the OS important anymore?

And if the OS is not important, are there any really good reasons why a desktop/notebook user should switch to Linux? Where is the killer application, where is the strictly Linux-only application that gives you so much benefit and is so uber-cool that it justifies the considerable investment in learning and re-adjusting your customs that switching to a different OS requires?