Oh man, the discussion contains so much nonsense, I don't even know where to start ranting about it.
Let's just take the scenario in question: a woman. A man. A child. The man doesn't believe he's the father - so there's a situation of distrust. The man gets a secret paternity test done - without the woman's knowledge and probably (since the child is surely still a baby) without the child's consent. In my eyes, a clear breach of trust. In the end, the relationship is already broken - by the mistrust.
So why are people railing against making secret paternity tests punishable? Because the alternative to a secret paternity test would be court proceedings and that would destroy the family. Seriously? Well, first of all, an alternative is also a voluntary and joint test. But if a joint test really can't be done, then the family is already broken - the court just makes the situation obvious, nothing more.
For me, the secret paternity test is just an expression of paranoia and inferiority complexes of men. And yet again, a practical general suspicion of these oh-so-evil women: they're all just sleeping around. Fits well with the Meisner remarks, because according to those, women are all just murderers.
If women are really that bad - why do so many men want to live with them?
Secret deception, snooping and suspicions have never saved a relationship. So there's no reason to particularly promote them - they are de facto an infringement on self-determination and therefore making them punishable would be completely correct.
But probably there would be a general strike of gene labs because a profitable source of income would disappear for them ...
At tagesschau.de - The News of the ARD you can find the original article.

Cardinal Meisner compares abortion with the crimes of Hitler and Stalin. 1 And who is pissed off? The Central Council of Jews in Germany.
Sorry, but I find Meisner's comparison absolutely outrageous for a completely different reason: he thereby puts women on the same level as dictators and mass murderers. And that weighs considerably more heavily than the violation of the Jewish monopoly on suffering ... From that perspective, Paul Spiegel's question - What can one expect from youth if a Catholic dignitary can relativize the millionfold murder of Jews in this manner and with impunity - looks quite different to me: what can one expect from a church that potentially puts 50% of its members under general suspicion of mass murder? But Aunt Catholica has always been rather limited in her worldview anyway.
I find the stance of Paul Spiegel equally concerning - doesn't he give any thought to the fact that Cardinal Meisner has greatly insulted women with his statement? Is this aspect of the whole story less important to him than pointing out that no one compares Jewish suffering in World War II to anything else because their suffering is unique? In this regard, the worldview of the Central Council of Jews is unfortunately also very narrow. What particularly bothers me is that this repeatedly provides argumentative fodder to the right-wing Nazi pack that is unfortunately still far too prevalent in Germany. 2 And no woman deserves to be compared to Hitler and Stalin - even if abortion doesn't fit into the worldview of these male-dominated organizations.
Not to mention, of course, the fact that I deny the Catholic Church any right to have a say in matters of general life because of its purely patriarchal and completely out-of-touch structure.
[1] When will we finally learn to overcome the history-falsifying totalitarianism idea - according to which Hitler and Stalin and various other dictators are lumped together into one pot? Hitler and Stalin have about as much in common as plague and cholera - both are deadly, both are sicknesses, but that's where the comparison ends. My toenails always curl up when I hear someone supposedly educated talk such nonsense ...
[2] It also produces very absurd consequences, such as the refusal to recognize other victims of the Nazi regime, like Sinti, Roma, or communist resistors (who, by the way, also had nothing to laugh about under Stalin). Which then, in combination with a memorial to the victims of National Socialism in general, actually only leaves an extremely embarrassing image.
At WDR.de there is the original article.
Well, it does work out. In his third article on optional static type declarations for Python, Guido van Rossum addresses all the comments and presents what I think is a compact and sensible proposal. Interfaces are a useful mechanism for stricter duck-typing - of course optional, so only where you need it. And automatic type adaptation is a good idea - it's about time that PEP 246 gets integrated into Python.
But the most important thing remains: it's optional. Whoever doesn't need it can leave it out.
Here's the original article.
I recently wrote about (P2984) the problems I've been having with Firefox on OS X. It has since turned out what's causing it. It's the Codetek Virtual Desktop Manager. As soon as it's active (I constantly have lots of windows open and otherwise can't find anything in the mess - and no, Expose wouldn't really help either), both Firefox and Thunderbird exhibit various misbehaviors:
- after startup the menu is empty. You first have to click in the background and then back on the application window for the menu to work properly
- the keyboard focus isn't always correct. Then you have to do the same as with the missing menu.
- after switching desktops (or also when normally hiding and showing the application again) the window is completely empty - only resizing it brings the content back.
As I said, this only happens with the desktop manager. Unfortunately I can't use Expose because I don't have 10.3. Besides, it wouldn't solve my problem: I need many parallel workspaces in which I have all the windows open for the respective task. Expose would only handle that very inadequately.
Bummer.

Hey, I didn't know about this yet: an emulator for various CPUs with just-in-time compilation and support for a whole mix of target and host CPUs. For example, emulating an Intel chip on PPC. Or conversely a PPC on Intel. Or ARM on PPC. And Sparc as a target is already in the works.
Particularly interesting for Linux users: it can do user emulation or system emulation. The latter does what Virtual PC does - present a virtual machine. The former simply offers the ability to run binaries for a different CPU on your own computer, even if you have a different CPU. For example, running Intel binaries on a Linux PPC - without major system emulation.
Due to the just-in-time compilation, the whole thing should also be significantly faster than Bochs. For OS X there's a graphical launcher that also handles the installation of qemu right away. Unfortunately only from OS X 10.3 onwards. Here's the original article.
For those like me who don't have time to chase after thousands of RBLs (lists of possible or alleged spam relays) to check whether someone has mistakenly listed their own server there again, these two links offer good services: they check a large set of RBLs all at once. The first link is the faster one:
Great. A classic of computer science literature (ok, a modern classic) is now readable online. The book is interesting because it explains many aspects of implementing a system environment suitable for Haskell or Miranda.
Unfortunately, it's only available online as scans in JPG format, so it's somewhat cumbersome to use - searching obviously doesn't work. But at least the table of contents is linked via an image map.
Here's the original article.