The FA is not about licenses in general. Instead, ESR makes a fairly narrow point about the long term economic efficiency of producing software. Assume for the sake of argument that the Free/Open-Source approach to development is superior to closed-source approaches. (Most of us, including ESR, think this is true.) In this case, he argues, the GPL has no economic advantages for software production over less restrictive licenses, though it is useful as a

... signaling behavior, like wearing a crucifix or yarmulke or pentagram - it helps build trust groups. But it has costs, too â" it creates a lot of needless fear from potential allies and users who suspect they wonâ(TM)t be able to control their exposure if they let it in.

He goes on to say

So the correct question to ask is this: Is the GPLâ(TM)s utility as a form of in-group signaling worth the degree to which fear and uncertainty about it slows down open-source adoption? Increasingly I think the answer is âoenoâ.

There's some useful discussion in the comments, but you'll probably have to wait until the slashdotting dies down before you can read them.

AFAICT, ISPs in the U.S. have been relying on the average bandwidth per user being fairly low. They knew that a few people would be downloading distros and other large files, but lots of their customers would only use email or read blogs. Some ISPs trusted this business model so much that they even started using "no bandwidth caps" as a marketing point. The other ISPs then felt forced to follow the same policy.

Long story short: they bet their businesses on internet usage patterns staying relatively stable.

Of course, internet technology keeps improving, and hence usage patterns keep changing. Internet video is now big (and still growing fast), and people can now (legally!) get their music, films and TV over the internet.

So the ISPs have lost that bet.

What happens next? I expect the ISPs to increase their prices and/or introduce bandwidth limits and/or go out of business.

(I live in Australia, where the ISPs always had usage caps, so my interest is purely intellectual.)

For once, I disagree with Ars Technica. In Python, integers automagically overflow into "long integers" (ie., BigNums). Therefore you can only compile integer operations into low-level opcodes (x86, LLVM, etc) if you somehow know beforehand that the no BigNums are involved and overflow is impossible. In general, you have to compile Python into calls on a python-specific run-time library instead of opcodes. (You can still produce code that runs much faster than CPython's stack-based bytecodes by using a register-based VM and by pushing type-based dispatch as early as possible.)

IMO, trying to generate language-neutral machine/LLVM code is a bad idea. The Parrot team seem to agree: Parrot byte-code will strongly reflect the source language; their aim is not language-neutrality but inter-language operability.

Great minds think alike! The Parrot team is already working on using LLVM for JIT code generation.

It was only on after reading the comments at Phoronix that I noticed the benchmarks used 64-bit (x86-64) kernels, not x86 as I had initially assumed. Maybe the use of kernel and compiler code that gets less testing than x86 is related to the odd performance quirks Phoronix found?

Given the huge number of Free/Open-Source games out there, it's natural that many of them suck. (And that's not counting the ones that were never finished.) It's also natural that commercial games win on immersive 3D graphics and other things that require big development teams. But there are some FOSS games that are absolutely terrific, and Wesnoth is one of them.

Many of the good things about Wesnoth are fairly obvious: quality music, good graphics, good in-game tutorials etc. Others are not so obvious: extensive playtesting, carefully tweaked scenarios. And one design choice that may look strange to modern console gamers turns out to be extremely clever: the hexagonal map and the lack of long-range attacks makes it relatively easy to write a superb AI.

Wesnoth has another unusual accomplishment: narrativist elements and hex maps in the same game.

CP/M took lots of inspiration from RT-11. DEC's command-language interpreters used "/" to denote options. (BTW, the CLI did this sort of parsing, not the command itself.) When the microcomputer took off, DEC was very popular so command syntaxes like "delete/confirm *.txt" or even "pip goodname.c=badname.c/r" seemed natural.

When directory-structured filesystems for microcomputers first appeared, there were two popular syntaxes for pathnames. Unix uses good old "dir1/dir2/file". DEC's high-end OSes used "[dir1.dir2]file" -- yuck. Apparently "dir1\dir2\file" was the closest anyone could get to the Unix syntax without introducing backward compatibility problems. I for one would have preferred the loss of backward compatibility.

I can't help wondering if Diebold's ATM software developers are using the Voting Machines division as a dumping ground for incompetents ... something like the B Ark in The Restaurant at the End of the Universe. How else can you explain an editable log (facepalm) on a voting machine?!

Another point: the company is trying to produce machines that they think county-level election officials will buy. Trying to produce a good electronic voting system would result in a very different system.

Companies aren't homogeneous. There must be lots of firms with really poor HR departments but really great IT or Engineering departments. The trick is to get past the HR people, and having the same name as some undesirable makes that harder.

Apple provide significant funding for two important academic projects, LLVM and clang. (I wouldn't be surprised to see clang+llvm supplant gcc on some architectures by 2010.)

There is a compromise between running as Administrator and limping along as a peon: use DropMyRights to run major internet-facing apps without full administrator access. (You patch the icons and Start Menu entries for the apps to run DropMyRights which then runs the .exe.) It's not a 100% solution, but it does help.

The main weakness of this approach is that Windows has dozens of ways to launch applications, and it's impossible to get DropMyRights to intercept all of them. There's a related tool, StripMyRights, which gives you two ways to make any .exe always run with limited rights, but I haven't tried it yet.

In actual fact we currently produce enough food for over 7 billion people. (Some is turned into ethanol, some grain is used to fatten up meat animals, some food goes to overweight people like me ... all because food prices are historically low.) The reason millions of people are starving today has nothing to do with global production shortages -- it's because of political failures.

Yes, Valve's insight is good sense, but it's not common sense: the big publishers are still paranoid about "piracy". Too many senior executives are unable (or unwilling?) to move away from the ship-boxes-to-retail-stores mindset (which even the big Music Manufacturers are now starting to discard).