I first tried to install the Radeon Omage Drivers from here, which helped a little, but still didn’t really work right: I was able to set some resolutions, but not all, and especially not 1680×1050, my external monitor’s native resolution.

The solution for that is simple: In Display Properties | Settings | Advanced | Displays, un-check the checkbox marked “Preserve Wide Aspect Ratio on attached displays.” Then it just works.

-DNDEBUG -g -O3 -Wall -Wstrict-prototypes -DWITH_APPINIT=1 -I/opt/local/include -I/opt/local/include/python2.5 -c _tkinter.c -o build/temp.macosx-10.3-i386-2.5/_tkinter.o
unable to execute -DNDEBUG: No such file or directory
error: command '-DNDEBUG' failed with exit status 1

Seems like the port doesn’t call gcc [args] to compile the binary, but only the [args], which often start with -DNDEBUG. (which, incidentally, is hard to google for, because of the minus sign.)

afb@macports.org has the reason for this:

It seems that those python ports are expecting $CC to be set in build as well as in configure, and fall back to random values such as “” or “cc” or “/usr/bin/gcc-4.0″ when it isn’t. This probably also explains why rerunning the build sometimes succeeds, if something else was adding a $CC to the environment that can be picked up as a lucky guess second time.

A workaround is setting the build.enc environment variable like so:

sudo su
password:
set build.env="CC=gcc"

and then run port. Here’s the ticket for more info.

With Windows Vista, Ubuntu 7.10 and Mac OS Leopard we’re experiencing the newest generation of operating systems. All three are (to a point) better than their predecessors and acquired tons of new features, both big and small.

There is a feature though, that I sorely miss in all the OSes as of today.

Sometimes when a friend visits me, I want to copy some files from ~/Music/PublicDomain and some from ~/Music/CreativeCommons and some from ~/Movies/OnlyLegalVideosHere over to their USB drive.

Or sometimes I want to get 13 different Linux ISOs from my file server.

I usually end up having 10 or more copy-jobs open at once, regardless of whether I use Linux, Windows or OS X, which vastly decreases my overall performance: My computer gets slower and the files take longer to copy. And my screen gets all cluttered up.

So I’d like to suggest an advanced form of the copy dialogue: Add a copy queue. What I want is something like this:

I copy a file somewhere, and while I see the little progress bar window, I can simply drag another file or folder onto the bar, adding it to its queue. The files will then be copied one after the other.

The above mock-up shows how this could look like. When a whole folder is dropped into the window, its files will be added to the queue individually.

This has a lot of advantages over the many-windows method.

• It uses your available bandwidth more effectively.
  A drive usually will be faster when it doesn’t have to fetch files from different blocks at once. So you have less search time and your file will complete more quickly. Over the network, you should also have less overhead copying just one file at a time.
• Your desktop won’t get cluttered.
  This is obvious. Instead of many small copy windows, you’ll have one big window you can handle or hide more easily.
• You can sort your files by importance.
  Just drag the most important file to the top of the list. Once it has finished copying, you can start working on it right away, while the rest of the files is still being copied.
• Time estimates get more accurate.
  When you have 10 files being copied at once, the time estimates in the different windows will usually be far off. Compare that to copying a folder at once, where the estimate is usually quite good. With all files to copy in one queue, the OS should be able to calculate the remaining time quite accurately.

So that’s my idea. As far as I know, this is not possible in today’s operating systems, but I could be wrong.

The only way to get that functionality I know of is to stack cp instructions in the console with &&, but that’s not very user-friendly.

Alright, dear OS writers, get going! I’d be delighted to see this in a file manager somewhere. What about you?

I’m posting this hopes of helping people who google this, because I just spent two fucking hours finding it out. Here goes:

If you have a Via Epia M (EpiaM, Epia-M) motherboard in your barebone computer, and you want to boot from a USB Stick or USB Hard Drive, you have to know three things:

1. In the Bios, set the boot device to USB-ZIP
2. Also in the Bios, activate the USB-Keyboard option
3. Finally, before you plug in either a USB keyboard or your USB drive, remove the power chord of your barebone computer! Don’t just turn it off; in order for the Epia M to recognize a USB drive or keyboard, you have to completely cut its power before you plug either of these in.

Those were infuriating two hours yesterday evening! HTH

In the next days, I’m going to post my experiences and maybe even a HOWTO on creating a software RAID on Linux.