A view from Huygens

NASA has build a video of the Huygens probe's descent into Titan's murky atmosphere. The movie was built with data collected during the 147-minute plunge through Titan's thick orange-brown atmosphere to a soft sandy riverbed on Jan. 14, 2005. In 4 minutes and 40 seconds, the movie shows what the probe saw within the few hours of the descent and the landing.

How to build Xephem-3.7.1 on amd64

I just installed the latest Xephem-3.7.1 onto my debian amd64. The problem with the default deb package in the Debian repository is that the version is quite old, at 3.4.5. Unfortunately, building Xephem-3.7.1 isn't quite straightforward; here's a small howto to craft the binaries (no package, unfortunately) :

* download the source from the Xephem site
* unpack the tgz tarball
* create and run this script in the ./xephem-3.7.1 directory :

#!/bin/bash
# Fix gcc flags
for i in libastro/Makefile libip/Makefile \ 
 libjpegd/Makefile liblilxml/Makefile GUI/xephem/Makefile ; 
do
  echo "Fixing CFLAGS in ${i}"
  sed -e "s~^CFLAGS[ ]*=\(.*\)-O2\(.*\)~CFLAGS= \1 \2 \ 
  ${CFLAGS}~" -i ${i} || die "sed failed"
done
sed -e 's~^CFLAGS[ ]*=\(.*\)$(CLDFLAGS)\(.*\)~CFLAGS=\1 \2~' \ 
  -i GUI/xephem/Makefile

* Remove the ./xephem-3.7.1/libXm/linux86 directory, as it contains 32bit libraries
* Create a symlink ./xephem-3.7.1/libXm/linux86 towards /usr/lib
* Install the lesstif2 and lesstif2-dev packages
* Build as described in the INSTALL file

That's it ! Copy your binary to /usr/bin, and you're ready to rumble.

Anonymous Wed, 01/24/2007 - 06:59

Tried to execute above script in a bash shell but trip up in line 8 with:

line 8: die: command not found

kristof Wed, 01/24/2007 - 09:30

In reply to by Anonymous

You could replace 'die' with an 'exit 1'. Your script will continue to fail, however, cause your sed command onthe same line is failing.

Denebe Sat, 01/27/2007 - 23:41

Hello,

I am a linux beginner. My OS is opensuse 10.2 x86_64, and I am trying to install xephem 3.7.2.
I wrote your srcipt, I saved it to "test", and I made it executable in xephem directory.
After I have a mistake answer:

thierry@linux-k2h5:~/Desktop/xephem-3.7.2> ./test
./test: line 4: syntax error near unexpected token `libjpegd/Makefile'
./test: line 4: ` libjpegd/Makefile liblilxml/Makefile GUI/xephem/Makefile ; '
thierry@linux-k2h5:~/Desktop/xephem-3.7.2>

Can you help me to find out what's wrong ?

kristof Sun, 01/28/2007 - 16:12

In reply to by Denebe

Looks like bash doesn't recognize the '\' character on the end of line 3. In fact, this character tells bash that the next line is part of the current line. So line 3 and 4 are one line. Try to remove all '\' characters, and place the following lines on one line.

denebe Wed, 01/31/2007 - 12:55

In reply to by kristof

Hello,

I din't write your script, I only used copy and paste. For a mysterious reason (for me), there was a dot at the end of the 4th line. Your answer show me my mystake. Now the srcipt work fine, thank you.

ZFS @Home

There's lot of rumours about Apple incorporating ZFS in Mac OSX. Yet there are many people that say ZFS is great for servers but people don't need it at home. Unix Admin Corner tries to explain why ZFS is possibly more needed at home.

ZFS mountroot, phase 1

Solaris Express 4/06 has been released, based on the Nevada build 36. The OpenSolaris community is already counting at Nevada build 39, but the real interesting part of build 36 is that it allows you to hand-assemble a system which boots from ZFS. The procedure is not for the faint at heart, but it clearly indicates that a ZFS root filesystem is getting real close.

OTOH, people are porting ZFS to Mac OSX.

Fake scientific papers

Dropped out of university too early ? Allways wanted a published scientific report with your name on it ? 'Controlling Digital-to-Analog Converters Using Amphibious Theory', 'Studying Forward-Error Correction and Model Checking with BUN' or 'A Case for Wider-Area Networks' are only some of the sounding titles of scientific reports the SCIgen robot can generate for you.

You're a science teacher, and hate the internet for it's ready to pick scientific papers ? You want to know in advance which students are going to deliver impressive reports, generated in a few seconds ? Fear no longer : here's the Inauthentic Paper Detector.

Note : I submitted one of my blog posts to this baby, and it rated my text as being classified as INAUTHENTIC with a 20.5% chance of being authentic text. I've allways known I was a lousy scientist :)