So, here I found how to do it: How to discard changes in VMWare player?

Q: Running a virtual machine, how can I ensure that the changes gets discarded after I close it?

A: Shutdown the VM cleanly. Edit the vmx and add this line if the VM uses a scsi-disk

scsi0:0.mode = "independent-nonpersistent"

or this if the VM uses an IDE-disk

ide0:0.mode = "independent-nonpersistent"

Alternatively, use this two lines:

snapshot.action = "autoRevert"
snapshot.disabled = "TRUE"

Several ways lead to Rome. Both ways should work, but don’t mix them.

So I found this two links, that explain the basics on how to implement bash completion for simple commands:

• An introduction to bash completion: part 1
• An introduction to bash completion: part 2

And I implemented my completion for ssh, which takes the hostnames from the file ~/.ssh/known_hosts

# Add bash completion for ssh: it tries to complete the host to which you
# want to connect from the list of the ones contained in ~/.ssh/known_hosts

__ssh_known_hosts() {
    if [[ -f ~/.ssh/known_hosts ]]; then
        cut -d " " -f1 ~/.ssh/known_hosts | cut -d "," -f1
    fi
}

_ssh() {
    local cur known_hosts
    COMPREPLY=()
    cur="${COMP_WORDS[COMP_CWORD]}"
    known_hosts="$(__ssh_known_hosts)"

    if [[ ! ${cur} == -* ]] ; then
    	COMPREPLY=( $(compgen -W "${known_hosts}" -- ${cur}) )
        return 0
    fi
}

complete -o bashdefault -o default -o nospace -F _ssh ssh 2>/dev/null \
	|| complete -o default -o nospace -F _ssh ssh

To load it, save the above script into a file called ‘ssh-completion’ then add “source ssh-completion” in your ~/.profile or ~/.bashrc. Some Linux distributions offer a directory where you can deploy your completion scripts, for example debian should have them like /etc/bash-completion.d/foobar.bash.

Then go on the command line and type:

$ ssh [tab]

If everything went fine, you should see a list of servers to which you recently connected.

I don’t go in detail to explain the above script, since there are good tutorials around for this purpose, including the links I included before.
I only want to say that the ‘compgen’ is an internal bash command, used in the above script: what it does is taking a list of words (-W ${known_hosts} option) from which to try to complete a given partial word (in the above example the ${cur} variable, which is the word at the cursor).

Since completion is usually implemented with bash scripts, it’s good to keep handy a bash reference manual.

Scared of bash scripts?

I am. I continuously forget the syntax of loops and conditional expressions: I think that the bash language is quite obscure and counter-intuitive. So I guess that is also possible to implement completion as part of your program using your language of choice (Java in my case). You can implement a feature in your program that reads and analyzes the COMP_WORDS and COMP_CWORD environment variables, and generate the list of completion words, as the ‘compgen’ bash built-in command does. And, if your language allows it, set those words into the COMPREPLY environment variable; in Java I think it’s not possible to alter environment variables, but you can just output the words on the standard output and use a brief script to set that output to the COMPREPLY variable.
I’d suggest to bind this functionality to a ‘–completion’ option that can be specified in the command line, as we usually do for the –help option.

See git-completion.bash if you want an example of a complex completion script… bash can be hard.

Update

I just typed ‘complete’ on my mac, and I discovered I have 408 commands binded to some completion script. Including a more advanced ssh script which implements the hostname suggestions.

If so, you can still have a look at those configuration files in the running environment using the unzip Unix utility.

$ unzip -p /opt/myapp/lib/my-app.jar log4j.properties
log4j.rootLogger=INFO,stdout

log4j.appender.stdout=org.apache.log4j.ConsoleAppender
log4j.appender.stdout.Threshold=INFO
log4j.appender.stdout.layout=org.apache.log4j.PatternLayout
log4j.appender.stdout.layout.ConversionPattern=%d{dd-MMM hh:mm} [%-5p] %m%n

The -p option uncompress and prints the file content to standard output.

I like to include resource files in the deployable artifact, unless I want those files to be modifiable without repackaging. Hiding most of configuration helps to keep things simpler for the end user, but still those files can be accessed for the technical support.

By the way, the above log4j configuration is the one I used to replace some System.out/System.err in a simple command line based application. I was thinking to remove the date part ‘%d{dd-MMM hh:mm}’ and make it look like the output of ant/maven, but then I thought that -after all- the date is quite useful since the application in question is ran by cron redirecting the standard out into a file, so we keep that as a log of the last run, to have some clue in case of errors.

Examples:

1. The unix server that hosts this blog, provides me a restricted shell to manage mysql and public web files with minimal access privileges. On this server I periodically update WordPress and other things, and I like to keep the changes under control and backed up with git. The problem is that on this server I haven’t git binaries, and I cannot install software and libraries required.
   Having the files of my WordPress installation under git version control saved me a lot of time in the past when somebody hacked into the server and changed the php files to do nasty stuff. I was able to easily check what was exactly changed since last legitimate update with “git status” command, and restore things as before.
2. I own a little NAS (network attached storage) which runs a mini distribution of Linux which doesn’t have a prepackaged version of git to install, nor I cannot use any package management tool like apt rpm or yum etc.

So I thought that I can compile by myself the git binaries and have them deployed on the target machine. It is possible, but there is an important note: when you build a software on a unix server, the resulting binaries are referencing the libraries installed there, so you cannot easily port the binaries between servers since they have dependencies.
What you need is a “statically linked” version of the software, which fortunately it’s not so difficult to achieve. Binaries which are statically linked are usually bigger in size and will possibly require more memory to execute, but they won’t require the specific libraries to be present on the executing computer, since the libraries code is contained in the binaries themselves.

Here is how I built a static version of git on an ubuntu virtual machine, that can be ported to other unix servers:

# let's make sure we have all we need to proceed
$ sudo apt-get install libexpat1-dev asciidoc libz-dev gettext curl

# let's create the directory to host the built artifacts
$ sudo mkdir /opt/git-1.7.4.1-static

# we are ready to download and unpack latest version of git sources
$ curl http://kernel.org/pub/software/scm/git/git-1.7.4.1.tar.bz2 | tar xvj
$ cd git-1.7.4.1/

# then compile and install the files in the target directory we created
$ ./configure --prefix=/opt/git-1.7.4.1-static CFLAGS="${CFLAGS} -static" NO_OPENSSL=1 NO_CURL=1
$ sudo make install
$ sudo make install-doc

On the above commands, the thing to notice is the CFLAGS=”${CFLAGS} -static” which is used to specify that the libraries must be statically linked with the binaries.

The last thing to do is create a tarball of /opt/git-1.7.4.1-static folder and copy that on the target machine; adding the /opt/git-1.7.4.1-static/bin directory to the PATH variable and the /opt/git-1.7.4.1-static/share/man to the MANPATH variable.

If possible, it’s a good idea to keep the same installation path on target machine (/opt/git-1.7.4.1-static) since this path gets hardcoded in some git scripts during the build process. But it shouldn’t give too many problems, anyway.

Here you can see how my prompt looks like:

"luigi@hal9000:~/bin" is my usual prompt (username@hostname:folder), the red part indicates current branch, and the asterisk means that I have some modified files to commit.
In the above sample, also you see what happens when I type "git cl[tab-key]". The tab key triggers command line completion, and I can see available git commands starting with "cl"; in this case clone and clean.
The completion works not only on the command names, but also on file names and git object names, for example on branch names: if you type "git checkout mas[tab-key]" you should get it expaended to "git checkout master".

To enable the GIT bash completion, on OSX, you need to execute the script located at /usr/local/git/contrib/completion/git-completion.bash. This script is installed together with GIT utilities on OSX. To have its features enabled when you open a terminal session you should add following lines to your .profile:

# Set git autocompletion and PS1 integration
if [ -f /usr/local/git/contrib/completion/git-completion.bash ]; then
  . /usr/local/git/contrib/completion/git-completion.bash
fi
GIT_PS1_SHOWDIRTYSTATE=true

if [ -f /opt/local/etc/bash_completion ]; then
    . /opt/local/etc/bash_completion
fi

PS1='\[\033[32m\]\u@\h\[\033[00m\]:\[\033[34m\]\w\[\033[31m\]$(__git_ps1)\[\033[00m\]\$ '

The GIT_PS1_SHOWDIRTYSTATE variable is used to indicate that the prompt should display the asterisk or other characters to indicate that the state of the working tree is "dirty" (something is changed and it has not yet been committed).
The script git-completion.bash enables bash completion and contains also some functions that can be executed when the prompt (defined in the PS1 variable) is shown. In this case the function displayed is named __git_ps1 and it displayes the current branch; the rest of the PS1 text is to display a nice prompt with colors.
For Linux user (at least for Ubuntu ones), it is just necessary to define the PS1 variable in the ~/.bashrc file, as the git completion script is usually included by default when installing git.

Another nice trick, that is not enabled by default, is the coloring for the following git commands: branch, status and diff. Having colors enabled helps to faster and easier understand the output. To enable colors issue the following commands:

$ git config --global color.branch auto
$ git config --global color.diff auto
$ git config --global color.status auto

Configuring above git properties will affect color display as in following examples:

Branch:

Status:

Diff:

Hope you find those tricks as useful as I do find them.

The trick, how explained here, is to download the sources and copy them into android-sdk/platforms/android-X/sources folder (where X is the API version).

For example, supposing I've installed the SDK into ~/opt/android-sdk-mac_86 on my Mac, and I am developing an application with Android 1.5 (API level 3). I should download the source files from Android website using git. Then I need to checkout the sources with tag name "android-sdk-1.5_r3", collect all the source files and move them into the directory ~/opt/android-sdk-mac_86/platforms/android-3/sources.

Here's my source tree under android-sdk-mac_86/platforms; in red color the directory where to put the sources:

luigi@hal9000:~/opt/android-sdk-mac_86$ tree -L 2  platforms
platforms/
|-- android-3
|   |-- android.jar
|   |-- build.prop
|   |-- data
|   |-- framework.aidl
|   |-- images
|   |-- samples
|   |-- skins
|   |-- source.properties
|   |-- sources
|   |-- templates
|   `-- tools
|-- android-4
|   `-- ...
|-- android-6
|   `-- ...
`-- android-7
    `-- ...

Creating a directory and copy some files inside it's not much complicate. But once you downloaded the sources, you'll discover that they are spread in several sub-directories. Also you need some practice with git command line utility to download the sources check available tags, switch to the proper source tree, then package all the source files into a folder or an archive.

I created following script, which runs on OSX (and hopefully on Linux as well).

luigi@hal9000:~$ cat bin/android-src-build.sh

#!/bin/sh 

SRC_DIR=/tmp/android-api

print_syntax() {
   echo "Syntax:\n\t$(basename $0) [option]"
   echo "\nOptions:"
   echo "\t-l prints available versions"
   echo "\t-v builds jar file with sources of specified version"
   echo "\t-c cleans up the sources from the temp directory\n"
   exit 1
}

error() {
   echo "Error:" $1
   exit 1
}

download_sources() {
    if [[ ! -d $SRC_DIR ]]; then
        git clone git://git.source.android.com/platform/frameworks/base $SRC_DIR
    fi
}

if ! which git > /dev/null ; then
    error "git not installed (or not in the PATH)"
fi

if ! which jar > /dev/null ; then
    error "jar not installed (or not in the PATH)"
fi

if [[ "$1" == "-l" ]]; then
    download_sources
    cd $SRC_DIR
    git tag -l
    exit 0;
fi

if [[ "$1" == "-c" ]]; then
    rm -rf $SRC_DIR
    exit 0;
fi

if [[ "$1" == "-v"  && ! -z "$2"  ]]; then
    JAR_FILE=$(pwd)/$2-src.jar
    download_sources
    cd $SRC_DIR
    if ! git tag -l | grep $2 ; then
        echo "Version \"$2\" not found"
        exit 1;
    fi
    git checkout $2
    touch $JAR_FILE
    find . -depth 2 -name "java" -type d -exec jar uf $JAR_FILE -C {} . \;
    exit 0
fi

print_syntax

You can download it here: android-src-build.sh.gz. (It's a bash script: download and execute it at your own risk!)

The script, is able to download the sources from the Android git repository, and, given the version number it packs it into a jar file, that can be lately be unpacked into the sources folder of the Android SDK, where indicated above.

When you run it without parameters it gives you following help:

luigi@hal9000:~$ android-src-build.sh 
Syntax:
        android-src-build.sh [option]

Options:
        -l prints available versions
        -v builds jar file with sources of specified version
        -c cleans up the sources from the temp directory

The -l parameter prints the list of available versions. If the sources have not been already downloaded, the -l will also download them. So the first time, it will take some minutes before displaying the available releases:

luigi@hal9000:~$ android-src-build.sh -l
Initialized empty Git repository in /private/tmp/android-api/.git/
remote: Counting objects: 106183, done.
remote: Compressing objects: 100% (35592/35592), done.
remote: Total 106183 (delta 61793), reused 104978 (delta 60893)
Receiving objects: 100% (106183/106183), 128.07 MiB | 1.52 MiB/s, done.
Resolving deltas: 100% (61793/61793), done.
Checking out files: 100% (8431/8431), done.
android-1.0
android-1.5
android-1.5r2
android-1.5r3
android-1.5r4
android-1.6_r1
android-1.6_r1.1
android-1.6_r1.2
android-1.6_r1.3
android-1.6_r1.4
android-1.6_r1.5
android-1.6_r2
android-2.0.1_r1
android-2.0_r1
android-2.1_r1
android-2.1_r2
android-2.1_r2.1p
android-2.1_r2.1p2
android-2.1_r2.1s
android-sdk-1.5-pre
android-sdk-1.5_r1
android-sdk-1.5_r3
android-sdk-1.6-docs_r1
android-sdk-1.6_r1
android-sdk-1.6_r2
android-sdk-2.0.1-docs_r1
android-sdk-2.0.1_r1
android-sdk-2.0_r1
android-sdk-2.1_r1
android-sdk-tools_r2
android-sdk-tools_r3
android-sdk-tools_r4
android-sdk-tools_r5

Now, let's say that we are developing on Android 1.5, we may want to create a jar file containing all the java sources for that version with:

luigi@hal9000:~$ android-src-build.sh -v android-sdk-1.5_r3
android-sdk-1.5_r3
Note: moving to 'android-sdk-1.5_r3' which isn't a local branch
If you want to create a new branch from this checkout, you may do so
(now or later) by using -b with the checkout command again. Example:
  git checkout -b 
HEAD is now at b69bed1... AI 150122: Merge from donut.   Fix permissions issue for uninstall of updated system applications.   If an existing update for a system application is uninstalled, when reverting back to the existing   version in system partition, permissions have to be granted again.   BUG=1893639
luigi@hal9000:~$ ls -l *.jar
-rw-r--r--  1 luigi  staff  4766477 May 17 15:59 android-sdk-1.5_r3-src.jar

As shown above, after issuing the command a file called android-sdk-1.5_r3-src.jar will be generated into the current directory.
Now the only thing remaining is to expand this file into the folder ~/opt/android-sdk-mac_86/platforms/android-3/sources:

luigi@hal9000:~$ rm -rf /Users/luigi/opt/android-sdk-mac_86/platforms/android-3/sources/
luigi@hal9000:~$ mkdir ~/opt/android-sdk-mac_86/platforms/android-3/sources
luigi@hal9000:~$ cd ~/opt/android-sdk-mac_86/platforms/android-3/sources
luigi@hal9000:~/opt/android-sdk-mac_86/platforms/android-3/sources$ jar xf ~/android-sdk-1.5_r3-src.jar

At this point, restarting eclipse should be enough to have the ADT linked with the source files.

About the platforms available into the Android SDK, under the platform directory, there are following folders: android-3, android-4, android-6, and adroid-7.
I am not 100% sure, but they should correspond to the source code identified respectively by following table:

android-3 = android-sdk-1.5_r3
android-4 = android-sdk-1.6_r2
android-6 = android-sdk-2.0.1_r1
android-7 = android-sdk-2.1_r1

Finally, if you want to remove the sources downloaded you can run the script with the -c (clean) option: it will remove the sources stored temporarily into the folder /tmp/android-api:

luigi@hal9000:~$ android-src-build.sh -c
(no output returned)

Alternatively, you may want to remove the files manually.

Now, after a refresh (or a restart) from Eclipse, the sources from Android classes should look much better:

Check the referenced articles for further information on how to download the android sources, and how the ADT plugin for Eclipse locates the sources for the Android SDK.

I really think that Google should release the sources nicely packed as per the android.jar and make sure they are properly referenced by the ADT. Since then, this is my way.

For questions, mistakes, etc. leave a comment.

References

- Browsing Android Source in Eclipse
- Android sources

It is always a good thing to have a second monitor, for example to read documentation while doing some development, or for listening to music or watching videos while doing things that do not require full attention.
I have an ancient (10 years old) Dell Inspiron 8600 laptop on which the keyboard and the touchpad are starting to lose efficiency, but still it has a great display and an unrivaled audio system. The keyboard is pretty different from the macbook I am using daily, and dealing with different keyboards may not be so enjoyable. Also being an old generation laptop, the keyboard it's not very pleasant to the touch, and sometime it loses some key press.

With Synergy, it is possible to control remote computers with a single keyboard and mouse. So now, I can work on my MacBook running OSX Snow Leopard and control also the second laptop, the ancient Dell on my desk running Ubuntu 10.4, without switching keyboard or mouse. A second thing that comes for free with Synergy is the clipboard sharing; this means that you can cut-and-paste across the computers you are using: it is very helpful.
You can find more information about synergy at the official site.

First of all, you need to install synergy executable files on all the computers you need to control. Download it from here. Then, ensure that the computer can see each others on the network, using ip address or dns names. If they can ping each others, usually it should be enough.
On ubuntu you can install synergy using apt-get, while on the mac you can download the executables packed into a tarball and unpack them in a any folder you like, in my case I used ~/opt/synergy-1.3.1.

I want to use my mac's keyboard and mouse to control the dell laptop. You can have a single controller computer and many controlled computers. The controller will need the synergy server, while the controlled computers will run the client.
So on my macbook I created a configuration file which describes computers I want to control and what edge of the screen is the "border" from which the mouse will jump on the other computers. I placed it under ~/etc/synergy.conf :

section: screens
    dell:
    macbook:
end

section: links
    dell:
        left = macbook
    macbook:
        right = dell
end

The above configuration describe my screens configuration and the links between the laptops. Basicallly I am telling that I have two screens, the dell one and the macbook. And for the dell I have a macbook on the left, and for the macbook I have the dell on the right. There are many other options, but here I am just using the defaults. The full documentation on the configuration options can be found here.

Then I have to start the Synergy server on the macbook using the command:

$ synergys --config ~/etc/synergy.conf

On the dell computer then I need to run the client, who will connect to the macbook (the controller laptop) and act consequently. To do so, I can connect the client launching one of the following commands

// supposing 192.168.1.123 is the ip of my macbook
$ synergyc 192.168.1.123

// in my network I can refer the server as macbook.local or just macbook.
$ synergyc macbook.local
// or
$ synergyc macbook

After doing so, you should be able to move your mouse from one computer to the other(s) just passing through the edges of the screens, and control all the computers on your desk.

If something goes wrong, you can use the -f option to run the client and the server in foreground, and the -d to enable debug tracing.

If you still have troubles with command line tools, you may want to try QSynergy, which is a graphical user interface over synergy.

Yet it is missing some "file sharing" feature; it would be very good if synergy was able to copy files dragging and dropping files from a desk to another, but at the moment, this is missing. Installing OpenSSH can do what is missing: file transfers (scp) and remote ssh logins should be enough to complete synergy desktop integration.

Here there is a video that visually explains what you can do with Synergy:

It looks like in the video she forgot to start the client; btw it is just needed to launch the synergyc executable as I explained above.

And here, a guy shared his mouse and keyboard over 7 computers using synergy:

Have fun.

Below I explain how to make it work.

Disable the USB storage

When the USB stick is behaving as USB storage it won't work as 3G modem. The first thing to do is to disable that.

Locate the device assigned to the USB stick:

luigi@hal9000:~$ df
Filesystem           1K-blocks      Used Available Use% Mounted on
/dev/sda3            199024992 113153688  75761392  60% /
udev                   2018164       296   2017868   1% /dev
none                   2018164       340   2017824   1% /dev/shm
none                   2018164       324   2017840   1% /var/run
none                   2018164         0   2018164   0% /var/lock
none                   2018164         0   2018164   0% /lib/init/rw
/dev/sr1                 26330     26330         0 100% /media/3Connect

The 3Connect one is the USB stick, so let's unmount it, to be able - later - to unload the usb_storage kernel module.

luigi@hal9000:~$ sudo eject /dev/sr1
[sudo] password for luigi: *****

now let's unload the usb_storage kernel module.

luigi@hal9000:~$ lsmod | grep usb_storage
usb_storage            66016  0
luigi@hal9000:~$ sudo rmmod usb_storage
luigi@hal9000:~$ lsmod | grep usb_storage

The first command (lsmod) checks that usb_storage is loaded, and see if there are other modules depending on it. There aren't, so we can remove the module (rmmod command). Last command (lsmod) is to verify that the module is successfully unloaded; if it doesn't display anything means that the usb_storage module is gone, so it's ok!

Load the usbserial kernel module

Now we can load the usbserial kernel module and we can associate that to the USB stick.

First of all, let's check how the module is loaded:

luigi@hal9000:~$ lsmod | grep usbserial
usbserial              43248  1 option

We see that there's the "option" module depending on the usbserial, so we need to first unload option then usbserial.
Let's do that with rmmod command.

luigi@hal9000:~$ sudo rmmod option usbserial
luigi@hal9000:~$ lsmod | grep usbserial

The lsmod at the end is just to ensure that usbserial is successfully unloaded. This time it should not display anything.

Now we need to know the manufacturer id and the product id of the usb device. Let's do that with lsusb command:

luigi@hal9000:~$ lsusb
Bus 004 Device 003: ID 05ac:8213 Apple, Inc.
Bus 004 Device 002: ID 0a5c:4500 Broadcom Corp.
Bus 004 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 001 Device 002: ID 05ac:8507 Apple, Inc.
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 003 Device 003: ID 05ac:0237 Apple, Inc.
Bus 003 Device 002: ID 05ac:8242 Apple, Inc.
Bus 003 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 002 Device 006: ID 19d2:0031 ONDA Communication S.p.A.
Bus 002 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub

Our device is the 19d2:0031 ONDA Communication S.p.A. I found that after some tries, pluggin and unplugging the USB stick and checking the differences printed by lsmod. Notice that if the usb_storage module is active the product id will be different, while the manufacturer id will be the same. The output of lsmod is not very informative and helping to understand the nature of the usb device, unfortunately.

Now we can load the usbserial device specifying the vendor id and product id of the USB stick.

luigi@hal9000:~$ sudo modprobe usbserial vendor=0x19d2 product=0x0031

If there are no error displayed, and the command ends silently, then it means that the module is loaded successfully.

Configuring the 3G connection

Clicking on the NetworkManager we should now be able to see the New Mobile BroadBand (GSM) Connection:

Click on it to configure. It will result in a Wizard dialog that will ask you information to "Set up a Mobile Broadband Connection". In my case I filled following information:
Country: I selected "Britain (UK)" ... I took some time to find it: I was looking for "UK" or "United Kingdom" or "England" or some other name...
Select your provider form a list: 3
Select your plan: Internet
Confirm.

After that, the connection is ready and you should be able to browse the internet. Clicking on the NetworkManager tray icon you'll see that you are now connected through it:

Improving the connection reliability

The connection is quite bad through the 3 network. Sometime you keep being connected but internet stops working. I thought to DNS problem so I set up OpenDNS following these instructions, and now it seems more reliable and usable, but still sometime the connection stops working, and you need to unplug the stick and restart from with the procedure described above.
I don't know why... if somebody knows, please leave a comment.

I'd like to buy a 3G USB Stick or 3G modem; if somebody have some hints on a good supported device, and stable, (easier to set up), please leave a comment.

For the guys who do not know about iTunes and iPod/iPhone foolishnesses, the problem is that when connecting trough USB they are not exposing the filesystem as usb disk. The second problem is how iTunes memorizes files: they get renamed in 4 char lenght filenames under a 3 char lenght folder... so you have files like:

iTunes_Control/Music/F26/UBFF.mp3
iTunes_Control/Music/F26/UEBM.mp3
iTunes_Control/Music/F26/UIPQ.mp3
iTunes_Control/Music/F26/UWCJ.mp3
iTunes_Control/Music/F26/WGOH.m4a

As result of this, if you have songs belonging to different authors and coming from different albums, but you was keeping in the same folder to listen together... you can't specify to iTunes to play a folder as a playlist, because folder and filenames gets chaotically "reorganized" when imported in iTunes. iTunes, then uses id3 tags to expose views to the user, grouping by author or album title, etc.

To connect to iPod filesystem with Linux, you need to jailbreak the iPhone or the iPod touch, and have OpenSSH running on it. It is not very difficult, but, afaik it's the only way to access remotely to the iPod filesystem, through network (wifi).

Once done that, you need to mount the filesystem using ssh protocol; this can be done with sshfs, which is the ssh protocol support for FUSE (Filesystem in USEr space) which allows you to mount several type of local or remote filesystem without needing root privileges.

$ sudo apt-get install sshfs

Then I created a folder under my ~/Music folder where to mount my iPod media

$ mkdir ~/Music/iPod

At last I made a bash script to mount and unmount the iPod media to the folder I just created. This is the content:

#! /bin/bash

case "$1" in
mount)
        echo -n foobar | sshfs root@Luigi-R-Viggianos-iPod.local:/private/var/mobile/Media/ ~/Music/iPod -o password_stdin
	;;

umount)
        fusermount -u -z ~/Music/iPod
	;;

*)
	echo "Usage: `basename $0` {mount|umount}"
	exit 1
	;;
esac

In the above script "foobar" is the password to use to ssh-connect as root on my ipod. Unfortunately I did not succeed in using public key authentication, even modifying the sshd configuration on the ipod, I do not know why... So I had to specify the password in the script.
The Luigi-R-Viggianos-iPod.local is the hostname used by my iPod. Alternatively you can use the ip address, but when you change network you will need to update the script; the hostname usually gets recognized in different WIFIs and mapped to corresponding ip address.

To get the ip address of the iPod connected to WIFI, there are some programs downloadable on the apple store, one is called, with much of fantasy, "IP Address". To know the hostname of the ipod, just connect to it through ssh, and you will see that displayed on the bash prompt line, in my case: "Luigi-R-Viggianos-iPod:~ root#", unfortunately apples likes to have weird unix system on their devices so you don't have basic commands like 'hostname' or files like /etc/hostname to check... did I already said that I hate OSX...?

At this point what I do is just running the script:

$ ipod mount
or
$ ipod umount

to mount or unmount the iPod to ~/Music/iPod folder. The iPod needs to be attached to a power source, the laptop trough usb or to a docking station, otherwise it will disable the WIFI connection when it goes on stand by; but the file access is fully wireless.

To play files, you need to launch Amarok or Rhytmbox and import the folder ~/Music/iPod. I use Rhytmbox and it indexes the folder. When the folder is mounted I can see the music on my iPod, and when it is not, the music just disappears.

If you want to export an album of MP3 files from the iPod, you just need to select files from Rhytmbox and drag and drop into Nautilus (filesystem browser). But you will have the weird names for the MP3 files set by iTunes. To batch rename the files you can install and use id3ren, which lets you to customize (-template option) how the filename should be renamed (using album title, song title, author, etc):

$ sudo apt-get install id3ren
$ id3ren -template='%a - %t - %s.mp3' *

If you like to backup the music on your iPod just use rsync or cp. But I don't know how to import new songs in the weird filesystem-format used by iTunes... For this, possibly the easiest solution is to use iTunes on Windows or OSX using VMWare, until Apple does not decide to release iTunes on Linux.

Have fun.

$ sudo apt-get install cryptsetup

When you plug in your USB disk, and mount it, you'll see an icon on your desktop. Right-clicking on it and selecting "Format" will let you to specify an encrypted filesystem type:

Clicking to "Format" button will ask you to type the password for the encrypted volume:

At this point the format will proceed and, after, the USB drive will be mounted with an open lock icon:

If you want to change the password it should be possible from the "Disk Utility" (System>Administration>Disk Utility, or type "palimpsest" at the command line), but it seems to be buggy, as it always gives me "Incorrect Passphrase. Try again".

Command line tools always helps in those cases.

First step, add the new password for the volume:

$ sudo cryptsetup luksAddKey /dev/sdb
[sudo] password for luigi: (my system admin password)
Enter any LUKS passphrase: (any password registered to this volume)
key slot 0 unlocked.
Enter new passphrase for key slot: (the new password)
Verify passphrase: (the new password again)
Command successful.

Second step, dump all the key password for the volume:

$ sudo cryptsetup luksDump /dev/sdb
LUKS header information for /dev/sdb

Version:       	1
Cipher name:   	aes
Cipher mode:   	cbc-essiv:sha256
Hash spec:     	sha1
Payload offset:	1032
MK bits:       	128
MK digest:     	c8 97 18 80 0c 0a 86 ed 8f 3c 85 03 e1 de de 2d 68 ed 70 a0
MK salt:       	16 ae 5b 05 2c 2b 02 d5 af 0d 71 d7 08 ba 51 fd
               	9f 98 cd 11 52 e8 14 44 71 4f 84 53 99 02 97 c7
MK iterations: 	10
UUID:          	c4ada688-3cae-4053-a1c6-781614ad683f

Key Slot 0: ENABLED
	Iterations:         	447466
	Salt:               	63 30 36 1f 87 83 f6 73 75 e9 a2 b2 dc f1 30 4c
	                      	09 67 1d e7 82 71 35 6c c4 df ce 10 0e 3b 42 2f
	Key material offset:	8
	AF stripes:            	4000
Key Slot 1: ENABLED
	Iterations:         	451004
	Salt:               	55 c4 d0 4f e9 24 d8 c0 2a cb b2 7f 09 a1 80 98
	                      	4f 4d 1a 5e 6f 1c d5 ad c2 30 a3 02 15 9d 1e bd
	Key material offset:	136
	AF stripes:            	4000
Key Slot 2: DISABLED
Key Slot 3: DISABLED
Key Slot 4: DISABLED
Key Slot 5: DISABLED
Key Slot 6: DISABLED
Key Slot 7: DISABLED

Third step: remove the old password:

$ sudo cryptsetup luksKillSlot /dev/sdb 0
Enter any remaining LUKS passphrase:
key slot 1 verified.
Command successful.

Unmount the volume and try the new password. It should work.

Of course it would be better to have the "change password" feature in the context menu of the mounted volume, or at least having the feature working in the "Disk Utility" program. But, you know, it's Linux, and you need to hack it sometimes :)

It seems also that LUKS is supported on Windows, see here.

Another good, portable, free, and more complete solution to encrypted filesystems is the great TrueCrypt.

For the experts, a useful command-line HOWTO on using LUKS: http://ubuntuforums.org/showthread.php?t=404346.