Wallpapers or MP3 Player?

Get the xmms-mp3 plugin, and you’re all set.

Get the xmms-mp3 plugin, and you’re all set.

As you probably know, one of the most popular things to do with gooey Linux is show off your desktop — it is also — if we Linux enthusiasts are honest with ourselves — arguably one of the only things it’s good for. :-)

Xmms is an engine, really. It’s basically just a Posix port of the best mp3 player ever made–Winamp 2.81, that is. You can look at it this way: XMMS is to audio playback as X-windows is to Graphical environments.

It should be noted than any skin that will work with Winamp 2.81 will also work perfectly with XMMS, so when you’re tired of shopping through all the alien-heads and sillyassed themes on freshmeat, check out the Winamp site.

As you probably know, one of the most popular things to do with gooey Linux is show off your desktop — it is also — if we Linux enthusiasts are honest with ourselves — arguably one of the only things it’s good for. :-)

Xmms is an engine, really. It’s basically just a Posix port of the best mp3 player ever made–Winamp 2.81, that is. You can look at it this way: XMMS is to audio playback as X-windows is to Graphical environments.

It should be noted than any skin that will work with Winamp 2.81 will also work perfectly with XMMS, so when you’re tired of shopping through all the alien-heads and sillyassed themes on freshmeat, check out the Winamp site.

The only thing I’d kind of disagree with is that my desktop at work is a GUI Linux. (FC4+KDE) I use it for everything from Cervasie (CVS frontend) to email to text editing (GVim).

I’m sure the majority of *users* are using it for the age-old Email+Web+Music type stuff, but we career people use the whole OS most of the time.

Of course, the server is a different story.

The only thing I’d kind of disagree with is that my desktop at work is a GUI Linux. (FC4+KDE) I use it for everything from Cervasie (CVS frontend) to email to text editing (GVim).

I’m sure the majority of *users* are using it for the age-old Email+Web+Music type stuff, but we career people use the whole OS most of the time.

Of course, the server is a different story.

We career people. That’s funny, I’ve had a pretty good career so far, and — not that I am a MS advocate by any stretch of the imagination, and I like certain things like Quanta, Kismet and Gaim — I think I can accomplish much the same things with TortoiseSVN/CVS, Notepad — and I can also use Dreamweaver, Photoshop, Maya, and other powerful W32 apps at the same time.

It’s a matter of personal preference, I suppose… but in my opinion, there’s nothing wrong with a proponent of OSS and the development of GUI Linux to admit that the de facto Desktop OS is also the better Desktop OS — if not for what it’s made of, for the body of applications designed to run atop it.
<div id=”Clipmarks_BorderDivLeft” style=”border: 2px solid orange; margin: 0px; padding: 0px; position: absolute; width: 0px; display: none; z-index: 99999″ />
<div id=”Clipmarks_BorderDivRight” style=”border: 2px solid orange; margin: 0px; padding: 0px; position: absolute; width: 0px; display: none; z-index: 99999″ />
<div id=”Clipmarks_BorderDivTop” style=”border: 2px solid orange; margin: 0px; padding: 0px; position: absolute; height: 0px; display: none; z-index: 99999″ />
<div id=”Clipmarks_BorderDivBottom” style=”border: 2px solid orange; margin: 0px; padding: 0px; position: absolute; height: 0px; display: none; z-index: 99999″ />

We career people. That’s funny, I’ve had a pretty good career so far, and — not that I am a MS advocate by any stretch of the imagination, and I like certain things like Quanta, Kismet and Gaim — I think I can accomplish much the same things with TortoiseSVN/CVS, Notepad — and I can also use Dreamweaver, Photoshop, Maya, and other powerful W32 apps at the same time.

It’s a matter of personal preference, I suppose… but in my opinion, there’s nothing wrong with a proponent of OSS and the development of GUI Linux to admit that the de facto Desktop OS is also the better Desktop OS — if not for what it’s made of, for the body of applications designed to run atop it.
<div id=”Clipmarks_BorderDivLeft” style=”border: 2px solid orange; margin: 0px; padding: 0px; position: absolute; width: 0px; display: none; z-index: 99999″ />
<div id=”Clipmarks_BorderDivRight” style=”border: 2px solid orange; margin: 0px; padding: 0px; position: absolute; width: 0px; display: none; z-index: 99999″ />
<div id=”Clipmarks_BorderDivTop” style=”border: 2px solid orange; margin: 0px; padding: 0px; position: absolute; height: 0px; display: none; z-index: 99999″ />
<div id=”Clipmarks_BorderDivBottom” style=”border: 2px solid orange; margin: 0px; padding: 0px; position: absolute; height: 0px; display: none; z-index: 99999″ />

Hey Aaron… what markup interpolation scripting are you using for your comments? I ask ’cause That wasn’t supposed to be a strike, but a pair of ndashes. :-)

Hey Aaron… what markup interpolation scripting are you using for your comments? I ask ’cause That wasn’t supposed to be a strike, but a pair of ndashes. :-)

bq. Josh, that’s funny… that’s the exact thought I had earlier today… that’s the only thing the GUI is good for… unless of course you work on a Solaris box where it seems like everything is geared ONLY toward a GUI install…. I can’t tell you how many days I’ve wasted in the past two weeks in McLean trying to install software after unsuccessfully remote installing from Rockville or Baltimore.

You know, one of my favorite things about FC is the Anaconda installer, which, if you know how to ask it, will let you do a network install (http,ftp,rsync or nfs) right from the mirror of your choice. Not that it’ll help you having to physically insert a bootdisk image, but not having to burn the distro and walk around with it is money, and cloning a farm from an nfs share can make your life a lot easier. Of course, I dunno if SuSe has a network method, but I never plan on using SuSe, so…

Out of desperation I’ve actually used an FC1 boot floppy to bootstrap a FC3 network install without any problems, so it’s something to consider if you have the bandwidth and a favorite fast mirror.

bq. Josh, that’s funny… that’s the exact thought I had earlier today… that’s the only thing the GUI is good for… unless of course you work on a Solaris box where it seems like everything is geared ONLY toward a GUI install…. I can’t tell you how many days I’ve wasted in the past two weeks in McLean trying to install software after unsuccessfully remote installing from Rockville or Baltimore.

You know, one of my favorite things about FC is the Anaconda installer, which, if you know how to ask it, will let you do a network install (http,ftp,rsync or nfs) right from the mirror of your choice. Not that it’ll help you having to physically insert a bootdisk image, but not having to burn the distro and walk around with it is money, and cloning a farm from an nfs share can make your life a lot easier. Of course, I dunno if SuSe has a network method, but I never plan on using SuSe, so…

Out of desperation I’ve actually used an FC1 boot floppy to bootstrap a FC3 network install without any problems, so it’s something to consider if you have the bandwidth and a favorite fast mirror.

Let’s see if I can hit all of these. The easy ones first:

Just because I run KDE doesn’t mean I can’t run both K-whatever and G-whatever apps. For the most part (obvious brain-damaged developers still statically linking against libraries that get replaced faster than yesterday’s underwear notwithstanding) you can run Gnome apps on KDE and vice-versa. Now obviously I’m going to use yum whenever possible, but this particular app came as an rpm. I installed it by hand, and it worked just fine.

Remote X…

Here we go.

Aaron: I don’t know the environment as well as I’d like, so I’ll be making some assumptions here… Let’s see how close I come.

The X subsystem is actually kind of backwards to the way we typically look at this stuff. You run the XServer on your machine, and the remote system is the client. That’s the actual true description of an X session. How so? The machine you’re on needs to run software that provides a location for the remote X machine to connect back to as a client. So, there’s a couple scenarios you’re probably experiencing.

1) You have a windows box on your end without X Server software and aren’t really sure how to proceed.

2) You have a linux box, and you’re getting the whole “permission denied” thing, and can’t figure where the security is configured.

In case #1, there is a pile of software from the very expensive to the free out there to run an X server on your windows based system. I’ve used many, and tend to like Esker’s SmartTerm X. Xoftware is another, and Cygwin is the free offering. Having one of these on your Windows box is always a good idea if you’re working in UNIX. At the very least, Cygwin can foot this bill.

In case #2 you have a situation where you simply haven’t enabled access. We’ll cover that later.

Once you have an XServer involved, you have to configure thecliennt to know how to reach you. It needs to know how to send a display to you, the server. We do this by exporting our display address. (I’ll use private addresses for example)

Solaris: 192.168.1.15
You: 192.168.1.27

while connected to the remote system (via ssh, presumably), you would execute a command telling that system where you are like so:

export DISPLAY=192.168.1.27:0.0

or, if your shell doesn’t support that

DISPLAY=192.168.1.27:0.0; export DISPLAY

In doing this, you’ve told your client “I’m over here”. This works on both windows-based and linux-based systems.

Next, there is a security system associated with X sessions. It’s very simple, but you still need to execute the command to allow X connections. That security system is called “xhost”. Xhost is used to add granular (and massive) rights to a system. It’s usage is like so:

xhost [[+-]name …]

Now, the simple running of the command: xhost + will completely open up your system to any X connection from anywhere on the wire. However, you can also man the command out to see some options, and the addition of users to an ACL and the like. This is all part of the XSecurity subsystem, which also has some pretty detailed man pages.

Finally, once you’ve run your X server, run the xhost command to allow access, you then SSH to the machine in question (you must tunnel X over SSH…this is pretty clear in SSH’s config files). Once on the destination machine, you can type a known X command (like xclock or xterm. This would include any programs you could run under X.

Another way to connect to remote boxes is VNC. VNC is supported on most platforms now, and is an excellent way of doing what you’re asking. It’s by far the best performance, and you actually see the desktop you’re trying to control instead of just the command prompt, and you needing to launch the various programs from thaqt location.

If youhave any more questions, I’d be happy to answer. It’s late, and this feels like it was somewhhat hard to follow, but I’m not really feeling well.

Let me know if I can be of more help.

Xhost Man: http://www.netadmintools.com/html/xhost.man.html

Let’s see if I can hit all of these. The easy ones first:

Just because I run KDE doesn’t mean I can’t run both K-whatever and G-whatever apps. For the most part (obvious brain-damaged developers still statically linking against libraries that get replaced faster than yesterday’s underwear notwithstanding) you can run Gnome apps on KDE and vice-versa. Now obviously I’m going to use yum whenever possible, but this particular app came as an rpm. I installed it by hand, and it worked just fine.

Remote X…

Here we go.

Aaron: I don’t know the environment as well as I’d like, so I’ll be making some assumptions here… Let’s see how close I come.

The X subsystem is actually kind of backwards to the way we typically look at this stuff. You run the XServer on your machine, and the remote system is the client. That’s the actual true description of an X session. How so? The machine you’re on needs to run software that provides a location for the remote X machine to connect back to as a client. So, there’s a couple scenarios you’re probably experiencing.

1) You have a windows box on your end without X Server software and aren’t really sure how to proceed.

2) You have a linux box, and you’re getting the whole “permission denied” thing, and can’t figure where the security is configured.

In case #1, there is a pile of software from the very expensive to the free out there to run an X server on your windows based system. I’ve used many, and tend to like Esker’s SmartTerm X. Xoftware is another, and Cygwin is the free offering. Having one of these on your Windows box is always a good idea if you’re working in UNIX. At the very least, Cygwin can foot this bill.

In case #2 you have a situation where you simply haven’t enabled access. We’ll cover that later.

Once you have an XServer involved, you have to configure thecliennt to know how to reach you. It needs to know how to send a display to you, the server. We do this by exporting our display address. (I’ll use private addresses for example)

Solaris: 192.168.1.15
You: 192.168.1.27

while connected to the remote system (via ssh, presumably), you would execute a command telling that system where you are like so:

export DISPLAY=192.168.1.27:0.0

or, if your shell doesn’t support that

DISPLAY=192.168.1.27:0.0; export DISPLAY

In doing this, you’ve told your client “I’m over here”. This works on both windows-based and linux-based systems.

Next, there is a security system associated with X sessions. It’s very simple, but you still need to execute the command to allow X connections. That security system is called “xhost”. Xhost is used to add granular (and massive) rights to a system. It’s usage is like so:

xhost [[+-]name …]

Now, the simple running of the command: xhost + will completely open up your system to any X connection from anywhere on the wire. However, you can also man the command out to see some options, and the addition of users to an ACL and the like. This is all part of the XSecurity subsystem, which also has some pretty detailed man pages.

Finally, once you’ve run your X server, run the xhost command to allow access, you then SSH to the machine in question (you must tunnel X over SSH…this is pretty clear in SSH’s config files). Once on the destination machine, you can type a known X command (like xclock or xterm. This would include any programs you could run under X.

Another way to connect to remote boxes is VNC. VNC is supported on most platforms now, and is an excellent way of doing what you’re asking. It’s by far the best performance, and you actually see the desktop you’re trying to control instead of just the command prompt, and you needing to launch the various programs from thaqt location.

If youhave any more questions, I’d be happy to answer. It’s late, and this feels like it was somewhhat hard to follow, but I’m not really feeling well.

Let me know if I can be of more help.

Xhost Man: http://www.netadmintools.com/html/xhost.man.html

Well, I posted a really long answer to Aaron here last night, but I can’t see it. What gives?

Well, I posted a really long answer to Aaron here last night, but I can’t see it. What gives?