HOWTO: Griffin Powermate in The GIMP

This Christmas, I was graciously given a slick Griffin Powermate from my brother.


It's a nifty little aluminum knob for your PC, with an extremely smooth touch and a blue LED that makes the bottom glow. There's support in the Linux kernel for the device, and it acts like a regular old input device. If you're crafty, you can edit your xorg.conf to make it control X, or you can use other software's built-in support for the device. Both the free DJ software Mixxx and The Gimp support the device. Mixxx's support needs a bit of work, but the device works great in The Gimp. However, there doesn't seem to be any documentation on how to actually set up the Powermate in The Gimp, so here we go:

1. Edit your ~/.gimp-2.2/controllerrc, and add the following to the end of the file:

(GimpControllerInfo "MIDI"
(enabled yes)
(debug-events yes)
(controller "ControllerLinuxInput"
(device "/dev/input/event2"))
(mapping
(map "button-0" "select-all")
(map "dial-turn-left" "context-tool-select-previous")
(map "dial-turn-right" "context-tool-select-next")))


If you have more than just a keyboard and mouse hooked up to your PC, you might need to change the "/dev/input/event2" to a different device (/dev/input/event3 or something).

2. Set up the permissions
for the Griffin Powermate device (run this in a terminal):

sudo chmod a+r /dev/input/event2

Again, replacing the device with the proper one if your's is different. (Also, see the note at the end of the HOWTO about setting this at bootup.)

3. Fire up The Gimp
and change the mappings in the preferences as you see fit:



4. That's it, enjoy!

There's probably a good way of setting the proper permissions (like in step 2) by default at bootup, but I can't figure it out. If anyone does figure it out, drop me a comment. In the meantime, use your favourite hack to run the command in step 2 at bootup ("/etc/rc.local" is a good enough place). :)

PDF + Cube = PDF Cube?



PDF Cube uses the OpenGL API to add spinning cube page transitions to PDF documents

PDF Cube is an OpenGL API-based PDF viewer that adds a compiz/Keynote-like spinning cube trasition effect to your PDF presentations (including Latex, Beamer and Prosper). You can also zoom on 5 predefined areas of any presentation page with a smooth zooming effect.


That pretty much says it all. It's still pretty alpha-quality, but it might be part of a cool solution to the presentation software dilemma some day. If you're interested, check it out.

GPL kernel modules, reverse DRM, and the future of Linux

I was doing my daily read of Digg today, and I stumbled across this: Linus Torvalds on the "GPL only modules" debate.

The issue at hand is whether the kernel should only be allowed to load GPLed modules (that is, drivers). That's right, some of the Linux kernel developers want to do away completely with binary-only (ie. closed source/proprietary) drivers, like the (good) Nvidia and ATI ones.

On the one hand, Andrew Morton writes:
Give people 12 months warning (time to work out what they're going to do,
talk with the legal dept, etc) then make the kernel load only GPL-tagged
modules.

I think I'd favour that. It would aid those people who are trying to
obtain device specs, and who are persuading organisations to GPL their drivers.

(Whereas the patch which is proposed in this thread hinders those people)


The patch in question would allow more components of drivers to be handled outside of the kernel, thus allowing developers of binary-only drivers (like Nvidia) to write GPLed skeleton drivers that actually just move their proprietary stuff outside of the kernel. So really, it doesn't solve anything, as Linus himself argues:

It will only result in _exactly_ the crap we were just trying to avoid,
namely stupid "shell game" drivers that don't actually help anything at
all, and move code into user space instead.

What was the point again?

Was the point to alienate people by showing how we're less about the
technology than about licenses?

Was the point to show that we think we can extend our reach past derived
work boundaries by just saying so?

The silly thing is, the people who tend to push most for this are the
exact SAME people who say that the RIAA etc should not be able to tell
people what to do with the music copyrights that they own, and that the
DMCA is bad because it puts technical limits over the rights expressly
granted by copyright law.

Doesn't anybody else see that as being hypocritical?


Linus then goes on to argue that decisions in the kernel should be based on technical merit, and shouldn't seek to limit how people are allowed to write/use drivers for Linux.

Because the history of Linux is so deeply intertwined with that of the Open Source/Free Software movement, many people feel that the kernel should only accept GPLed modules. The idea is that this would put pressure on companies like Nvidia and ATI to release GPLed ("Free") drivers. However, the flip-side is that it may well just end up in the same companies simply shunning Linux all together, or just moving their proprietary code out of the kernel. (The only thing the latter case seems to accomplish is a philosophical cleansing of the kernel.) Furthermore, proponents of this approach seem to be ignoring the other important component of "Free" - that is, being able to do what you like with it.


A "Free" kernel should let the user insert any modules they want in it, regardless of the license. It's almost reverse DRM - Instead of restricting users from freely using something, users would instead be restricted to only using something "Freely" (that is, only in ways that are compatible with their definition of "free"). Lastly, it just so happens that allowing binary kernel modules (as is the case currently) is the best technical decision, at least in the short-term, because Linux does not have the consumer base to persuade vendors to release open source drivers. Think about it - Linux is most popular in server market right now, and all of the popular binary drivers (Nvidia, Ati, Wifi) are needed on desktop Linux, not on servers.

Moving forward, the Linux kernel must continue to walk the narrowing line that balances between commercial success and philosophy. The next few years are certainly going to be interesting, particularly if ATI and/or Nvidia open source their drivers. Keep on the look-out.

HOWTO: Banshee 0.11.3 on Ubuntu

Update: Banshee 0.13.1 can be installed in Ubuntu 7.10 (Gutsy Gibbon) by running "sudo apt-get install banshee".



Banshee 0.11.3 was released today and features a bunch small new features and bug fixes, but also includes some pretty big performance enhancements. If you're a Banshee user, it's definitely worth upgrading to this new release for the speed improvements alone (switching to your library doesn't take a ridiculously long time anymore.) Since I've already written a HOWTO for Banshee 0.11, I decided to updated it for Banshee 0.11.3. This guide was written for Ubuntu 6.10/Edgy Eft, but there's only one little change necessary for it to work on Ubuntu 6.06/Dapper Drake (which is explained inline below.) On with the installation instructions:

:)

IMPORTANT NOTE:
If you have the QuinnStorm repositories enabled (for Compiz/XGL stuff), you might encounter this compile error: "/bin/grep: can't read /usr/lib/libXrender.la: No such file or directory" or something along those lines. The necessary fix can be found here. (I just ended up removing the "/usr/lib/libXrender.la" part of that line and it fixed it, and I think that's probably a safer route.)

To follow this HOWTO, just punch (ie. copy and paste) the commands listed into a terminal. Good luck!

1. Install prerequisites

First, make sure you have the universe repository enabled. If you're unsure, here's instructions on how to check and enable it.
Next, install the software required to build Banshee:

sudo apt-get build-dep banshee
sudo apt-get install libavahi-cil mono libgconf2.0-cil
sudo apt-get build-dep libipoddevice0
sudo apt-get install libgtop2-7 libgtop2-common libgtop2-dev libsgutils1 libsgutils1-dev
wget \http://banshee-project.org/files/libipoddevice/libipoddevice-0.5.2.tar.gz
wget http://banshee-project.org/files/ipod-sharp/ipod-sharp-0.6.2.tar.gz
tar -xvzf libipoddevice-0.5.2.tar.gz
tar -xvzf ipod-sharp-0.6.2.tar.gz


Note: If you're an Ubuntu 6.06/Dapper Drake user, the above "sudo apt-get install..." line might fail. If it does, try running the following:
sudo apt-get install libsgutils libsgutils-dev
(Note: If this fails, just keep going to with the rest of the HOWTO...)

Now, in order to have iPod support, we're going to install libipoddevice and ipod-sharp:

cd libipoddevice-0.5.2
./configure --prefix=/usr
make
sudo make install
cd ..

cd ipod-sharp-0.6.2
./configure --prefix=/usr --disable-docs
make
sudo make install

(The above iPod steps can be safely left out if you don't need iPod support...)

2. Download Banshee 0.11.3

wget http://banshee-project.org/files/banshee/banshee-0.11.3.tar.gz
wget http://www.banshee-project.org/files/banshee-official\
-plugins/banshee-official-plugins-0.11.3.tar.gz

3. Extract and configure

tar -xvzf banshee-0.11.3.tar.gz
tar -xvzf banshee-official-plugins-0.11.3.tar.gz
cd banshee-0.11.3
./configure --prefix=/usr --enable-avahi --disable-docs

I suggest leaving Avahi enabled here as I did so that DAAP sharing works. (It lets you share your music library with iTunes, Limewire, etc. users, as well as listen to other peoples'.) iPod support should be automatically detected if you followed the iPod steps above.

4. Build and install Banshee

make
sudo make install


5. Configure, build, and install the plugins

cd banshee-official-plugins-0.11.3
./configure --prefix=/usr
make
sudo make install


6. Run Banshee!

Either from the console run "banshee" or launch it from the "Applications->Sound & Video" menu in GNOME.


That's it! I've tested this on Ubuntu 6.10/Edgy Eft machine, but if this doesn't work for you, leave a comment and I can try to help you figure it out. :)

Lastly:
Digg!

3D Album Art in Banshee

Hot on the tail of Apple's Coverflow, Lukasz Wisniewski is developing a plugin for Banshee called Fleow:



That is hot. (Thanks Gabriel!)

Ecksdee 0.0.9 Release and Developer Interview

For those of you who remember my Next-Gen Linux Game Roundup, you might recall Ecksdee as one of the games that was highlighted. In the months that have passed, the developers of Ecksdee have been hard at work tweaking and adding new features to the game, which has started to evolve into a playable fast-paced racer. A few days ago, the team released Ecksdee 0.0.9, which features the slick new Ciy map (video) along with some new ships.

With this new release, I had the opportunity to ask two of the Ecksdee developers, Vincent Knecht (vknecht) and Amir Taaki (genjix), a few questions:

Q. What was your original motivation for creating Ecksdee?

VK: I had free time when Amir came up with his hovercraft prototype, and thought it deserved a game project, like XRacer. At that time I was making a visualization application with Crystal Space, and wanted to do something more
fun with CS...

AT: I found myself working with the Crystal Space community, and on a under development game called Crystal Core. Since it was set in the future I imagined some type of hovering vehicle would be useful and began work on a CEL property class for hovering (basically like a kernel module that can be loaded in your game to do stuff). To go with it I made a demo to demonstrate it.

Since Vincent is an active Crystal Space developer, he managed to catch wind of it and started asking questions about the status of it - "Is it a game?", for which I had no ideas about :D Needless to say a week later I got a tarball back with a build system and proper application architecture... This was a great blank canvas to start working from since now everything was properly organised as it should and from there that was real good kick starter. I doubt I ever would've gone at this on my own otherwise.

Nonetheless after some time, the game started to play but looked crap -there were no ships, a poorly designed main menu... Luckily Pascal (who is also a Crystal Core artist), made us all the ships (we've taken some out for this release), the Ciy level and other textures and special effects.

Q. How has your decision to use the Crystal Space engine impacted the development of Ecksdee?

VK: Well, I've be following Crystal Space development and learning it since a few years. When we started, Crystal Core (technical demo for CS) was one of the best source of inspiration to kickstart Ecksdee. So I think Ecksdee wouldn't exist without CS :-)

AT: Very much, all of us come from the Crystal Space community. Using CEL is the biggest help since it simplifies many worked concepts in game development.



Q. What about Python?

VK: Currently, we're using Crystal Entity Layer XML scripting, which is great for simple things. However, we will switch to CEL python scripting to get more control and possibilities. Hopefuly, it will be easier to read and modify too.

Q. You're one of the only open source games that uses Nvidia's Cg Toolkit. How has this enhanced Ecksdee's graphics?

VK: Cg support is actually a Crystal Space feature, so all projects using CS use Cg to some extent. So far, we don't have really specific shaders, and use the ones provided with CS. Afaik, Cg is about shader portability, and is supported by CS since a few years.



Q. Can you tell us more about the Blender integration you're working on? How will it make content development easier?

VK: Basically, it's about having a straightforward way to get Blender models and maps running in the game. Blender and blender2crystal export script are the recommended tools for Ecksdee content creation. There are a couple of HOWTOs in our wiki about that.

AT: Hopefully Ecksdee will become more and more easier to modify in the future. I am also a blender2crystal developer and want to make it easier for people to quickly build levels and ships to make a huge load of quality content rather than developing a game like a traditional closed team.

At the moment you can set a few properties in blender, and click "Run" in the b2cs overlay and just start playing your level. With ships you can make a ship, drag it to a specific directory and the game will automatically show it in the menu.

For those who are interested I encourage them to check out the Blender conference video about Crystal Space, the blender2crystal page and the artists documentation section of our site.

Q. Lastly, what's on the roadmap for Ecksdee in the future?

VK: First there will be some porting from XMLScript to Python, especially for menus and head-up-display handling. Sure, there will be new tracks and ships. We also will look at networking and multiplayer support, though we cannot say now when that will be added.

AT: The addition of python will allow some interesting things like little drones following your ship (as a weapon system) and easier extending of the game for new features and options to the main menu among other things.

~

To wrap up, I'd like to thank Vince and Amir for the time they've both graciously spent on Ecksdee and this interview. Ecksdee is in good hands, and I'm looking forward to the future of the project. Finally, if anyone would like to get involved with Ecksdee (be it artists or coders), the Ecksdee documentation is a good place to start.

Links:

Show Ballmer the Money



"Mr. Ballmer, Since you live in a fantasy world, thought we'd pay some fantasy money for your fantasy claim."


I think Tristan Sloughter might be on to something.

If you haven't been following the whole Novell-Microsoft deal, Novell's customers basically get protection from being sued by Microsoft. Microsoft believes that the Linux kernel infringes intellectual property it owns (heard this before?), and that Linux users are liable.

That's right, Microsoft wants you to think it's going to sue you for using Linux. (It's yet to be shown if they actually have a case or not. Many Linux advocates believe they don't.)

In either case, I find it ironic that Novell, who takes a strong stance against proprietary drivers in their kernels, admits that it thinks there's "proprietary" intellectual property in heart of it's Linux distribution, SUSE. That doesn't make very much sense, does it?



Little associated with Microsoft seems to make sense these days...

We need better presentation software


Let's face it: OpenOffice.org Impress is a lame, boring presentation program. What if you want to do something really fancy? Something Keynote fancy?

Well, if you're a Linux user, you're out of luck. OpenOffice.org Impress is the best you can do, but it's not for a lack of other people trying. A few years ago, GNOME's Agnubis was a potential candidate for a some new presentation software, but it unfortunately never took off. Even more recently, Criawips (aka AbiShow) seemed to show some promise, but never ended up making it very far. Currently, the project looks dead.

So in 2006, if you're going to make a presentation on Linux, it's going to look like it's from 1995. If you're a serious developer looking for a new project, the libraries seem to have fallen into place since Criawips in order to make this a viable project:
  • Cairo provides a nice SVG 2D graphics canvas.
  • GStreamer (which has finally reached maturity) makes it easier to handle embedded audio and video.
  • OpenGL (which has been there all along) is also a fairly straightforward graphics API (can do 2D as well as 3D), and can be used to do funky 3D transitions and such easily.
  • Hell, even SDL would be a good choice for the graphics API.
  • If you're feeling rather C#, Mono's Tao provides bindings for both OpenGL and SDL. I can't vouch for how strong GStreamer# is, but it at least seems to exist.
Maybe someone can fill me in as to why nobody's created the killer open source presentation application yet. The libraries are there to handle the fancy stuff, and the simple stuff (allowing creation of text-boxes and insertion of images) should be straightforward to do.

Discuss.


Digg!

Edgy and Beyond

It's been a few days since Ubuntu 6.10/Edgy Eft was released, and you haven't upgraded already, here's a few reasons to:
So what are you waiting for?
Upgrade to Edgy now, or if you still haven't given Ubuntu a shot, now would be a good time. You won't regret it.

If you're interested in finding out what's on the drawing board for the next Ubuntu, version 7.04/Feisty Fawn, here's some things we might see:
If there's one thing that's clear to me from running Ubuntu 6.10 for over a month now, it's that Ubuntu is still getting better and seems to be making great strides toward becoming the most usable operating system on the planet. I can't imagine where we'll be in three years.

Speed updates...

Some happenings from the Linux world over the past few days:
  • Banshee 0.11.1 was released. Check out what's new in on the release page.
  • The Ubuntu 6.10/Edgy Eft RC (Release Candidate) came out today. (Essentially Beta+1). More release notes on how things are shaping up.
  • A beta of Flash 9 for Linux was finally released. Even more release notes if you're interested. Long story short: A/V sync is fixed.
  • Ubuntu 7.04 = Feisty Fawn. From Mark Shuttleworth himself:
    The main themes for feature development in this release will be improvements to hardware support in the laptop, desktop and high-end server market, and aggressive adoption of emerging desktop technologies. Ubuntu's Feisty release will put the spotlight on multimedia enablement and desktop effects. We expect this to be a very gratifying release for both users and developers. Detailed planning will take place at the developer summit next month in Mountain View, California. Please join us there to help shape the Feisty Fawn!

  • EasyUbuntu, Automatix, and now.... AfterBirth. *sigh* I wish the Ubuntu team would just give me a button to push that installed w32codecs. I might add that w32codecs is no more illegal than any of those Windows codec packs like the Nimo one, which can be pretty handy, and nobody seems to be getting in any hot water for (yet).

NVIDIA Root Explot

The following says it all:
   The NVIDIA Binary Graphics Driver for Linux is vulnerable to a
buffer overflow that allows an attacker to run arbitrary code as
root. This bug can be exploited both locally or remotely (via
a remote X client or an X client which visits a malicious web page).
A working proof-of-concept root exploit is included with this
advisory.

Yeah, so apparently we don't need Internet Explorer and Windows anymore to have malicious software silently installed on our computers - we just need NVIDIA's closed-source graphics driver. Update: Fortunately, the bug has been fixed in NVIDIA's 1.0-9625 beta driver. The thing is, you need Xorg 7.1 to run that, so everyone running Ubuntu 6.06/Dapper Drake is still vulnerable.

On that note - are there any analogous driver exploits in Windows like this? I didn't even think something like this was possible in Linux...

(Thanks Hubert...)

Portland, TC:E, GWoot, Automatix 2, Mute

First off, it's been quite a busy couple of weeks for me and I'd like to apologize for the lack of news. Luckily, the world doesn't revolve around me and so some interesting stuff happened in the last few weeks:

  • The Portland project made their first release. The Portland project is a kind of vague teaming up of various desktop environment developers with the aim of making it easier for software vendors to create Linux software that worked consistently across the various desktop environments (KDE, GNOME, etc.) out there. The first release consists of a daemon that applications can interface with through DBUS and some command line tools to perform various tasks like installing an application's icon or disabling the screensaver. The way a programmer would normally go about doing stuff like this is specific to each desktop environment (in general), and so Portland's DAPI solves that problem by offering a single unified way of going about that. I don't think there's anything in Portland right now that would convince Adobe to release a supportable version of Photoshop for Linux, but making companies like Adobe's lives easier is what Portland aims to do. I think it's a small step in the right direction, but once it reaches critical mass, it'll become a crucial part of the Linux desktop.
  • True Combat: Elite 0.49 was finally released after a long period of waiting. TC:E is a realistic combat mod for Enemy Territory that's similar to Counter-Strike. However, TC:E has plenty of unique features like the iron-sights system that make it a fun, memorable play. Several of the maps have been updated in this version, as well as new ones being added. Additionally, several of the weapons were tweaked and there are some new snazzy graphics effects as well.
True Combat: Elite 0.49

  • For any Woot.com fans, someone created GWoot, which might be handy.
  • Automatix 2 for Ubuntu 6.10/Edgy Eft has been released. This thread has some extra screenshots of it, and it seems like it's come quite a ways from the cheap bash script it once was. Good job Automatix team!

    Automatix 2 for Ubuntu 6.10 looks much better than the original


  • The under-appreciated Linux Rock Star blog uncovered Mute, which is a new rewrite of the incredibly powerful Jeskola Buzz. Buzz is a Windows-only modular synth "tracker-on-steroids" whose source code was lost in 2000, but continued to be developed through clever add-on hacks. The Buzz community has been waiting for something to replace it for a while, and it looks like Mute is a good contender. (Buzztard is another one to get excited about too.) While you're at it, check out the Mute screenshots - it's a pretty impressive clone of Buzz. (I currently use Buzz under WINE for music production as a hobby, but unfortunately it has some issues.)

    Mute's user interface is almost identical to Buzz

Also, I upgraded to Edgy last week. If I told you it was easy to do, I'd be lying. Long story short: Wait until it's officially released unless you're prepared to file bug reports. It's working great now though and I've been able to play with some brand new software that wouldn't run in Dapper. I'm enjoying the built-in AIGLX now too - No more hacks to make OpenGL games run with Compiz/Beryl. :)

Freespire's sketchiness and a word on DistroWatch

Linspire's sister distribution Freespire is reportedly using sketchy tactics to inflate their rank at DistroWatch... again.
I have to admit that I've seen the advertisement in question on Linux Revolution, but the URL pointed to the Freespire wiki.



DistroWatch's ranking index provides a rough sketch of the relative popularity of almost all of the various Linux distributions out there. Why would you want to be near the top? DistroWatch's 50,000 hits/day seems like a decent reason (it makes my blog look like, well, a blog). After keeping an eye on it for many years, it seems like at least the top three distributions are probably correct in their popularity with desktop users. The index is also a good way to find out what the next big Linux distribution is going to be (openSUSE has jumped quite a ways recently).

Firefox Copyright Dilemma

"I used to be cool"

There's been quite a bit of noise in the Linux community lately over this 'bug' in Debian. That's right, the Mozilla Firefox name goes under the same crappy copyright license as the artwork, so Debian isn't "allowed" to use the name. (This is why Ubuntu and Debian ship Firefox with that uninspiring blue globe icon instead of the Firefox icon.)

So what's Ubuntu going to do about it?
Well, the users clearly want the official logo in addition to keeping the name, and I don't blame them. Firefox made the transition from Windows to Linux quite a bit easier for myself, and I'm sure others have found the same thing. The matter is still up in the air, as the only official word on this has been been from Canonical's Matt Zimmerman saying he's "discussing the relevant issues with representatives from Mozilla."
In the meantime, I was shocked to discover that those crazy GNU people are now maintaining "Gnuzilla" and "IceWeasel", the latter of which I originally thought was a joke.

I think I can honestly say that I wish IceWeasel was joke, because this whole Mozilla trademark thing is ridiculous. Aren't we all playing on the same team?

OT: Fake USB Drives - I was right

In the last week, three people have come forward and posted comments on my fake USB stick article saying they're had similar experiences at Factory Direct (the retailer I bought it at). If there was any doubt in my mind that it was really false advertising and not just a faulty USB stick, this cleared it up.

The weird thing about this is that the guys who branded the USB keys in California probably didn't even know they were fake. My theory is that the original manufacturer sold these to various companies who branded them and resold them. One (sketchy) guy ended up counterfeiting them by branding the Sony logo on them, and that's why there's all these fake Sony ones around that look exactly like the one I had. Other companies, like the one who branded mine got screwed because they bought the fake keys and didn't know it until it was too late. (Do they even know it now?)

If anyone else has had a similar experience at Factory Direct, let me know here, or if you want to do something about it, contact me at linux*NoSpam*revolution0@yahoo.ca, and we'll see what our options are.

Skype 1.3 (Linux) Released

Skype 1.3 for Linux has finally been released, now with ALSA support. Still no video support, but I'll take what I can get. (More details over here)

If you've been using the 1.3 betas, not a whole lot has changed with the official release, but if you're using an older version, you'll be quite thrilled with the new version. There's a detailed changelog available for anyone who's interested in that as well.

People always rant about Skype and say "use OpenWengo instead", but the problem is that nobody uses OpenWengo (relatively speaking), so who am I going to talk to with it? The fact that it's open source means nothing if the software is useless due to a lack of users. (Plus, if you live in North America, Skype's free calling to the U.S. and Canada can't be beat.) Yeah, I might be playing devil's advocate here, but for what it's worth, Skype is a pretty good application.

Random News (Beryl release and video, Edgy Beta)

Here's some interesting recent happenings:
  • MacSlow's got a video of Edgy running Beryl, showing off blurred transparent windows.
  • On that note, the first version of Beryl (0.1.0) was released (yeah, it's still lacking a website). If you've already got the QuinnStorm repositories enabled in Ubuntu, a "sudo apt-get install beryl emerald emerald-themes" should do it for you. (Launch it with beryl-manager.) If you don't have XGL or anything set up yet, I suggest waiting until Ubuntu 6.10 comes out at the end of this month, as it'll be much easier to set up then. Beryl does feel a bit snappier than the old QuinnStorm compiz though and things seem less buggy so far. (The nice thing about Beryl is that Quinn's team can now make their own releases and stabilize them beforehand. No more buggy development compiz.)
  • Ok, last bit of Beryl news for today, I promise! Linux.com has an interview with QuinnStorm on the new release and the divergence from Compiz. It doesn't contain a whole lot of new information, but there is one gem tucked away at the end:
    "I would tell a layperson to look forward to a desktop that can really outshine what both other major players in the field offer, especially once X gets its input redirection code in, but even before that we'll be able to really catch some attention."

    The article doesn't explain what X's "input redirection code" is, but it rang a bell with me. I've read that in Compiz (and now Beryl), when you click on a wobbling window, your click won't "land" on the right spot. It turns out this is actually a limitation in the X server, so I'm pretty sure Quinn was referring to the solution to this problem. I'm sure this'll open up some interesting ways to interact with windows though. (Also, my memory is terrible, so someone correct me if I'm wrong on this one...)
  • The beta of Ubuntu 6.10 Edgy Eft has been released. The official Ubuntu release notes details the big noticeable changes and includes some screenshots as well.

Gaim 2.0 Beta 4 coming soon

I just spotted on the Gaim development blog that a fourth beta of Gaim 2.0 is coming soon. The blog has some screenshots of the new features that are worth taking a peek at (global buddy icon integrated into UI, different indentation). Check them out!

If you're still running Gaim 1.5, I strongly suggest trying out the new Gaim 2.0, even though it's still in beta. The UI overhaul makes it much easier to use - no more nagging "away" screen.

Lastly, Planet Gaim has launched in order to make it easier for avid fans to keep track of Gaim's development. Good stuff.

WINE 1.0 in early 2007?

The latest issue of the WINE Weekly Newsletter was released last week and has some interesting details about where WINE's headed.

For those not familiar with WINE, it's a software project that allows Micrsoft Windows applications to run under Linux (and other *NIX OSes). It's been in development for about 13 years now and is the result of a tremendous effort by many people. Over the past few years, WINE's compatibility has been improving at an impressive rate, and I consider it to be at that pivotal point where you can download a random Windows app from the internet and expect it to install and actually run (a far cry from the situation three years ago.) Watching a Windows application install significantly faster in Linux than it does in Windows is a sight to behold. :)

Propellerhead's Reason is an example of a complex Windows application that has only recently started working in WINE (thanks WINE team!)



Going back to the newsletter, check out these juicy tidbits about WINE 1.0:

As far as 1.0 goes, there's a lot of things that would be nice to have completed. The default registry needs a bit of help, the IDL compiler (widl) needs to be more feature complete, and Win64 would benefit from printf format fixes. Copy protection would be really nice to get in the tree and Ivan Leo Puoti reiterated that he had patches available. Alexandre responded that there are some critical parts of the design that need to be fleshed out because "the Wine maintainer is a pain in the ass." Finally, Alexandre called for more help with packaging. Most developers don't use the packages and they really could use more eyes with more testing done.

Despite that, a 1.0 release seems to be near (and yes, I should know better than to write such things.) The current plan is to continue working on Direct3D and get it stabilized. After that, a code freeze will begin and the plan is to have it last about 2 months. The target date for that is approximately the end of this year. Alexandre mentioned we have some projects going on right now that are rolling along nicely and it's probably worth letting those continue rather than lose momentum.

Post-1.0, a stable branch will be maintained and new development will eventually begin on a development branch. Using git will help a lot with this and fixes can be cherry-picked back in the stable branch. As far as version numbering goes, there's no clear plan for how it will work, although Jeremy White jokingly suggested "Wine 2007" for the release instead of 1.0.



There's also some stuff on WINE's Direct3D support:



Currently the rendering code is now shared between all versions of Direct3D from version 1 through version 9. There's been a huge improvement in shader code with shaders implemented with GLSL and the GL_ARB_*_program extensions. There's support for up to shader model 3.0.

But what would a Direct3D presentation be without some eye candy? Stefan showed off screenshots of some games. There was also a small contingent of DirectX folks in attendance with some really high-powered laptops that could show off the games. It's quite impressive to see the latest and greatest games running on Linux. Jon Parshall extensively, um, "tested" World of Warcraft throughout the conference (did you finally make it to level 48, Jon?) Tom Wickline had 3DMark2000, 3DMark2001SE and 3DMark2003 running all of there test. There is still some artifacts in the rendering of a couple of the test, but the DirectX guys knew what was to blame for it. Stefan showed off the Microsoft DirectX logo "proving" DirectX is being properly detected.


But perhaps the most interesting bit is this:


Direct3D10, which will ship with Windows Vista in a few months, doesn't seem to be a large cause for concern. At first glance it appears to be more of an evolutionary change rather than revolutionary. New shader support will be needed, but extending ours once OpenGL supports it should be pretty easy. Stefan mentioned Microsoft is currently offering a lot of incentives for Windows developers who develop D3D10-only games since they'll only be usable on Vista - there's no plan to backport D3D10 to XP. Dan Kegel asked if that means we should port Wine's forthcoming D3D10 implementation to Windows, which would be relatively easy when we switch to WGL.

(emphasis added)

That's just a really cool idea, and something that I think people'd appreciate.

If you're interested in other upcoming changes in WINE or would like a peek into the WINE developers' world, take a gander at the full issue here.

Dear Lazyweb: Backing up Email in Evolution

Dear Lazyweb,

My university generously provides us with a whopping 25 megabytes of email storage, and has been sending me messages on a daily basis that my inbox is 93% full.

I use Evolution as my mail client, and it's an IMAP mail server.
How can I download my 23 megs of emails and a stick them in a tarball?

If anyone has any ideas or hints on this, I'd appreciate it very much if you dropped me a comment. :)

Thanks!

OSS Usability and Linux: What?

I found an interesting article in the Digg queue today titled, "Windows & OSS: The Usability Problem".

After giving it a read, I realized that I just don't agree with most of the author's points. Allow me to unleash the hounds:

Lack of standardized user interface

Users of open source operating systems are spoilt for choice: Gnome, KDE and Xfce only to name a few desktops and Blackbox, WindowMaker, AfterStep, FluxBox, fvwm and mwm just to name a few of available window managers. Yes, diversity is generally a good thing, but consider how confused an average Windows user must feel when all the programs look and behave differently among different desktops.


Well, I've got some ground-breaking news for you: It's been a long time since I've used a distribution of Linux that actually asked me which desktop environment I wanted to use. I think it's safe to say that most Linux users use either KDE or GNOME [1].

On all my systems, except on the Solaris boxes, I am running KDE. Why? Because it lets me concentrate on getting the work done and does not bother me with trivial tasks like mounting and unmounting devices and the like.


I've got some more news here: That's why everyone runs GNOME too. Both desktop environments offer a complete set of good applications for daily activities that are easy to use.

No, I am not using Gnome because of the poor GTK architecture and the lack of basic stability. On all my systems, from Laptops to workstations, Gnome did not perform well at all.


Poor GTK architecture? Lack of "basic stability"? Since the author carefully failed to provide evidence to support his argument, I'm going to ask all the GNOME/Ubuntu users out there: Have you experienced a lack of "basic stability"? Perhaps someone can point me to the source of this basic instability?

Diversity among Linux kernels

While the author does make some valid points here, I think many people view the Linux kernel the wrong way. Vendor specific kernels allow vendors to do better Q&A testing in order to ensure their distro works as stably as they'd like, while also letting them add new features in-between kernel releases. For example, the Ubuntu kernel often has features backported from the next "unstable" kernel release, to the benefit of the users. As well, new features and bugfixes that the Ubuntu team finds are sent back "upstream" to the official Linux kernel [2], so that other distributions can benefit from them as well. That's how open source is done these days. Essentially, new features and bugfixes are only vendor-specific until they're sent upstream and included in the next official Linux kernel release.

Hardware known to work on one system does not work on the other due to missing drivers or modules


Let me ask the reader the obvious question: If your hardware works in Ubuntu, would you expect it to work on OS X? So is it fair to expect that it would also work in Fedora Core? Fedora is a different operating system from Ubuntu.
I'll finish this thought below...

Effectively no software and hardware certification standards

The author again makes valid points here, but I'd like to mention my own thoughts:
As far as I can tell, a piece of software that's included in the default GNOME desktop environment is as close to being "certified software" as possible. It's guaranteed to be stable, have a consistent/friendly UI, and in general, be useful. I'd rather have my apps be GNOME-certified and Ubuntu-certified rather than LSB-certified. See where I'm heading with this?

Poor stability of many user programs

The days of horribly unstable Linux apps are mostly gone, and that's entirely due to the quality standards that have been set by distributions like Ubuntu and Fedora. It's up to the distributions to make sure they package stable software, and the major ones do!
Also, that big paragraph about Sally installing an RPM doesn't really apply to apt-based distributions like Debian and Ubuntu. (If the package wasn't already in a repository, then a properly created package downloaded from the web wouldn't have a problem.) :P

I'm not even going to start on the documentation stuff. (Actually, I lied: GNOME software is documented well.)

My Conclusion

Over the last few years, I feel there's become a greater distinction between "Linux" and Linux distributions. What exactly is Linux these days?
When I use Ubuntu, I'm using a Linux kernel enhanced by the Ubuntu team. When I browse the internet, I'm using GNOME and Firefox, both of which were tweaked by the Ubuntu team. The Ubuntu team didn't choose all of this software and tweak it accordingly just to be different: They did it to tackle the usability problem and the standardized user interface problem. They did it to provide a rock-solid kernel with the best hardware compatibility out there, to ensure users have stable software, and to ensure that software is quality software. Did I mention they do it every six months too?

The article in question is another example of an article that would have held its ground three years ago, but my, how things have changed.


[1] Desktop Linux 2006 Survey Results
[2] Search the 2.6.18 kernel changelog for the word "ubuntu" to see what I mean.

Goodbye XGL, Goodbye AIGLX (Hello Xorg 7.1?)

Something awesome happened today.

NVIDIA released a beta of their new 1.0-9625 drivers. I'll just skip to the good part, as NVIDIA's James Jones explained it:

Neither Xgl or AIGLX are required to use compiz with the NVIDIA drivers now that they natively support GLX_EXT_texture_from_pixmap.

Xgl is an X server that renders using OpenGL and runs on top of another X server. It was the first X server available to support GLX_EXT_texture_from_pixmap.

AIGLX stands for Accelerated Indirect GLX. It is not related to compiz or GLX_EXT_texture_from_pixmap at all, except that support for GLX_EXT_texture_from_pixmap in the open source DRI OpenGL drivers required it. NVIDIA has always supported Accelerated Indirect GLX rendering.

NVIDIA supports both Direct AND Indirect rendering with the GLX_EXT_texture_from_pixmap extension. Users should not need to install any additional software to run compiz with new NVIDIA drivers. Please see the just-created sticky thread covering the basic setup steps.

That's right. These drivers feature hardware support for the mystical "GLX_EXT_texture_from_pixmap" extension, which now allows your videocard to accelerate your wobbly windows, among other things.

The catch? They require Xorg 7.1 or later. That means Ubuntu 6.06/Dapper Drake users are out of luck, unless they feel like compiling X from source (not such a good idea.) The good news is that Ubuntu 6.10/Edgy Eft features Xorg 7.1, so NVIDIA users will be able to benefit from this when Edgy is released (or if they're running Edgy already).

If you're using Xorg 7.1 and want to give it a shot, here's the sticky with instructions. (Also, this feedback thread might come in handy.)

Your move, ATI.

Update: A reader has pointed our that AIGLX has been merged into Xorg 7.1, and it's now enabled by default. So while it's not quite "goodbye" forever to AIGLX, you don't actually have to do anything manually to get it (and you shouldn't have to worry about it).

GNOME 3.0 Mockup: "May-B"

There's an interesting mockup of GNOME 3 over on GNOME-Look.org that someone did:


The mockup looks like a combination of Gimmie, (U)SLAB, and Mathusalem (elements which should probably be merged into GNOME sooner or later.) The tags/files panel on the right looks it could use some work (layout, size, fonts, etc.), but the overall concept seems like it's pointed in the right direction.

On a related note, the new DesktopThoughts blog (who's author created the mockup) aims at discussing ideas for "the next desktop". If you're interested in seeing what people come up with or you've got some ideas of your own, be sure to give it a visit.

Now, we just need to make sure that the GNOME logo doesn't evolve into some scary, evil-looking thing like the one in the background of that desktop...

Edgy CDs, NVIDIA preview, Neuros OSD, Linux 2.6.18

Lots of Linux news happening around the web today:

HOWTO: Banshee 0.11 + Ubuntu

Dec 9/06: Banshee 0.11.3 has been released! Updated HOWTO here.
Updated Sept 26, 2006 (Added iPod support stuff)
Updated October 16, 2006 (Minor fixes)


With the Banshee 0.11 hot off the press, and no .DEBs for Ubuntu 6.06/Dapper Drake in sight, I figured people would probably appreciate some instructions on how to install the new Banshee. Well, here we go:


IMPORTANT NOTE:
If you have the QuinnStorm repositories enabled (for Compiz/XGL stuff), you might encounter this compile error: "/bin/grep: can't read /usr/lib/libXrender.la: No such file or directory" or something along those lines. The necessary fix can be found here. (I just ended up removing the "/usr/lib/libXrender.la" part of that line and it fixed it, and I think that's probably a safer route.)
Ubuntu 6.10/Edgy Eft Users: Updated Banshee packages will probably hit the Edgy repositories, so just hold tight for a bit and hopefully an updated package will get pushed through the usual Ubuntu update notifier.

To follow this HOWTO, just punch (ie. copy and paste) the commands listed into a terminal. Good luck!

1. Install prerequisites

First, make sure you have the universe repository enabled. If you're unsure, here's instructions on how to check and enable it.
Next, install the software required to build Banshee:

sudo apt-get build-dep banshee
sudo apt-get install libavahi-cil mono
sudo apt-get build-dep libipoddevice0
sudo apt-get install libgtop2 libgtop2-dev libsgutils libsgutils-dev
wget http://banshee-project.org/files/libipoddevice/libipoddevice-0.5.0.tar.gz
wget http://banshee-project.org/files/ipod-sharp/ipod-sharp-0.6.2.tar.gz
tar -xvzf libipoddevice-0.5.0.tar.gz
tar -xvzf ipod-sharp-0.6.2.tar.gz

Now, in order to have iPod support, we're going to install libipoddevice and ipod-sharp:

cd libipoddevice-0.5.0
./configure --prefix=/usr
make
sudo make install
cd ..

cd ipod-sharp-0.6.2
./configure --prefix=/usr --disable-docs
make
sudo make install

(The above iPod steps can be safely left out if you don't need iPod support...)

2. Download Banshee 0.11

wget http://banshee-project.org/files/banshee/banshee-0.11.0.tar.gz
wget http://www.banshee-project.org/files/banshee-official\
-plugins/banshee-official-plugins-0.11.0.tar.gz

3. Extract and configure

tar -xvzf banshee-0.11.0.tar.gz
tar -xvzf banshee-official-plugins-0.11.0.tar.gz
cd banshee-0.11.0
./configure --prefix=/usr --enable-avahi --disable-docs

I suggest leaving avahi enabled here as I did so that DAAP sharing works. (It lets you share your music library with iTunes, Limewire, etc. users, as well as listen to other peoples'.) iPod support should be automatically detected if you followed the iPod steps above.

4. Build and install Banshee

make
sudo make install


5. Configure, build, and install the plugins

cd banshee-official-plugins-0.11.0
./configure --prefix=/usr
make
sudo make install


6. Run Banshee!

Either from the console run "banshee" or launch it from the "Applications->Sound & Video" menu in GNOME.

That's it! I've tested this on an "almost" fresh-install Ubuntu 6.06/Dapper Drake machine, but if this doesn't work for you, leave a comment and I can try to help you figure it out. :)

Digg!

Banshee 0.11 Released!


It's been a while since the last Banshee release, but the wait was well worth it. Among the new features in Banshee 0.11 are better tagging support (including writing for all the readable formats), the ability to import a selection or all the music from a digital audio player into Banshee, one-click track rating, an improved preferences dialog, and a new song-change notification bubble (using libnotify, so they should have a somewhat consistent look with the rest of the desktop).

A song-change notification



There's also three new official plugins:
  • Recommendations - This uses the awesome Last.fm service to recommend you different artists based on what you're listening to. The plugin also tells you some interesting stats about the artist you're currently listening to.

  • Podcast - Banshee now handles podcasts in all their glory. The interface is very smoothly integrated into Banshee and is easy to use. As of the beta, I noticed that you couldn't drag-and-drop a podcast onto a digital audio player, but I haven't tested it yet in the final release.


  • Downloading a podcast while playing music



  • Mini-mode - Sick of Banshee taking up a sizable chunk of screen space? This new plugin allows you to flip Banshee into "mini-mode", which provides a compact player interface instead of the usual library view. The mini-mode interface is really well designed, check it out!

The new mini-mode view



As I said before in my Banshee 0.11 preview, this is the Banshee release I think everyone's been waiting for. Banshee is now a fully featured music playback and management application, and has quickly become a shining star in the world of open source software. It's fast development, great polish, and intuitive interface make it a great example of open source done right. Is it perfect? Well, it's still a little sluggish, but I think it's well worth the trade-off for the excellent feature-set.

Update: I've posted a some instructions on how to install Banshee 0.11 in Ubuntu 6.06. Enjoy! :)

Compiz gets forked real good: Beryl


It was time. QuinnStorm (who maintained her own tree of compiz and provides awesome Ubuntu packages) has forked off Compiz into a new project, Beryl.

It's been clear for a while that QuinnStorm's version of Compiz has diverged from the simplicity present in the first release of Compiz. Quinn's tree has included all sorts of community developed plugins, most of which add shameless bling with little contribution to enhancing usability.

That's the difference as I see it: Novell's Compiz will continue to be a stable compromise between bling and usability, and Quinn's Beryl will become the community-driven bling-machine. And that's not to say that Beryl won't be stable, it's just that Novell puts more effort (and justifiably at that) into creating a top-notch user experience - Something that comes across as being an afterthought in Quinn's tree of Compiz.

At the very least, the next six months are going to bring us the answers to two important questions this fork has posed:
1. Will Beryl plugins remain compatible with Compiz? (mikedee from the Compiz forums asked this one)
2. When other distributions like Ubuntu add some sort of standard AIGLX/XGL support, will they offer Compiz or Beryl as the window manager of choice?

I honestly don't know the answers to either of these questions, but as development progresses, it'll certainly be interesting to find out. (In the meantime, follow this thread.)

Why I love GNOME (part 1?)

If someone asked me, "What's the most underrated feature in GNOME?", I'd undoubtedly show them this:



Did you see that?

All I had to do to extract that archive was right-click on it, and hit "Extract here". That's it. I didn't have to open an application and choose where I wanted to extract the archive to or anything. (It's probably the feature I miss the most when I use Windows or OS X...)

Oh yeah, and I forgot the coolest part: If the archive you're extracting contains a bunch of files that aren't grouped in a parent folder, it knows to create a new directory to throw them in (so you won't end up with files littered all over your desktop.) Neato.

What do you think is the most underrated feature in Linux, GNOME, or whatever desktop environment you use?

Ubuntu Deja Vu...

Get GNU/Linux and Linux.org

There's an interesting post on Digg that hasn't quite made the front page yet:

Is Linux.org hurting Linux?


If you've ever visited Linux.org, you'll probably have an opinion on this.
My opinion?
Absolutely.

The majority of the people who've taken the poll on the Ubuntu Forums seem to agree.

Is there something better?
Absolutely.

It's clean, simple, eye-catching, and the information that the community wants to send to potential users is easy to get to. While Get GNU/Linux does have some typos and some GNU-isms (I think the general public might not get the free beer/free speech thing), it's certainly a great start. I think the official Ubuntu site in particular does the best job I've seen at explaining Linux and the concept of open development:

Ubuntu is a complete Linux-based operating system, freely available with both community and professional support. It is developed by a large community and we invite you to participate too!

The Ubuntu community is built on the ideas enshrined in the Ubuntu Philosophy: that software should be available free of charge, that software tools should be usable by people in their local language and despite any disabilities, and that people should have the freedom to customise and alter their software in whatever way they see fit.

This makes more sense to me than Get GNU/Linux's:

Gnu/Linux, or simply Linux, is an alternative to Microsoft Windows�. It is easy to use and gives more freedom to users. Anyone can install it: Linux is free as beer and as speech.

The last thing I'd like to throw into the mix here is that the GNU label is just plain ugly. The general public doesn't need to know what GNU is (nor will they care). Like I mentioned in the comments of the Digg article, there's a reason it's just "OS X" and not "GNU/OS X" [I stand corrected... While OS X does ship with some GNU utilities, it's apparently not warranted to call it "GNU/OS X" since it isn't 'free'. The Wikipedia article on GNU seems to agree with me at one point though: That GNU/Linux meant GNU utils + Linux kernel... (hence GNU/OS X, by my initial logic)]


(And if I may be so bold, look at the Ubuntu site again. How many times is the distribution referred to as "Ubuntu" rather than "Ubuntu Linux"? Simply put, anyone who doesn't know that Ubuntu is Linux won't care that is, and anyone who does care already knows.)

Release: MythTV 0.20

After 7 months of hard work, a new version of MythTV (0.20) has finally been released. This release includes support for a bunch of new pieces of hardware, more DVB stuff, and some other cool stuff like OpenGL accelerated menus.


I've been using a copy of MythTV 0.20 from SVN for a couple of weeks, and it seems rock solid so far. My favourite features of the new version have to be the OpenGL menus (which now smoothly fade in/out to one another) and the new "Internal" player. The internal player can be used in MythDVD and MythVideo to play DVDs, DivX/XviD, and MPEG videos. The internal player even features support for DVD menus, which is pretty kickass. There's also two big advantages in using the internal player over something like MPlayer for video/DVD watching:
  1. You don't have to add to your LIRC configuration file (.lircrc?) and set up an extra set of remote button bindings for the extra program (ie. MPlayer)
  2. The OSD that shows up when you pause, adjust volume, etc. is the same now as it is when you're watching TV in MythTV. This visual consistency adds a lot of polish to Myth.
:)

Aside from this stuff, there's tons of other features/fixes in this release, and you can read about the rest of them in the MythTV 0.20 Release Notes.

Formatting your USB stick in Linux

After all that dealing with all that fake USB key stuff, I ended up picking up a 5 gig USB stick from Z-Cyber (it was cheap). Long story short, I ended up having to format it, and I haven't been able to get it properly formatted since. I could get Linux to see the sole FAT32 partition on it, but not Windows. Oddly enough, even if I formatted it with Windows, it still wouldn't show up properly on Windows machines.

Anyways, I think I've finally found the proper way to format your USB key in Linux. This extremely helpful article walks you through it.

I think I kept making two mistakes before I found that article. The first was that I kept forgetting to actually format the partition using mkfs.vfat (fdisk just creates the partition). The other problem I had was that I forgot that /dev/sda and /dev/sda1 are different. The second one (/dev/sda1) is the "first partition" on the drive, while /dev/sda is the whole drive. (I had ended up adding accidentally creating a new partition on /dev/sda1, which made a partition in a partition or something screwy like that... I should have taken a screenshot of what Nautilus thought the drive contained... trust me, it was bad.)

Anyways, I pretty much feel like a total idiot for forgetting my partitioning basics, but I'm going to take that as a sign that desktop Linux's usability has gotten better, seeing as it's been a pretty long time since I've had to do any crazy manual partitioning like this. (Now, if only GParted worked properly on my USB key in the first place, I wouldn't have had to figure this out... :P)

Updates from the Web: GNOME 2.16, Exaile, and more

I've been pretty busy over the past few days, and I apologize for the lack of updates. In the meantime, here's some interesting articles I've spotted:
  • Ars Technica has a First Look at GNOME 2.16
  • The 2X TerminalServer seems to be getting a bit of press (ie. it popped up on digg, hardly the press though.) It's another open source (but commercial) implementation of the NoMachine server. Honestly, the NoMachine server kicks ass, so I'm not quite sure why you'd want to use this one. (It's a breeze to install in the latest version too - just 3 .DEBs)
  • Red Herring has an interview with Michael Robertson, founder of MP3.com and Linspire, where he talks about Linux, Linspire, and the music industry. I found this part of the interview particularly interesting - Way to dodge the question Michael:

    Q: And so we�ve got to ask you, you�ve recently made CNR�your software update service for Linspire�free. If I�m an Ubuntu user, am I going to be able use this in order to fill up my machine with capabilities like the ability to do DVD playback, Flash, Quicktime? Is that a possibility?

    A: If you�ve ever tried to install software on Linux, [you know] it�s really difficult. There�s no easy installer, like an XP user would be used to, and Click and Run goes beyond that, right? One click and everything is downloaded and installed, icons on the desktop, etc. So, absolutely, I think that�s something that makes a lot of sense and so we�re definitely looking at something like that.

  • CRN has an editorial I think I agree with: "Advice to Linux: Kill the Penguin" (especially the part about the acronyms thing)
  • A new version (0.2) of the Exaile Media Player has been released. It's apparently described as "Amarok for GNOME". If any long-term readers out there remember my thoughts on Amarok, they'll know that I have beef with the sideways tabs. Worst interface element ever? Quite possibly. Well, now GNOME users can indulge themselves in useless sideways tabs that hide functionality that shouldn't be hidden away. (Compare Banshee's GUI with Exaile's GUI.) Lastly, someone should also quietly tell the author of Exaile that the only good Hooverphonic album is the first one. :P
  • ... and if you made it this far, Charlie isn't the only who's just gone through a breakup. (Albeit mine was for different reasons...) :(

Linux Distros Timeline

I spotted a cool timeline over on Digg that shows when each major Linux distribution was created and was branched.
It's neat to see how recent Ubuntu really is relative to all the other distros, and to see how many forks/branches there are.

I think MythTV is trying to tell me something....

I think my MythTV box is trying to tell me something:

Desktop Linux Poll Results

DesktopLinux.com's 2006 poll results are in.

The most popular distro?
Ubuntu

The most popular desktop environment?
Read on.

There's some other interesting stats there as well. I'm suprised that Gentoo has more users than Fedora. It's nice to see Xfce doing pretty good as well (I used it for a few months a couple of years ago, before I switched over to GNOME/Ubuntu).
I'd like to think these numbers are decently accurate too, seeing as the number of votes is almost 15,000. I can't see posting a link to the poll in a distro-related forum having skewed the numbers too much...
Interesting stuff though.

What kernel would Jesus compile?

Does the bluescreen of death make you spew forth fire and brimstone, or are you the type that just says a quiet prayer? Now, God can be on your desktop, with Ubuntu: Christian Edition.

Read on at APC Magazine...
(It's funny)

Upstart - Goodbye Init

I spotted something delicious over on Planet Ubuntu recently:
upstart is a replacement for the init daemon, the process spawned by the kernel that is responsible for starting, supervising and stopping all other processes on the system.

What's so yummy about this? Well, they've got a damn good project outline, explaining why there's a need to replace the ancient init daemon, how upstart is designed, and how it's different from existing solutions such as Apple's launchd and initng. Lastly, it also explains what the state of the project is and what direction it's headed in. This is almost enough to make sure a project succeeds right there - well, that, along with a good leader and lots of free time.

While Upstart is now in Edgy's Universe repository, it's not actually going to be standard in the next Ubuntu. I'm surprised that a next-gen init system hasn't landed in a mainstream distro as of yet. (After a shallow googling, the only distro I could find with InitNG was the obscure Berry Linux.) Most of these next-gen init systems cut startup times, which would make a noticeable difference to the end user. Unfortunately, most of these new init systems just aren't quite ready for public consumption, but when they are, you can bet we'll finally see Linux boot much faster.