I like bleeding-edge technology. The newer, the better, as there are usually fun little things to discover and rarely does anything bleeding-edge actually work perfectly, giving me ample opportunity to mess around with things.

Enter Fedora 12. I’ve played with the initial Alpha release, several of the Snap releases, and as of this morning, the Beta release. So far, at least as far as my MSI Wind 123 goes, they’re all unusable.

The biggest problem so far is the display – as soon as GDM kicks in, prior to the login screen loading, the screen starts to ‘pulse’. To be more specific, if you were to tap the ‘Brightness Up’ and ‘Brightness Down’ keys fairly quickly between two different levels, you’d get the same effect. This continues incessantly, and makes it impossible to use the GUI. If I boot in runlevel 3 (direct-to-console), everything is fine, so the issue is restricted to Xorg.

As far as the beta goes, there’s been a little regression. In the Alpha and Snap releases, I was able to get all the way to the desktop, despite the pulsing display. Now, with the beta, it takes just over 2 minutes just to get past the boot animation, and then it stalls before the login screen loads. Whether or not this has anything to do with the fact this is a livecd on a USB stick (made with the Fedora live-usbcreator for Windows, with persistent overlay), I’m not sure. I’ll probably download another spin (maybe the KDE version) and see if I have any more luck.

Update: Progess! On a whim, I removed rhgb from the kernel line when booting (press Tab to cancel automatic boot, and then again to edit boot options) and was actually able to get to the desktop, albeit with no improvement to the speed of things. The screen still pulses, however I did notice something interesting: while the screen pulses (briefly before the login screen appears, then it stops until you choose a user, and then it starts again), if I switch to another terminal (CTRL+ALT+F2, for example), the login name starts to fill with “^@” repeated several times, and then stops. If I then switch back to the GDM terminal (CTRL+ALT+F1), the pulsing stops.

At least that gives me something to go on.

Further Update (Dec 12, 2010): The Wind U123 works perfectly on Fedora 14 with no additional hacks/changes needed.

And while it may be a step in the write direction, there are a few little things that just bug me about the current iteration of the Moblin netbook distribution.

  • Power Button – Where is it? The short answer is there isn’t one – your only option is to simply use the physical power button on your netbook. Also, the shutdown terminal command is strangely absent. Thankfully, though, the reboot command is still there. It would be excellent if Shutdown/Reboot/Sleep were included in the options presented when you click the battery status icon.
  • myzone – There’s no way to customize the myzone tab, which is the default location on start up. You get three columns, and no control of them or choice in adding other quick content. Not very handy for a start page, in my opinion.
  • Persistence – The detailed instructions page that tells you how to put Moblin on a flash drive/SD card/whatever doesn’t explain how to add a persistence layer so that your changes survive a reboot. All of the instructions/utilities supplied simply bite-copy the image to your device and that’s that. Supposedly you can use the Fedora LiveUSB Creator to do this, but I’ve yet to try it. Having an official utility would help, and something that I hope to see coming down the road.

So for the time being, Moblin is something that’s neat to play around in – it has an amazing interface that’s refreshing and definitely easy to use for Linux (or computer-in-general) newbies – but still has some growing to do before I’d use it in place of a full distribution.

With any luck, they’ll work on these things for the final version of Moblin 2, or if not that, than hopefully for Moblin 3.

Edit: Also useful information, the root password for Moblin 2 is ‘moblin’. And for some reason, although sudo is installed by default, the standard user (also ‘moblin’) isn’t in the /etc/sudoers file, so you can’t use sudo unless you manually add it. That needs to get fixed as well.

I recently picked up an MSI Wind U123, and have been doing my usual try-every-operating-system-I-can-find-and-see-what-happens tests. Along the way, I’ve tried Windows XP (was installed by MSI, lasted a few minutes before I got rid of it), Windows 7 (worked fine out-of-box, no extra drivers needed), Moblin v2 (not a big fan of the UI, but ran very well from an 8GB SDHC card), Fedora 12 (big issues with this one, but it’s only an Alpha release), and now I’m at Fedora 11.

I dumped the Live image of Fedora 11 (Gnome desktop manager) on to my SD card (the U123 supports booting from SD cards, something my old Acer AspireONE couldn’t do), and installed it in less than 10 minutes, boot times included. Almost everything works out-of-box, however the Gnome UI was horrifically huge. Most windows would fill the screen and beyond, making it impossible to see buttons, let alone click them.

The solution? Change the default DPI. To do this, login, then click the System menu, followed by Preferences and Appearance. In the resulting window, click the Fonts tab, then the Details button at the bottom. In the new window, change Resolution to 96 and hit Enter on your keyboard (as the Close button is just out-of-view). Everything should be back to normal.

In addition to this, I also changed all of the font sizes to 8, but that isn’t strictly necessary – I just like to get as much on the screen as I can.

Beyond that, everything else works perfectly, including sound, wireless, the SD card reader, and even the webcam.

Update (Dec 12, 2010): The Wind U123 works perfectly on Fedora 14 with no additional hacks/changes needed.

So there’s been a lot of news lately about Google’s newly announced operating system cleverly called ‘Chrome OS’. What is it? In short, it’s a bare-bones Linux distribution (yes, it uses a Linux kernel for both the ARM and x86 versions) that puts focus on web applications. What does this mean to the consumer? Probably not a whole lot.

Here’s the thing: there’s a cycle that goes on between people/businesses having Desktop Systems and Thin Clients. This has been going on for ages. With a desktop system (which most people are used to), everything is literally at your finger tips. All of the software is installed on your local computer, and that’s where you do most of your work. If you’re in a business, chances are you save your files on to a remote server, but that’s typically in the same building.

With Thin Clients, things are a little different. In days of old, you had a terminal, which consisted of a monitor and keyboard. This was networked to a mainframe-type setup, which housed everything. These days, Thin Clients typically have a very minimal operating system (either something *nix based, or Windows CE) that allows you to use Citrix, Remote Desktop (terminal services), etc… to connect to a server that contains all of your apps. Not a lot of difference.

Chrome OS, from everything that’s been said, definitely appears to be of the Thin Client sort – basically, you boot it up and you’re on the web. Done. Likely you’ll be able to muck about with installation, and it will hopefully still support Offline applications, however the goal appears simple: once you’re connected, you’ll be using Google Apps. Yep, you’ll be in the ‘cloud’, with all of the benefits and problems that go with it.

Depending on your needs, this is all well and good. If you have a broadband connection, you’re laughing. If you have a netbook, the target platform, this is supposed to be Nir-freaking-vana. However, I’m not convinced. Not yet, anyways.

The problem is, if this thing were to launch right now, I can’t imagine it would do horribly well. With cell phones capable of tethering, 3G USB sticks, and WiFi hotspots everywhere, internet access really isn’t a problem. And now that netbooks have matured and are actually usable (my 8GB SSD Acer AspireOne proves that it doesn’t pay to be an early adopter) and Windows 7 nearly out the door, what’s the point of yet-another-OS?

The pre-releases of Windows 7 run amazing well on netbooks (there are even rumors that Microsoft is making an ARM port of it). Fedora 11 is brilliant (even on my slow-as-hell AspireOne), and I’ve heard that Ubuntu rocks the platform. Intel is pushing Moblin (I’ve tried it, and I really don’t care for the UI). Even OS X runs well, according to the Hackint0sh crowd. All that said, why is Google bothering with making an OS of it’s own? Will it be a fork of one of the above distros? Will it retain binary compatibility with it’s upstream brothers? Until it’s out, we won’t know for sure.

All that I can say is that I don’t think it deserves the hype. We need to sit back, wait for Google to bless us with a beta, and then start hailing it as the best think since sliced bread or just another OS.