Scott Dowdle's blog
I've been building a Fedora Remix for some time now. If I remember correctly I started around Fedora 9 and have continued to build them with each new release. I'm on Fedora 13 now. I usually rebuild the remix every time a new set of updates comes out. So far I had rebuilt the i686 and the x86_64 remix 46 times each... and then someone reported some problems with the last couple of builds. I didn't notice because I had been on vacation and was doing the rebuilds remotely without testing the final product. I figured if it built ok, it was probably ok... because I hadn't previously had any problems with any builds.
I wrote a rather long response to a posting I saw on Fedora Planet entitled, "Death of the Year of the Linux Desktop". I'm sharing it here as well.
- - - -
The desktop is dead? Some disagree. See this very compressed video on the "Death of the Desktop":
What makes someone a winner and what makes someone a loser? Linux on the desktop has tens of millions of users. While that might not be double-digit market share it is still a significant number of users. Anyone who looks at those numbers and calls them losing must have thought the game was winnable to begin with. It wasn't. FOSS does not spend hundreds of millions on advertising and billions on under-the-table deals with hardware makers... FOSS simply does not operate on the same playing field. The free hand of the market happens to be attached to a twisted arm.
In reality there aren't any winners and losers because it isn't a race. While those who are in it for money have certain ways to measure success, in FOSS you just make the best software you can and hope users will appreciate it. Of course the squeaky wheels always make the most noise and there are lots of complainers out there (see discussions on KDE 3.x vs. 4.x as an example) but that doesn't mean that vast majority of us FOSS users aren't very happy.
Recently a GNOME survey (aka the Neary report) came out that showed who contributes to GNOME and at what levels. Not so oddly enough the results of it turned out similarly to periodic Linux kernel surveys done by LWN and Greg KH. The results being that Red Hat is the top named contributor.
It just so happens that Canonical (the sponsor of Ubuntu) typically does not fair so well on such surveys and as a result they are often criticized for their perceived lack of upstream contributions.
I don't usually repost mailing list messages but just got this one in my inbox from the OpenNode folks. Since I'm a big virtualization geek, I'm sharing. Haven't heard of OpenNode? Here's a brief description before I get to the status update email:
OpenNode is a open source server virtualization solution providing easy to use (CentOS / RHEL based) bare-metal ISO installer and supporting both OpenVZ container-based virtualization and emerging KVM full virtualization technology on the same physical host.
So, OpenNode is a lot like Proxmox VE except OpenNode is based on CentOS and uses libvirt, virt-manager, and other Red Hat standard tools.
You have to look closely but yes, that is an OLPC. The darker color, antennas down, and hard plastic keyboard really make for a different look, eh? I have no idea why the desktop background on display is from Fedora 7. I believe the idea for offering a model for older kids with a different keyboard came from seeing some altered OLPC units that modders had done. I wish I had some bigger pictures so I could see the keyboard better. It looks very similar to my Acer keyboard except for the function keys.
I first blogged about webm the day Google released it. It has taken some time but now I have full support for webm in my preferred Linux desktop distro (Fedora 13). I've been doing some testing and I have to say I'm impressed.
Why even care about webm? Because I prefer to use royalty-free file formats that are based on open standards and free / open source software. Any other questions? :)
I'll cover both webm playback and encoding.
I saw an announcement the other day about a development OS release (os16) for the OLPC XO-1 laptop that basically brings it into parity with the release on the XO-1.5. I downloaded it, got a developer key, unlocked an OLPC, and figured out how to install it. Once you become familiar with the process, it is actually easy and straight forward. I even played with the FORTH-based firmware for the first time.
The main new features in the OS16 devel release are:
- Based on Fedora 11 (was 9)
- 2.6.31 Kernel (was 2.6.27)
- Includes "Switch to GNOME" option
- Additional productivity Apps
- Updated Sugar release
Of course the hardware in the XO-1 has not changed but the new software still runs quit well in 256MB of RAM with no SWAP.
Just wanted to mention a few news items from the OpenVZ Project.
Updated vzctl - vzctl 3.0.24 has been released. Even though the version number only changed from 3.0.23 to 3.0.24 there are a ton of changes, fixes and some feature additions. Of special interest is the --swappages option as well as being able to refer to a container by its name rather than requiring the CTID with vzmigrate. Over all it was a long overdue, much appreciated update.
Updated Official OS Templates - The last wiki notice is dated April 27th but looking today at the dates on the OS Templates they appear to have been updated May 27th. One thing to note is that there are now OS Templates for Ubuntu 10.04 which I'm sure Ubuntu folks will be happy about.
Beta Fedora 13 OS Templates - And speaking of OS Templates, Kir just released Beta OS Templates for Fedora 13. On the day Fedora 13 was released I tried creating my own OS Templates by taking Fedora 12 containers and updating them but ran into a snag. With Fedora 13 a lot of new stuff has been added to the init setup and some of it causes a container to just hang during startup. I was glad to see the beta OS Templates released. I created containers from them, made my own changes, and then uploaded those to the contrib section.
As luck would have it, later in the afternoon the Fedora Project released a whole bunch of updates and among them was a new initscripts package. I suspected that when I upgraded my container whatever changes the OpenVZ folks had made to the init setup that made it work in a container would be wiped out and I was correct as upgrading the initscripts package did make the container get stuck in the init process upon container reboot. I ended up filing two bugs: 1566 and 1567. I joyfully await their resolution.
2.6.32 devel kernel - There have been a few releases of the 2.6.32 devel kernel and it appears to be making progress. While there have been a number of OpenVZ devel kernels that died on the vine, 2.6.32 should be different mainly because it is the kernel in the upcoming Red Hat Enterprise Linux 6, the upcoming Debian 6, and in Ubuntu 10.04. I have no guess as to when it'll be marked stable. My guess would be sometime after RHEL 6 is released.
***Please note that any URLs mentioned (and the information they contain) in this posting are time sensitive and will surely be outdated not long after posting.
It appears that the netbook shown in the video is a slightly newer build of the one I have (Acer Aspire One D150) and should fit in a number of 10.1" netbooks. Unfortunately I don't have $275 to spare to try this screen out myself. This is the first OLPC-type screen available as a mod.
You can order one from here if desired.
I think I'll wait until the Pixel Qi screen (or one like it) is standard in new netbooks. In this case I don't think it is too practical because, to the best of my knowledge, you can't get to the eReader mode (monochrome at a higher resolution) until you have a BIOS update that will allow the backlight to be completely turned off. Supposedly though, in color mode, they are still very readable in direct sunlight.
I'm investigating virtual desktops at work. My goal is to do a small proof-of-concept trial of VDI this summer, see how it works out in practice, and then expand it or dump it as appropriate into the future.
What I like about both products:
- Uses the Linux-based KVM hypervisor
- Supports both Linux and Windows Desktop VMs
- Designed with VDI in mind rather than bolted on