Saturday, March 29, 2008

Ubuntu 8.04 Beta: Compiz Coverflow?

I don't know how, but I've bumbled into a new (to me anyway) Compiz feature, a Coverflow-like ability to view multiple windows on a desktop. I installed KSnapshot in order to get a delayed screenshot of the effect, shown below.

Note the black bar running horizontally across the center of the image. This is an artifact of KSnapshot. It's not a part of the desktop. The effect is triggered by the key combination [Windows][Tab], or [Command][Tab] on the Apple keyboard. When you release the [Window] key on the keyboard then the window in the center is the upper window on the desktop. Very very cool. This is a Compiz feature worth having enabled. And the movement is very smooth, tight, professional, and fast without the crazy wobbles that first afflicted Compiz. And to think all this is running on a slow (1.8GHz) 32-bit Athlon XP with only 512MB of DRAM.

I am properly impressed!


It's available under the CompizConfig Settings Manager (ccsm), which you have to install via synaptic. The specific effect is Shift Switcher under Window Management.

Ubuntu 8.04 Beta: A second install

When I upgraded my 7.10 installation to 8.04, audio support broke during the process. When I went Googling for possible causes (and hopeful fixes) I ran across the rather snarky comment from a developer that 8.04 was an alpha, and as such if you wanted everything To Just Work then you needed to install outright rather than upgrade. Considering that technically he was correct (but flying in the face of personal positive experience with past upgrades) I stuck the 8.04 beta CD in rhea and performed a completely new install. And sure enough, sound support worked again.

The problem was the video screen. I have, up until this release, never had any problems with configuring high resolution after Ubuntu was installed. We're not talking about successful guessing with the Live CD, we're talking about a complete installation (or upgrade) with all the proper bits. And I'm talking about alphas, betas, and final releases. Until now, that is, with Ubuntu 8.04.

I run my monitor, an ancient Sony E400, at 1600 x 1200. System | Preferences | Screen Resolution in beta 8.04 doesn't know what kind of monitor I have and fixes it at 1280 x 1040. If you look below you'll see everything I'm complaining about except the resolution.

To fix this problem I went looking for nVidia settings via synaptic, found it (nvidia-settings), and installed it. I then had the tool I needed to fix the resolution, NVIDIA X Server Settings.

When it's installed it winds up on the System Administration menu. When you open it the first time it shows up on the desktop in 'cramped' mode.

Selecting any of the entries (such as X Server Information) opens up the application to its proper screen size.

The entry we're interested in is the second, 'X Server Display Configuration'.

You'll note that the nVidia tool does identify the monitor attached to my system and it also properly identifies (and allows me to set) both resolution, refresh, and (on the 'X Screen' tab) color depth. I was able to dynamically set and test screen parameters just like I would have under Windows XP. The only problem was saving the changes. Xorg still uses a text configuration file (/etc/X11/xorg.conf), and to save the changes I had to first save them to my home directory, then, as root, move it to /etc/X11. Just to be on the safe side I made a backup of the original configuration file before moving the new one over. After those changes I reset the X server ([Ctrl][Alt][Backspace]) and everything was back to normal screen-wise. And before you comment, I'm well aware of Applications | Other | Screen and Graphics Preferences. I've invoked it, worked with it a bit, and frankly it's little better than Screen Preferences. Two different screen resolution applications in two entirely different locations that both perform poorly is a major usability issue that needs to be fixed before 8.04 goes live. But based on history and experience I doubt it will. If you've got an nVidia graphics card or chip on your system, then install the nVidia management tools. You'll be glad you did.

One other annoyance concerns gnome-screenshot. For some inexplicable reason it is removing the window border on individual window captures, regardless of how I set it. It didn't start this until 8.04.

Rather than leave this post on a sour note, I'd like to extol the virtues of one of Ubuntu's and Debian's better (if not killer) applications: synaptic. This application, and the underlying repository architecture make managing applications a true joy. The more I work with all the tools the more I discover, and the more I like. For example, I was grossing about how there was no history trace for installations, upgrades, and removals of software. I was wrong. Under the Synaptic Package Manager, on the File menu, is the History selection which brings up the History window.

With the history window you can either browse (like above) or search and then browse the results. Synaptic has been a powerful tool for some time, and combined with Add/Remove under Applications, gives the user a quite powerful pair of application management tools. It makes finding and installing applications drop-dead simple compared to Fedora and openSUSE.

Finally, I'd like to write several postive comments about Firefox 3 beta 4. It really is fast, faster than Firefox 2. What's more, a number of little Firefox annoyances have been fixed. One that always gets to me is when I'm writing a blog entry with Blogger's editor and I want to add a link. On earlier versions of Firefox, when I clicked the 'Link' button to add a URL to a selection of text, the cursor would display it's selection image (the clinched fist), and sure enough, it would be carrying the URL of the Link icon. I got into the habit of clicking outside the Link dialog to make it drop, otherwise I'd wind up with a URL to that little icon in the dialog's text box. That seems to have been fixed with Firefox 3 beta 4. It's still a bug on Firefox 2, because europa is still running Ubuntu 7.10 and Firefox, and i encounter it all the time.

Hey. Like I've said many times before. A bad day with Linux is better than a good day Windows. And there are far more good days with Linux than bad.

Linux last OS standing, but is it really better?

The last notebook left unbeaten in the Pwn 2 Own contest is a VAIO VGN-TZ37CN running Ubuntu 7.10. The other two, a MacBook Air and a Fujitsu U810 running Vista Ultimate SP1, have fallen. The MacBook was the first and fastest to fall via a Safari exploit. The Vista machine fell due to a bug in Java. I have no idea what version of Java that was, but it's disturbing that (1) we still have security problems with Java after all these years and (2) Vista would allow a compromise in Java to compromise the operating system.

The fact that Ubuntu hasn't been hacked yet does not prove Ubuntu's superiority to either Mac OS X or Vista. There was this interesting comment from the show:
Some of the show's 400 attendees had found bugs in the Linux operating system, she said, but many of them didn't want to put the work into developing the exploit code that would be required to win the contest.
So the bugs are there ready to be exploited. All somebody needs is real motivation.

Friday, March 28, 2008

Microsoft's chickens are coming home to roost

From the article "Study sees Microsoft brand in sharp decline".
Microsoft's brand power has been in sharp decline over the past four years, an indication the company is losing credibility and mindshare with U.S. business users, according to a recent study by market research firm CoreBrand.

According to the CoreBrand Power 100 2007 study, which polled about 12,000 U.S. business decision-makers, Microsoft dropped from number 12 in the ranking of the most powerful U.S. company brands in 2004 to number 59 last year. In 1996, the company ranked number 1 in brand power among 1,200 top companies in about 50 industries, said James Gregory, CEO of CoreBrand.
Gregory said that a decline in and of itself is not indicative that a company is losing its mindshare or reputation among customers. However, what's significant in Microsoft's case is that the decline has been consistent over a number of years, and has plunged dramatically in a brief time.

"When you see something decline with increasing velocity, it's a concern," he said.
Gregory could only speculate as to why Microsoft's reputation has been declining, since his firm does not ask people that specific question. He said the "underwhelming" response to Windows Vista might be one reason, and Apple's clever "I'm a Mac, I'm a PC" advertising campaign -- which paints Windows in an unfavorable light -- may be another.
Let's look at the recent history of Microsoft for some possible answers. Note that what follows is incomplete, but it's a good start to building a bad taste in one's mouth.
  • 1993 - In August the Department of Justice opens an investigation into whether Microsoft is abusing its monopoly in operating systems.
  • 1994 - On July 15 Microsoft settles with the Department of Justice and consents not to tie the sale of other Microsoft products into the sale of Windows. Microsoft was free to continue to integrate "new features" into Windows. Microsoft would use that hole to integrate Internet Explorer in the coming years.
  • 1995 - Sun releases Java 1.0 and the HotJava browser on May 23. Mark Andreessen of Netscape announces during the festivities that the Netscape Navigator browser would include a Java Virtual Machine. Netscape IPOs on August 9. Netscape's shares start at $14 and finish the first day of trading at $75. Microsoft releases Chicago, i.e. Windows 95 August 29. Microsoft is already aware of Netscape Navigator's threat to commoditize the operating system, and begins to work on its own browser, Internet Explorer. Version 1 was released as a part of the Windows 95 Plus Pack. Internet Explorer 2 is released in November with support for Windows 95 and the old Macintosh operating system.
  • 1996 - In March Microsoft licenses Java from Sun, which sews the seeds for a future Sun/Microsoft lawsuit over Microsoft's Window-specific extensions to Java. This is another expression of Microsoft's embrace, extend, and extinguish philosophy towards all competitors. In August Internet Explorer 3 is bundled with Windows for the first time with the release of Windows 95 B (OSR2). Microsoft is ranked number 1 in brand power.
  • 1997 - In August Microsoft releases Internet Explorer 4 for Windows 9x and NT 4. Bundled with IE 4 is Microsoft's latest version of Java, supposedly based on Java 1.1. October 7, Sun files a lawsuit against Microsoft over Microsoft's implementation of Java. Microsoft's implementation fails to implement two key interfaces, RMI (Remote Method Invocation) and JNI (Java Native Invocation). In addition Microsoft has modified the Core Java class libraries with the addition of 50 methods and fields that are not a part of the public Java API. As a consequence any developers making use of Microsoft's extensions will create Java applications that will not operate properly on other JVM implementations, and implementations that depend on RMI and JNI will fail to operate properly on Microsoft's JVM. Microsoft's Java implementation has effectively broken 'Write Once Run Anywhere' (WORA). Sun's lawsuit is its final attempt to force Microsoft to bring its JVM into compliance.
  • 1998 - In January Netscape offers Navigator 4 for free and makes Navigator's source open via the Mozilla Project. Netscape also lays off a large number of employees due to poor performance, especially against its chief competitor Microsoft. In May the Department of Justice and 20 state Attorneys General sue Microsoft for illegally thwarting competition in order to protect and extend its software monopoly. Microsoft releases Windows 98 on June 25. Internet Explorer 4 is now fully integrated with the operating system, specifically the GUI and file manager. In September Microsoft released Internet Explorer 5. In October the Department of Justice also sues Microsoft for violating the 1994 consent decree by bundling IE 4 with Windows. On November 24 AOL announces it will acquire Netscape Communications for a stock swap valued at the time at $4.2 billion. This 'purchase' is seen in time to a large waste of money on AOL's part, and eventually contributes to AOLs future fall from grace within Time Warner. Because Netscape is dead, Microsoft no longer feels competitive pressure. It's Internet Explorer release schedule slows drastically. It will not release another version of Internet Explorer until 2001.
  • 1999 - Microsoft releases Windows 98 Second Edition (SE). This version is a bug-fix of Win98, and is considered by many to be the most stable Win 9x release. On November 5 the presiding judge over the Microsoft trail, Judge Jackson, releases his findings of fact. In it he states that "Microsoft's dominance of the personal computer operating systems market constituted a monopoly, and that Microsoft had taken actions to crush threats to the monopoly, including Apple, Java, Netscape, Lotus Notes, Real Networks, Linux, and others."
  • 2000 - In February Microsoft releases Windows 2000. On April 3 Judge Jackson issues a two part ruling. His "conclusions of law were that Microsoft had committed monopolization, attempted monopolization, and tying in violation of Sections 1 and 2 of the Sherman Act, and his remedy was that Microsoft must be broken into two separate units, one to produce the operating system, and one to produce other software components." On September 26 Microsoft attempts an end-run on the ruling by sending their appeal to SCOTUS. SCOTUS refuses to hear it, instead sending it back to a federal appeals court. In July Microsoft releases .Net 1.0 pre-beta. Originally code-named COOL (C++-like Object Oriented Language), .Net and C# in particular is Microsoft's answer to Java. In August the European Union widens its anti-trust case against Microsoft based in part on anti-competitive complaints from Sun Microsystems. It is, at the time, the third action brought against Microsoft in Europe. In September Microsoft releases Windows Millennium Edition (Me). An update to Win 98 as well as a stopgap on the way to Windows XP, it was so filled with bugs and instalbility that it was also referred to as the Mistake Edition.
  • 2001 - In January Sun settles its lawsuit with Microsoft. Under the terms of the lawsuit "Microsoft will pay $20 million to Palo Alto, Calif.-based Sun, terminate all Java licenses, and agree to a permanent injunction against the use of the Java Compatible logo." Microsoft cannot use Java within .Net nor j#. Microsoft can continue to sell existing code based on Java 1.1.4. Meanwhile the D.C. Circuit Court of Appeals overturns Judge Jackson's remedies in part due to interviews Judge Jackson gave regarding the trial. Judge Jackson in turn responded by saying that "[Microsoft executes] proved, time and time again, to be inaccurate, misleading, evasive, and transparently false. ... Microsoft is a company with an institutional disdain for both the truth and for rules of law that lesser entities must respect. It is also a company whose senior management is not averse to offering specious testimony to support spurious defenses to claims of its wrongdoing." In May Microsoft begins development of Longhorn, the future Vista and next release of Windows. In October Microsoft releases Windows XP, the version that combines the Win 9x line with the Windows NT line. This release met with considerable criticism, in particular due to security holes, requiring Microsoft to release two service packs to address (SP1 and SP2). It wasn't until the release of SP2 that Windows XP was widely deployed, especially in business. Internet Explorer 6 was also released along with Windows XP. In September the Nimda worm begins to spread as an email attachment. Users of IE 5 and 5.5 are infected simply by reading the email. It certainly isn't the first, but it signals the sophistication (if you will) of virii writers and illustrates just how extensively and quickly a worm can spread across Windows clients and servers. Also in September the Department of Justice announces it is no longer seeking to break up Microsoft. In November the DoJ seeks a settlement with Microsoft. The monopoly conviction stands, but the remedy now consists of Microsoft opening its APIs to the world and of appointing a three-person panel to oversee compliance. The biggest victory for Microsoft is that Microsoft can still tie other software with Windows in the future under the rubric of "innovation".
  • 2002 - .Net 1.0 is formally released in February. Windows XP SP1 is released. Not much changes with Windows XP, and uptake is still slow. Most users, especially business, prefer to continue using Windows 2000. In November the government accepts the final resolution in the Microsoft antitrust case. Nine states (California, Connecticut, Iowa, Florida, Kansas, Minnesota, Utah, Virginia and Massachusetts) and the District of Columbia do not agree with the settlement, stating it does not go far enough.
  • 2003 - A big year for Windows worms, starting with SQL slammer in January that dramatically slowed down general internet traffic until it was finally found and removed on most systems, continuing with Blaster, Sobig, and Sober and many others. Longhorn was supposed to ship late this year. It does not. Instead many of the Longhorn/Blackthorn developers are re-tasked to enhance the security of Windows XP and Windows Server 2003, which was released in April.
  • 2004 - More Window worms. January opens with MyDoom, the fastest spreading mass-email worm evah. In May we get Sassar, which winds up bringing a number of companies to their knees. In March Microsoft is hit with a record $613 million fine by the European Commission for abusing its "near monopoly" with Windows. In July Microsoft pays the fine, placing the money in escrow while Microsoft continues to appeal the ruling. August arrives with the delivery of Windows XP SP2, and the widespread adoption of Windows XP, introduced two years earlier, finally begins. At the same time Microsoft announces it is revising its plans for Longhorn. The original Longhorn code base, based on Windows XP, is scrapped, and development essentially starts anew with the Windows Server 2003 SP1 code base. A key feature, highly touted since the mid-90s under other names, called WinFS, is postponed (and eventually dropped). The scope and sweep of Longhorn features are pulled back across the board. This is also the year that Microsoft decides to finally get serious about Windows security.
  • 2005 - Longhorn is named Vista in July. A comprehensive Vista beta program is started. Also in July Microsoft releases a media-player-free version of Windows XP for the European market in an effort to placate the EC with as little effort as possible. The EC is not amused, and in December threatens to start fining Microsoft 2 million euros/day until Microsoft starts giving rivals more access to its systems. Meanwhile, in October, Sony is found to have distributed Windows root kits with its music CDs in an effort to keep purchasers of its CDs from ripping songs from those CDs. In November Sony tries to 'fix' the problem with a removal kit that is as bad as the original root kit. Sony eventually recalls all affected (infected) CDs. The problems with the Sony root kit fiasco are many; it's poorly written, allowing for other malware to piggy-back on the Sony root kit, forcing the Department of Homeland Security to issue an advisory calling Sony's root kit a security threat. Although technically not Microsoft's fault, the root kit is easily installable due to security design flaws in Microsoft Windows which have not been fully addressed up to this point. And just in time for Christmas, Microsoft launches XBox 360, beating Sony and Nintendo to market with a next generation game console.
  • 2006 - In February Windows Vista CTP is released. It is considered feature complete at this point. In March Microsoft announces that it will push the release of the consumer version of Vista to January 2007. In November Microsoft releases Vista to key business users. Meanwhile, XBox 360 consoles are beginning to fail early and fail often. Some are estimating that failure rates are as high as 30%, which is 10 times the industry standard. The standard XBox 360 failure mode has a name: the Red Ring of Death. In November Microsoft releases its answer to the iPod, the Zune. Many take note of Microsoft's draconian DRM practices, in particular the three-day or three-play rule. The Zune has built-in WiFi, and uses it to allow transfer of songs between Zunes. The only problems are the three-day/three-play rule and the fact that it's applied to every song that's transfered, not just songs purchased from Microsoft. Further, Microsoft, by creating the Zune and the Zune Marketplace, has in one fell swoop cut off all previous partners of its "Plays For Sure" music initiative. Zunes are interesting at first, but as the truth spreads via the web, interest drops markedly.
  • 2007 - Windows Vista is released. Controversy immediately erupts about PCs labeled as "Windows Vista Capable". Tagged earlier in 2006 in order to help sell new systems over the 2006 Christmas holiday, punters now have their shiny new Vista to install, only to discover that Vista Capable is not good enough, and in most cases won't run what many believe Vista should provide - the Aero interface. Bill Gates claims that they have sold over 40 million copies of Vista, nearly double what they sold of Windows XP, during the same period. There's just three problems with this; computer manufacturers are now selling twice as many machines as they did back in 2001 when XP was released, many users (especially business) are actually having Vista replaced with Windows XP, electing to keep the Vista license until such time as it's good enough or Vista SP1 is released, whichever comes first, and the majority of the copies are OEM. Regardless the tech press from all quarters is relentless in its harsh criticism about the poor performance and poor hardware (driver) support of Vista, as well as the maddening User Account Control. In June, Microsoft takes a $1 BILLION dollar charge on its income statement to help pay for bricked XBox 360s. And in spite of a Zune refresh, iPod still rules. Microsoft is ranked number 59 in brand power.
  • 2008 - February is a bad month for Microsoft. The Windows Vista Capable lawsuits are granted class-action status. The emails released due to discovery are damning to both Microsoft and Intel; they show that Microsoft executives knew they had a problem with Vista, and that even for them Vista was a sorry excuse. The European Union fines Microsoft a record $1.4 BILLION for defying sanctions imposed on it by the EU for anti-competitive behaviors. It's just the end of March, but 2008 is turning into a banner year for Microsoft bad news. Who knows how low their brand power will sink?
I know I've forgotten many important facts, or just didn't put them in due to my fingers getting tired of typing. You can add them as comments if you like. But when I hear somebody make completely unbelievable and clueless comments such as they don't understand why Microsoft's brand has been in "sharp decline", you just have to wonder what alternate universe they've been living in all this time. I've outlined 15 years (year-by-year, every year) of Microsoft dirty tricks and poor business practices, and I can assure you it ain't just Vista, baby. The question shouldn't be why their brand is in 'sharp decline' but how they've managed to even stay in business. And the answer is their operating system monopoly and the inordinate monetary reward for having it.

  1. Windows Products and Technologies History
  2. History of Microsoft Windows
  3. Windows Vista
  4. United States Microsoft Antitrust Case
  5. .Net Framework
  6. Google results: History of Java
  7. European Union Microsoft competition Case
  8. Windows Bugs
  9. Timeline of notable computer viruses and worms
  10. Anything but speechless; 100 things people are really saying about Vista
  11. Living with Vista, one year on
  12. Tracing Microsoft's Vista Capable Debacle
  13. etc...

Yeah, that Safari browser for Windows is just peachy

With all the lame FUD being slung about how Apple is "forcing" Safari on poor unsuspecting Windows users, and how the EULA only allowed for one copy of Safari (just one!) to be installed only on Apple machines (which has now been fixed), you'd think that it couldn't get worse, PR wise, for Safari and Apple.

Unfortunately it can. A contestant at the CanSecWest security conference's PWN 2 OWN hacking contest broke into a MacBook Air using a Safari exploit in just two minutes. This is no different than last year, when a MacBook was hacked and broken into via a - you guessed it - Safari exploit. Yes, I know that those exploits are on the Mac, and I'm talking about Safari on Windows. But those kind of exploits on one platform (Mac OS X) make me take pause about running it on Windows. I can already get that kind of security with Internet Explorer. But I want better, which is why I use Firefox. Having said that, I hope that Mozilla's upper management can get a better grip on things and quit complaining about Apple's practices. Take all that boundless energy and put it towards advancing and enhancing Firefox. Compete on merit, not on FUD.

Sunday, March 23, 2008

Question to Dennis Howlett: Why are you so bloody stupid?

In a post titled "Question to Mozilla CEO: what do you fear?", blogger Dennis "Howler Monkey" Howlett tries to back up a very important question with out-and-out gross distortions of the facts. For example, in the first paragraph:
The kerfuffle over Apple’s decision to include the Safari 3.1 ‘update’ with an iTunes update installer for Windows made me smile. John Lilly, CEO Mozilla Foundation seems outraged at Apple’s decision. It only adds to the hilarity. Why? In enterprise land, neither browser is likely to be the number one choice down at the PC/laptop image factory. Not even close. Check the stats at W3schools to see what I mean. Mozilla comes in at a miniscule 1.1% while Safari has just hit the 2% mark.
I did check the stats over at W3schools, and you know what I found? Mozilla (Moz) does indeed come in with a mizerly 1.1%. But look over to the left column next to it labeled Fx (for Firefox) and you see that Firefox has a 36.5% share. Add the two together and you get 37.6%. That's a good third of the market.

Earth to Howler: The Mozilla Foundation is responsible for both the Firefox and Mozilla browsers, and it's been primarily (and now exclusively) Firefox for years now. Where the hell have you been?

The the Howler Monkey further howls:
And while I’m in Mozilla flame mode, exactly when will Firefox stop leaking memory like a sieve by default and routinely bringing my machine to a grinding halt? That would be a much better story in my opinion because right now I’d prefer to use Safari, short though it may be on all those Firefox plug-ins, add-ons and themes. Why? Because in enterprise land, performance matters, not all the bells and whistles.
Let's talk about memory leaks for a moment. I'll be the first to admit that Firefox consumes memory like a drunk in a licker store. But is Safari 3.1 any better? Let's take a look, shall we?

Oh my goodness! Safari running on my Windows XP SP2 system, with six tabs, consumes almost 400MB. Firefox 2, while by no means svelte, consumes in the high 200s on this system, and that's with considerably more tabs open to the same sites. Why is this? Because I use three Firefox plug-ins to keep memory consumption down on my systems: Adblock Plus, Adblock Filterset.G Updater, and NoScript 1.5.2. These three plug-ins keep ads, especially Flash-based ads, off my web pages. Safari, with nothing like them, renders everything, and in the process sucks up even more memory than Firefox 2.

Safari 3.1 on Windows is indeed a good, fast browser with regards to speed and quality of rendering. But when it comes to overall quality, Firefox 2 is as good if not better. Safari provides yet another choice for browsing, but as for me, I'll happily stick with Firefox.

Saturday, March 22, 2008

I, for one, welcome our new Android overlord

It's been a while since I played with Google's Android SDK (and I do mean play), so after after it surfaced recently in the tech press I decided to crank it back up and see what was different. So I downloaded the latest Eclipse release (3.3.2) and the latest Android SDK (M5 RC15), installed everything, and fired Eclipse IDE back up again.

First, before you can use the SDK in the IDE you have to install the Eclipse Android plugins, which are at If you've worked any with Eclipse then you know you have to add the https address as a new remote site, then install all plugins found there. One of the two sets of plugins, the editors, requires the Eclipse WTP (Web Tools Platform) plugins as well, so when you go to install new features select the Google Android site as well as teh Europa Discovery Site. When you select both the Google entries the editors selection will have an error indicator. Click WTP. then the "Select Required" button to pick up all the other supporting Eclipse plugins.

After the installation of the plugins I shut down Eclipse then unpacked the SDK. I guess I should have turned that around (unpack the SDK then install the plugins), because then I had to restart Eclipse and under Window | Preferences I pointed Android to where the SDK was located. That still wasn't enough, however. Remember that I worked with an earlier version of this tool kit back in November 2007. As a consequence I had to perform a three-step upgrade on the earlier Eclipse Android project:
  1. Under project Properties | Libraries, remove the old android.jar and add the correct one in the new SDK.
  2. Shut down Eclipse and change directory to workspace/.metadata/.plugins/org.eclipse.core.runtime/.settings, and modify four files:[dms | apt], and The last line of every file points to the location within the Android SDK for its specific tools. Modify the stem of those lines to the correct SDK directory location.
  3. Restart Eclipse and open the project. The project will be in error because of schema changes to AndroidManifest.xml that render it invalid. You can follow the problem entries to each incorrect line of XML. Before you do that you might want to navigate to the Android API Changes Overview and read it. In a nutshell you need to rename certain XML attributes to 'android:name', such as the 'class' attribute in the activity tag, and the 'android:value' attribute in the action and category tags.
Once you've done that you can run the ported Android project, like I've done below.

I know that there are Android plugins for NetBeans 6, but I've been drawn back into Eclipse for this and several other projects where I found Eclipse to be a better tool set. And that's a shame considering the promise and current capability of NetBeans. But I've always believed you use the right tool for the job. I'm not here to show disrespect to NetBeans.

Operating systems are not cars

Every once in a while I come across the argument being made that OS choice is like car choice. It's usually in a forum involving Linux, where one poster will lead off that there's too much choice and the follow-up will read something like:
But how will we ever choose what car to buy and drive with all this choice.. it's too much..

Or maybe, the user can focus on the device and features it provides rather than the embedded OS. The people who will focus on the embedded OS are not usually the type that are scared by choice.
Operating systems and environments are different enough that if cars truly were built the way we build our software there would be far fewer cars and they'd be a lot more expensive. The key to ubiquity is the interface, and the simpler the better. All cars come with the following common user interface:
  • Ignition or starter (true on/off)
  • A steering wheel
  • A break pedal on the floor
  • An accelerator pedal on the floor
  • A gear selection
  • Speedometer
  • Gas (or petrol) gauge, indicating full to empty
  • Rear-view and driver side mirrors
And that's for automatic transmission cars. For manual transmission cars the gear selection becomes a little more complicated (more gears) and a clutch pedal is added to the break and accelerator on the floor. Using a manual shift requires a bit more coordination and attention on the part of the driver, but the key point is this: once you've learned either (or both) skill sets on any one vehicle, you can be pretty darn sure you can use the same skill set with just about any other car manufactured. And not just on the current crop of automobiles, but even older vehicles that were manufactured back before you were born. That's what standards can really accomplish if manufacturers allow it.

But what about user environments and the operating systems that host them? Let's do a quick run down of today's operating systems:
  1. Windows (in all it's versions)
  2. Windows Mobile
    1. Win 3x
    2. Win 9x
    3. Win NT
    4. Win 2k
    5. Win XP
    6. Vista
  3. Palm OS
  4. Mac OS X
    1. Leopard
    2. iPhone/iTouch embedded Mac OS X
  5. Linux
    1. KDE
    2. Gnome
    3. Xfce
    4. Fluxbox
    5. ...
  6. Linux Mobile
    1. Maemo
    2. Android
    3. ...
  7. Solaris
  8. *BSD
Why do I list them this way? Because each major division (Windows, Mac, Linux, etc) is significantly different enough that moving from one to the other requires significant re-training because there are enough major and minor differences between them to perform the same tasks. They may look nearly identical at first blush, but put someone in front of a Mac (or Linux or ...) who's been sitting in front of Windows for the last 10 years and you're going to be spending a lot of time getting them up to speed and assuring them they're not stupid. Even Microsoft's changes from major version to major version have involved enough changes that a lot of people have not upgraded and will never upgrade because what they have works for them and they do not want to have to unlearn one set of skills and learn a new set to perform the exact same tasks.

And that is the lesson the auto industry learned at the start of the twentieth century, and made absolutely sure they never forgot. Not only do car companies have to compete with other car companies, but they need to convince customers to trade up. And the only way that they can do that is to make sure that (1) the new cars are at least as good as the old cars, and (2) that Ma and Pa Kettle don't have to learn anything different at all to drive the new and shiny car just like they drove their old trade-in klunker. Violate either of those and you will not sell cars. Or at least not enough to survive.

If us software folks made our user interfaces along the same lines then we really could compete with Microsoft and Apple. And that leads to some interesting consequences. We Linuxers like to point out the superior merit of our favorite OS and our favorite desktop environment. As do Windows and Apple fanatics. But if we all stuck to the same user interface and truly competed on merit, then how successful would we really be? I have some partial answers that will anger just about everybody:
  1. If we had truly interchangeable operating systems for our hardware, then there would be churn that would make the mobile industry look timid by comparison.
  2. In a world of real competition, Vista would be seen by all as the unmitigated disaster it truly is. Right now, because of Microsoft's monopoly on the desktop, they can ship 100 million 'somethings' and charge good money for it, then claim success with Vista as well as billions in gross receipts per quarter. But in a world of real competition, without the special contracts with computer manufacturers, said computer manufacturers could pick another OS and UI and install it in place of Vista on new machines, and wouldn't that have a powerful effect on Vista and Microsoft? Do you think for one minute that Microsoft would have waited as long as they did to release Vista SP1? Do you think Vista would have taken as long as it did to appear? Do you think Vista would have even seen the light of day? Real competition has a poweful Darwinian effect on products and services in markets where there is competition. The weak truly do not survive.
  3. Mac OS X has its supporters, and it's continuing success is based on both hardware and software, deftly integrated, from one manufacturer: Apple. Apple has successfully re-invented itself as the true alternative to Windows by providing the complete package. Jobs has long recognized that limited standards loosely adhered to results in a wide ranging user experience (from mediocre to truly horrible). This is a good observation. The problem is Jobs solution. While many complain (and rightfully so) about Microsoft's monopoly at the software level, Jobs has extended control into hardware, and with the iPhone SDK, even into how new applications are written and accepted for use on the newest platforms. But if Apple were truly open and could be interchanged with Windows or Linux on x86 systems, do you really think it would succeed? It probably would as an OS, but the mystique and the buzz would be considerably diminished. And one thing is for certain. If the operating systems were truly interchangeable across the Mac hardware platform as it would be across any other x86 hardware platform, then the premium hardware price Apple currently commands would be under severe pressure to go down. And Apple literally can't afford that.
  4. Linux, Linux, Linux. And Linux. Linux is so fragmented across its various distributions, let alone its various desktop environments, that to say Linux would succeed would be to ask "which one?" Right now Linux has an official 0.6% of the desktop environment. Linux is doing much better on mobile devices such as cell phones and in the cheap ultra-portable market, such as Asus' Eee, but that's because the manufacturers picked one and only one distribution to support (or they created one from scratch, furthering fragmentation). One thing is for certain. In a truly competitive market there would be far fewer distributions and there would have to be a lot more consistent quality across the board, especially with regards to hardware support. I've run too many recent releases where hardware that worked just fine in an earlier release suddenly stopped working in the next. Quality, especially at the driver levels in the kernel and X, is the Achilles heel of Linux.
Operating systems are operating systems, not automobiles. Our 'freedom' in the OS space results in too much effort expended when moving from any given OS to another, and in my not-so-humble-opinion is actually holding true innovation back. One lesson we can take from the auto industry is to agree on one set of core controls (the 'UI') that is consistent across brands and time so that true cross-usability is achieved. Then innovate in areas that really add to the experience. When we do that in the OS space then we'll have achieved something truly significant, and we can get on with advancing the state of the art in meaningful ways.

Friday, March 21, 2008

Looks like this Democrat's voting McCain

After ceaseless political faux pas from both Democratic candidates as well as the DNC, I have come to the considered conclusion that I'll once again vote Republican like I did for Ronald Regan. Twice. It's been a long slog getting to this point, and I've been wanting to vent about this. But before I got started I found all of my reasons for switching so sweetly summed up over at Time Magazine.

In an article titled "Will Dean Cost the Dems Florida?", article author Tim Padgett clearly articulated every problem that the current Democratic National Committee under Howard Dean has created and gotten itself into up to this point. I'm going to quote a number of paragraphs from that article, but let me lead off with this one first.
[Howard Dean], along with the other sage bosses of the DNC, has left Democratic voters in what is arguably the nation's most crucial swing state feeling dissed, disenfranchised and, it now seems, disinclined to back whomever the Democratic candidate is in November.
You got that right. The self-righteous hypocritical bastards "in charge" of the DNC decided that they would punish both Florida and Michigan for moving up the date of their primary elections. Florida moved up to January 29th. Why did we move up? Because Florida, with over 18 million citizens (fourth behind California, Texas, and New Your) and 27 electoral votes got tired of being over-shadowed by New Hampshire, with less than 1/10th our population (1.3 million) and four, yes, count 'em, four electoral votes.

We got some serious Big Buck issues to contend with here in Florida. We've got a booming population, hurricanes (in 2004 and 2005 no less than four passed over Orlando where I live, and we are inland), a fire season second only to California in destructiveness, tornadoes, and tourists. (Note to European tourists: this does not include you. Please visit often and bring lots of €uros). What has New Hampshire got? Hell if I know, but they only pop up once every four years on the political radar when it comes time to pick a new president, and the conventional wisdom is that if you don't win New Hampshire then you don't win the presidency. Well folks, recent history shows that in 2000 and 2004, if you didn't win Florida (and Ohio) then you didn't win the presidency. New Hampshire is just one out of 50, and a pretty damn small and inconsequential one at that. In the end they didn't help decide squat.

But I digress. Tim writes so much better than I, so let me continue to quote what he so pointedly reported.
Then there's the question of all the prodigious Flori-dough. Prominent Florida Democratic donors and fundraisers are now threatening to withhold or seek the return of hundreds of thousands if not millions of dollars from the national party if at least some of the state's delegates are not reinstated.

Those risks were already apparent when Dean and the DNC made their original fateful decision. Last May, Republican Florida Governor Charlie Crist and the state's GOP-controlled legislature — fed up with what they call an absurd presidential primary process that gives small states like Iowa and New Hampshire inordinate clout — decided to leapfrog Florida's primary from March to Jan. 29. The move violated Democratic as well as Republican party rules, but many if not most Florida Democrats also supported it. Still, the DNC ruled that all 210 of Florida's Democratic nominating delegates would be annulled. It exacted the same draconian punishment on Democrats in Michigan, which moved its primary to Jan. 15. The DNC also allowed Iowa, New Hampshire and other small, early-primary states to force Democratic candidates to pledge not to campaign in Florida or Michigan.

Dean has consistently argued that the integrity of party rules is at stake. But that seemingly principled stand rests on shaky ground. In a New York Times op-ed article this week, Michigan Senator Carl Levin and Debbie Dingell, a Michigan member of the DNC, pointed out that one of the perennially pampered primary states, New Hampshire, also broke newly established party rules last year by defensively moving its own primary to an earlier date — and the DNC allowed it. Even discounting that apparent hypocrisy, Florida Democrats insist that the moves by their state and Michigan should have indicated to the DNC that the rules were antiquated and flawed, and therefore required some flexibility. "I detest this ruling," says U.S. Representative Robert Wexler, Barack Obama's Florida campaign director. "This should have been a wake-up call to the party that the primary system needs to be more representative and democratic."

Either way, the DNC's lack of foresight is astonishing, even more so now that Florida and Michigan have rejected the idea of costly and less than reliable primary revotes. After all, the Republican National Committee annulled only half of Florida's GOP delegates — a more measured ruling the DNC could have mirrored. And while Democratic rivals Obama and Hillary Clinton couldn't set foot in Florida in January, John McCain and his Republican competitors campaigned there and scored valuable face time with Florida independents, with McCain even winning the endorsement of the popular Crist. Despite all that, Florida Democrats campaigned with cardboard cutouts of Clinton and Obama and then turned out to vote in record numbers for the primary election, which was remarkably blooper-free by Sunshine State standards. All the while they were convinced that in the end the DNC would never shut out their delegates, especially with Clinton and Obama running neck-and-neck into the spring.

Their simmering frustration now is beginning to boil. "None of this makes sense to me," says JoAnne Bander, a Democratic activist in Miami and a Clinton supporter. "I feel completely disenfranchised. How can they dismiss the turnout we produced and keep treating Florida as if it were some marginal consideration?"

Perhaps because Dean and the DNC painted themselves into a corner. They can't easily lift the Florida-Michigan sanctions after all the authoritarian chest-thumping they did last year. Yet if the party heads into Denver without a clear nominee — and needing the votes of Florida and Michigan to decide the issue — their peremptory action will seem even more ridiculous, making the leadership of the so-called people's party look like a clique of arrogant patricians thwarting the popular will.

What's worse, Dean and the DNC now look all but AWOL when it comes to resolving a mess they did so much to create, leaving it to the states to figure it out. Nor have Crist and the Florida legislature been much help after they were the ones who led the state into the primary rebellion in the first place. (In this week's poll, Florida Democrats lay equal blame on Dean and the Florida GOP.)

With any hope of a revote in either Florida or Michigan all but dead, the candidates themselves aren't helping to resolve the mess they too helped create by going along with the DNC. Clinton accepted the DNC ruling last year when she was the front-runner; but now, because she won the Florida and Michigan primaries but trails Obama in the delegate count, she's the earnest champion of voters like Bander. She backed the idea of revotes in Florida and Michigan; but this week rejected a compromise solution offered by Democratic state Senators in Florida that would take the GOP tack and reinstate half the delegates (giving Clinton 63 and Obama 42) and perhaps divide the other half equally between the two candidates or divvy them based on the popular vote that has so far been tallied nationally. Clinton insists instead that short of a full revote, all the Florida and Michigan delegates must be counted and seated as they stood in January.

The Florida compromise was also rejected by Obama, who didn't even put his name on the Michigan ballot in January. Otherwise, not surprisingly, he is keeping sheepishly quiet — and a bit unpresidential — about the whole thing, looking to many as if he is simply trying to run out the clock by raising objections to a proposed revote in Michigan.
And these are the fools you want in charge of the executive branch? I don't know who are the bigger bunch of fools; the Republican conservatives like Rush Limbaugh who would rather see the Democrats take the White House because "... the country will suffer with either Hillary, Obama or McCain, I would just as soon the Democrats take the hit ... rather than a Republican causing the debacle," or our fine group who are so splendidly and consistently incompetent "Dem-witted" Dim-ocrats?

Beta spotting: rhea upgraded to Ubuntu 8.04 Beta 1

Well, I stuck my neck out again and upgraded rhea, running Ubuntu 7.10, with Ubuntu 8.04 using the command 'upgrade-manager -d'. And in spite of what the upgrade dialog says, it took twice as long to download and install everything.

So far, I've discovered that sound is broken (nothing will play) and Totem, because of the sound problem, will not play back any video. VLC works fine in its place. Everything else works as near as I can tell, especially Compiz. Rhea has the nVidia 7600GS video card, and the restricted drivers provided by Ubuntu work flawlessly with Compiz providing a smooth and satisfying experience.

As an experiment I booted rhea with the Ubuntu 8.04 beta live cd, and sound works. So something during the upgrade farkled up the sound. I have a number of choices at this point:
  1. Live with it and hope that future upgrades magically fix the sound problem without breaking anything else.
  2. Do a fresh install of Ubuntu 8.04 beta and then accept the upgrades up until release.
  3. Re-install Ubuntu 7.10 yet again, gaining further experience in that arena.
  4. Accept greater risk and install openSUSE 11 alpha 3, thus moving back to the Dark Side.
Since it's Easter weekend (this is Good Friday), I'm going to punt and just let it ride through the weekend. After all gcc was upgraded to 4.2.3 (after it was discovered that 4.3.0 exposed a but in the kernel) and Java is at version 1.6.0_04 build 12. Playing with the shiney new tools will keep me happy for the time being.

Thursday, March 20, 2008

Alpha spotting: openSUSE 11 Alpha 3 stirs cautious optimism

The third alpha of openSUSE 11 was announced yesterday, so while I was puttering about the house before bedtime I downloaded the ISO, specifically the KDE live CD. I then burned the ISO and rebooted europa with my newly minted CD. That's when I ran into my first problem.

I don't know when it first cropped up, but the kernel driver for my NEC DVD/CD burner seems to be broken. While I can use it as a normal DVD/CD burner and player after Linux has booted, using it as the boot device results in failure. It refused to boot in either normal or safe mode using the NEC. It's a good thing I have two DVD/CD devices on europa. The second is an older LiteOn DVD/CD device that will only burn CDROMs. So I swapped the openSUSE CD to the second drive and sure enough, it booted without further ado. openSUSE finished at the graphical KDE 4.0.2 desktop. I poked it prodded it for about 10 minutes and didn't discover anything horribly wrong, then shut 'er down and got some sleep.

This morning I decided to give openSUSE something harder to work with: my Gateway M685 notebook with the NVdia GeForce Go 7800 graphics chip. Just about every distribution I've laid hands on in the past 12 months has had issues booting on the M685. The last known good distributions to boot successfully and completely on this machine were openSUSE 10.2 and Ubuntu 7.04. By completely I mean coming up in full graphics desktop with full screen resolution of 1680 x 1050. Just about every distribution since then has either come up in the Black Screen or required that I boot in "safe graphics" mode. Today I'm happy to say that this release of openSUSE successfully and completely booted on the M685.

After going though failure after failure my expectations were pretty low. When it came up successfully I just had to sit back in amazement. This is using the built-in Xorg driver, so I owe somebody in the Xorg group a heart-felt thank you for making this work once again. I just hope it stays fixed.

I don't have a lot of time, so I can only perform the most cursory of tests, but I've already found one problem; any attempts to start Konsole results in a crash. That's no problem as I can just switch to a virtual terminal. And this is, after all, an alpha release.

Another pleasant surprise occurred with Firefox. The Firefox version in Alpha 3 is Firefox 2 is not a bad choice. I was able to write this entry without any issues. I was quite pleased with how the Flash plugin installed without any issues. When I went to the CNN website to view video Firefox popped up a dialog and downloaded and installed Flash without any problems whatsoever. Not even a restart was required. Once installed I was able to view flash video. Even the sound was excellent. The overall experience was equal to Windows, including (unfortunately) the ads. But then that's what Adblock Plus is for.

I'm still not sure about KDE 4. While I have no problem living on the bleeding edge, I tend to like my bleeding edge reasonably useful. KDE 4 is just different enough from KDE 3 that I would tend to install KDE 3 on openSUSE 11. But then again, my impressions are with a rapidly evolving desktop, and KDE 4 is advancing with incredible speed. Who knows how it might behave 30 days, or even a week from now?

I knowingly tempt the Fates by saying this, but based on what little exposure I've had to it so far, I think openSUSE 11 just might be a really good release.

Saturday, March 15, 2008

And now for something completely different...

Rhea is a 'constrained' system, primarily in its memory. It has only 512MB. In today's world that's nothing, unless you've got a pocket system along the lines of an iPhone. But the typical desktop system (and notebook too, for that matter) ships with between two and four gig of memory.

And that, from my perspective, is an incredible amount. For personal uses our operating systems and our applications shouldn't keep demanding ever greater amounts of resources every time there's a new release, such as faster processors and more memory just to perform the same tasks. And we've been performing the same tasks since we identified what we wanted to do in the late 1980's, especially in business. In PC gaming it seems to get worse a lot faster. We've grown so enamored over the rendering and the look of the game that we've forgotten plot, game-play, and just general enjoyment (but that's the subject for another rant).

With that mind-set I stumbled across "Building a highly functional desktop with lightweight software" on The author gave a very high-level overview of Fluxbox and other light-weight applications for running on a constrained system with 128MB of memory. And these were applications running with Ubuntu. Perfect, I thought. So I read the article and then attempted to install the applications, starting with Fluxbox. All of the following work was done in a separate account from my regular Gnome login.

When I say high-level, I mean high level. The article gave no indication of versions or any special steps. My system runs Ubuntu 7.10 and I installed Fluxbox 1.0.0 from the (universe) repository. I also installed fbpanel 4.9 from universe as well as IDesk version 0.7.5 from universe. I then stopped to test the Fluxbox desktop and make sure everything worked up to that point.

It's a good thing I did. Fluxbox is a bare-minimum desktop. It consists of a set of menus that can be selected from the desktop with the middle and right mouse buttons. There is a single tray across the bottom of the screen. That's it. It turned out that the version of Fluxbox installed from the 7.10 repositories did not install all the tools, and importantly did not create the desktop menus. I found all this out from "HOWTO: get a Fluxbox menu (and customization)" on the Ubuntu forums. The key to fixing the missing menus under Ubuntu 7.10 is to run (in either a terminal or virtual screen) "sudo update-menus". Once you've done this then when you log in using Fluxbox you'll have the menus you need.

I also installed IDesk and PCManFM. IDesk provides the ability to add icons to the desktop via very simple text-based files, while PCManFM provides a highly functional file browser that isn't Nautilus. From a shell I ran Nautilus after installing and fixing Fluxbox, and when it started it started up the Gnome background as well. PCManFM just starts. I've got one screen shot showing a number of applications running. From left to right, clockwise, you have Firefox, PCManFM, Gnome System Monitor 2.20.1, and Gnome Terminal 2.18.2. These are applications provided from the Ubuntu 7.10 repositories. I have three icons on the desktop via IDesk for PCManFM, Firefox, and Gnome Terminal.

There are some differences between my setup and the author's in the original article:
  • PCManFM uses the Tangerine icon set, not Tango.
  • I installed FBPanel and tested it, but in the end decided not to use it. It added nothing to the functionality of the desktop that Fluxbox base and IDesk didn't already provide.
  • I'm using Firefox simply because, in spite of its warts, it's IMHO the best all-around browser, especially with regards to properly rendering every website I go to (and I go to a lot).
  • The Fluxbox style I'm using is MerleyKay, which is not what the author chose.
Overall Fluxbox reminds me a lot of twm, except that I like Fluxbox with IDesk a lot more.

I going to give some very, very basic benchmarks comparing the two desktops running on rhea. Rhea is built around an Athlon XP 2500+ (1.8 GHz) processor with 512MB (333 MHz FSB) of memory on an nVidia nForce2 Ultra 400+ - based motherboard. Times were measured with a digital stopwatch. Here are my measurements taken from running full blown Gnome and Fluxbox running on rhea:

First login after reboot (sec)4.517
Startup after first login (sec)Nearly instantaneous10
Memory consumption after first login, with desktop and Gnome System monitor only (MB)80120
Memory consumption with shell, Firefox (three tabs), system monitor, and PCmanFM (mb)160230
Application startup1 sec or less3 sec or less

The statistics on memory consumption don't tell the full story. Memory consumption with Gnome tended to grow over time to the point that swap was hit. Memory consumption with Fluxbox stayed flat over time, and swap was never touched except when running 1080p film clips (Indiana Jones, Wall-E, and Iron Man) from Apple's trailer site.

Performing this experiment has been fun. I've learned a bit more about Ubuntu and a lot about Fluxbox. The Fluxbox environment makes a good final desktop for those who want to really strip a distribution down to its bare essentials. It isn't better than Gnome (or KDE) and it isn't worse. It's just different, and you have to both appreciate and accept its differences. In the end, no matter how much I grouse about the problems with Linux I still find it fun, powerful, fun, highly flexible, fun, and open. That's why I stick with it.

Friday, March 14, 2008

Alpha testing, Day 5: Like a monkey with a hand grenade

Well, the grand experiment with Ubuntu 8.04 alpha 6 has come to an ignominious end. That last batch of 10 updates included an update to libc6, which, once I finally got it installed, completely hosed the system. Like the monkey with a hand grenade who figures out how to pull the pin but doesn't know what to do next, the whole thing just blew up in my face. I couldn't log in (it gave multiple malloc errors before crashing back to the prompt in safe mode) and if you tried a normal boot it hung at the point where it was trying to start logging. I tried various 'recovery' methods, but in the end I:
  • booted into Ubuntu 7.10 live (the DVD), mounted root, opened up a shell, and backed up my home directory
  • re-installed Ubuntu 7.10
  • restored my home directory
  • got on with life.
This is the first time in years (and I'm talking back before 2000 now) that I've so throughly hosed a Linux installation that I had to just back up and re-install. And it took a lot of work to get there, let me tell you. Not with an update, but some seriously inspired 3-in-the-morning-with-far-too-little-sleep hacking. Not that it took that long to re-install and re-configure, mind you.

In the process of re-installing I discovered that updating an existing Ubuntu installation may not be such a good idea. For example, the amount of 'material' on the root device after a clean re-install was cut by over 60% when compared to what I had after the update from 7.04 to 7.10. What's more, the performance on rhea is even snappier than it was before. Of course, I learned long ago never to trust SuSE updates, and every time I upgraded it was always as a clean install. And then, of course, there's Windows. No, the reason I felt comfortable upgrading Ubuntu is because it appeared to be the most successful; that is it rebooted and continued to operate but with the newer bits running. Now that I've been provided the opportunity to compare a fresh install vs. an upgrade, looks like I'll be doing fresh installs from now on.

When I compare what I've experienced so far with Ubuntu 8.04 alpha with 7.04 and 7.10 alpha testing, I'm surprised that the earlier runs went as well as they did. And that frankly concerns me a bit. I would have thought that 8.04 testing would have gone a bit better because this release is intended to be Canonical's Long Term Support, or LTS, release. But this testing sequence has been the roughest, with one little niggling issue after another leading up to the update marathon that led, in turn, to the libc6 debacle and complete collapse. But I'm going to wait at least for the betas before I attempt to load it again, and reserve judgement on 8.04 until then.


Remember folks, I'm describing my experiences installing and running the alpha code. What I described above hasn't happened to me with released production code.

Thursday, March 13, 2008

Alpha testing, Day 4: Bunches and bunches of updates

Yep, the day started off bright and early with 108 updates in the pipeline for Ubuntu 8.04 alpha 6. Since I had to go to work I waited till I got home and loaded them up. Buried in that large update were more updates for the kernel. As soon as those updates installed, then another group of 66 showed up. Then a set of three. Right now I'm looking at another 10 in Update Manager. Three of the updates (108, 66, and 10) have updates to the kernel (2.6.24-12). So far none of the updates have fixed the breakage that is currently occurs with the nVidia drivers and kernel 2.6.24-12. I've looked at the Xorg log and dmesg, and the only problem I see is in the Xorg log where Xorg fails to 'initialize GLX extension (Compatible NVIDIA X driver not found)." I have no idea what that means just yet, but I'm sure there's a bug out there already about it. I just haven't had time to chase it down. In the mean time I just continue to boot with 2.6.24-11.

Tuesday, March 11, 2008

Alpha testing, Day 3: A whole slew of updates

One hundred sixteen updates were sitting in the update manager, ready to come barrelling on in. One update in particular (and one I looked for specifically) was for the kernel:
Version 2.6.24-12.20:

[Ben Collins]

* Enable CONFIG_SOUND at least, so alsa build in lum works
- LP: #200338
That one was good for a smile. Maybe sound will work again after the update.


Rebooted rhea after all the updates had been installed and discovered that the latest version of 2.6.24-12 is even worse than the prior release of 2.6.24-12. Not the sound, but the video. When rhea tried to come up in the graphic desktop it failed and I was left staring at the big 'you must configure the desktop' screen that currently passes for fixing a busted graphics subsystem.

Bugger that.

So I rebooted back into the slightly older but much more functional 2.6.24-11 kernel. Not much else exciting to report.

Monday, March 10, 2008

Alpha testing, Day 2: A new kernel, new problems

Got up this morning and checked to see if rhea was still okay after its upgrade. It was still operational and reporting that about 66 or so new upgrades were ready to drop in, including an upgrade to the kernel (2.6.24-12-generic). So I let all 66 install, rebooted, and low and behold I had no sound. So I rebooted back to the prior kernel (2.6.24-11-generic) and sound worked again.


So I rebooted back into 2.6.24-12-generic, and looked for sound and snd entries in dmesg. Sure enough all the sound devices (drivers) had 'unknown symbols' (lots of them), indicating the kernel hadn't been configured properly before it was built. Oh, well. The -11 kernel worked, so I figured I'd just go and comment out the -12 entries in /boot/grub/menu.lst. No big deal, right?


When I tried to run 'sudo vi menu.lst' I got a failure in sudo: 'unable to resolve host rhea'. I checked in /etc/hosts, and sure enough the 127 entries are missing rhea as an entry. So, how can I gain rootly powers when sudo seems broken? Turns out that gksudo still worked, so I use that to fire up gedit and edit the hosts file. Then I can use sudo to edit menu.lst. The sudo issue was already reported (Bug #195308) back in late February.

The fun you have running new software. I'm real curious how 2.6.24-12-generic got out the door with the sound subsystem as broken as it appears.

Sunday, March 09, 2008

Acid 3, Firefox 3b3, and Ubuntu 8.04 Alpha 6

After reading about the new Acid 3 test, I mossied on over and ran the test. Even though it didn't get a 100/100 score I was never-the-less impressed with how well it performed.

I'm somewhat surprised by the scores on Anomalous Anomaly. AA's chart shows Firefox scoring higher than 3.0b3, while my testing shows just the opposite. What's more 3.0b3's rendering is a lot closer to the reference rendering than's. Regardless of the scoring there's one thing I will say about Firefox 3. It's fast. Very, very fast.

rhea upgraded from Ubuntu 7.10 to 8.04 alpha 6

Well, I took the plunge today and upgraded my 'lesser' system, rhea, from Ubuntu 7.10 to Ubuntu 8.04 alpha 6. The upgrade required over 1,200 files be upgraded/deleted/added, and over two hours of time for the whole process to complete. The time was long because the upgrade stopped twice for confirmations and I wasn't sitting in front of the keyboard to immediately make the decision. The good news is that the upgrade succeeded in spite of itself (more below). That's rather remarkable considering that I started with an Ubuntu 7.04 alpha 4 initial installation on this system, and I've been upgrading it every sense.

This is an upgrade to an alpha release, so a few odd events did transpire. Here's my short list of those odd little happenings.
  • I don't know when or why, but I had VirtualBox OSE (Open Source Edition) installed. It must be old age but I don't ever remember installing it. In any event the attempts to upgrade VirtualBox caused multiple installation failures (and halting of the process with dialog boxes). It appeared that the entire upgrade was a failure because of the virtualbox failures, leading me to have minor heart failure. In the end I calmed down and figured that if the upgrade was a failure I could re-install 7.10 from scratch. But after rebooting the system the upgrade was a much bigger success than the failure messages indicated.
  • The upgrade process halted and waited for me over samba. The dialog wanted to know if I wanted to keep my current samba configuration file. This was, again, one of those times I was away from the keyboard, and it just sat there until I got back. I kept the configuration file.
  • After the first reboot, I was pleasantly surprised to see that all the graphics effects still worked. Rhea has an nVidia 7600GS video card, unlike europa, which has an ATI X1950 Pro video card. I can only assume that the system was using the older drivers because I had the little 'nasty nag' dialog on the desktop telling me I had to enable the non-free drivers in order to use my card. I clicked on the icon, and the icon simply disappeared. I then went to the 'Change Desktop Background' dialog (right-click on the desktop) and clicked on the 'Visual Effects' tab. Under Ubuntu 7.10 there were four selections on that tab, with the forth being Custom. With this release the hook for custom Compiz settings is not there. In fact, none of the three selections were selected, so I selected Extra. It was at that point that I did get the custom video driver dialog and was able to 'enable' the use of the nVidia drivers (in spite of the fact that they seemed to be in place and running already). That blew my custom Compiz settings from 7.10, so I had to run ccsm from a shell to get back my custom settings and behaviors.
  • I'm concerned about upgrading to Firefox 3. My Adblock Filterset.G Updater is disabled at the moment because it's not compatible with Firefox 3. I've got other Firefox plugins on europa, which is why I am hesitant to upgrade europa.
Nits notwithstanding, the end result is pretty good. All my settings transferred unaltered. All the necessary codecs and drivers upgraded and transferred as well, so that all my multimedia continued to work. Again, this is an alpha and such problems as I've encountered can and will occur. But overall the complete experience is good. So far, on a 'slow' 1.8 GHz Athlon XP 2500+ and 512MB of DRAM, the system shows excellent performance, even with Firefox 3 open with multiple tabs and playing an MPEG4 movie. Obligatory eye candy follows.

From what I can tell I may be able to have full Compiz support on europa with its ATI card. I believe, from what I've seen so far with synaptic, that the repositories now contain the ATI 8.02 (February 2008) drivers. I think, however, I'll wait, at least until the second beta. Europa does yeoman work and rhea makes an excellent test bed.

Wednesday, March 05, 2008

Ubuntu notes

ATI Driver Update

Version 8.3 was released today (March 5th), and I installed it. Nothing special to report. Performance was no different than before, and DRI (AIGLX) support is still broken under Ubuntu 7.10 (Gutsy). The release notes give the official changes, and the Unofficial Wiki has been updated.

Linux Mint 4 KDE

I'd like to note a regression with Linux Mint 4 KDE. I booted the live CD on my Gateway M680 notebook and discovered that both wired and wireless networking would not work. I then booted the same notebook into Windows XP (already installed), openSUSE 10.2 (already installed), and Linux Mint 4 Gnome live CD to check out the networking hardware with those operating systems and found it still worked. I then booted back up into Linux Mint 4 KDE one last time and found networking was still inoperative. Everything else I bothered to test worked just fine. It's a shame, really. I was looking to replace openSUSE 10.2 on that machine with Linux Mint 4 KDE. I'll wait for now, but for what I don't quite know.

Sunday, March 02, 2008

Casual viewing: Mandriva 2008 Spring RC1, Fedora 8

In a January post I uttered the asinine comment that I was not going to bother with Mandriva after that posting because it had failed twice in succession to boot on my machines, especially europa. I forgot about that utterance and downloaded Mandriave 2008 Spring RC1 KDE One, and booted it on europa. I can say that it not only booted and ran, but the overall quality ranges from good to excellent.

One of the key features I feel is important and which I always test with a new release is the ability to play back some video (any video, frankly) that I have on my system.

As you can see above Mandriva found and mounted my existing partitions, including my home partition and all its data. I was able to play back a movie with Kaffeine that I had ripped with K3b (MPEG4 video, MPG3 audio) without any problems. The ability to just Do It is still a wonder, both from a technical and ideological perspective. It failed to play back a DVD. Kaffeine started up and played the opening screen and FBI warning, then failed to continue with a "not enough permissions" error message. Keep in mind that europa has two DVDs, so I was using the second as the player while the boot CDROM was in the first. DVD playback failure is not such a big deal; after all, Simply Mepis failed as well. I will say, however that Linux Mint 4 had no problems with DVD playback.

And as you can see from the capture above, Firefox comes equipped to play back nearly any Flash-based video stream.

I tried to boot Mandriva both with and without Compiz. It did boot into both, but because of Mandriva's selection of the ATI driver and my card (X1950 Pro) it performed a lot better without Compiz. This is a problem with the ATI driver and 3D effects enabled, not with Mandriva. In both cases Mandriva used the ATI driver.

Fedora 8

I wanted to retry Fedora 8 to check some facts. I booted Fedora 8 booted on rhea (with nVidia 7600GS). I wanted to check and see what would happen if I tried to play ripped video (see above). When I double-clicked the video Fedora attempted to play it back via Totem. I was presented with a dialog that gave me three choices; free playback via GStreamer, an MPEG Playback Bundle for 16 euros, and an MPEG 4 part 2 decoder for 7 euros. Wow. I selected free-as-in-beer GStreamer option, and all I could play back was the audio portion.

Then I tried to play back the same type of CNN video in Firefox. Instead of it just playing, I was presented with a dialog to download Adobe Flash. It went through the motions of getting it, then failed to install it.

that I know is built into Fedora is not a bad distribution. Once installed it's a handsome and easy distribution to work with as long as what you want is already installed and configured correctly. But its support of video playback is crippled, and I personally don't like to have somebodies hand stuck in my face asking for money to purchase a codecVLC and MPlayer (Totem needs to be abandoned, and Kaffeine needs to be cleaned up or dropped alongside Totem).


My initial judgement notwithstanding, Mandriva 2008 Spring may just turn out to be a pretty good release and a decent distribution, especially if you want good video support off-the-CD. If the final release is as good as this release candidate then this will make the second consecutive Mandriva release that actually works from the get-go and is a pleasure to work with. Based on my admittedly limited experience and not-so-humble opinion I rank it up there with Mint 4 and Simply Mepis 7.

Fedora 8 either needs to provide the codecs and drivers necessary to fully enable today's computer hardware, or someone needs to spin off an alternative distribution with everything needed, the way Mint is an alternative spin of Ubuntu. It's an excellent quality distribution that's been deliberately hamstrung by misguided ideology, and it shouldn't be that way.