[user@localhost Project]# ./bin/wmic -U “administrator%password” //192.168.1.133 “SELECT SizeStoredInPagingFiles FROM Win32_OperatingSystem”
CLASS: Win32_OperatingSystem
Name|SizeStoredInPagingFiles
Microsoft Windows Server 2003 Standard Edition|C:\WINDOWS|\Device\Harddisk0\Partition1|2095104
[user@localhost Project]# ./bin/wmic -U “administrator%password” //192.168.1.133 “SELECT FreeSpaceInPagingFiles FROM Win32_OperatingSystem”
CLASS: Win32_OperatingSystem
FreeSpaceInPagingFiles|Name
2081296|Microsoft Windows Server 2003 Standard Edition|C:\WINDOWS|\Device\Harddisk0\Partition1

The basic format of the output is that the first line contains the name of the table you’re getting data from (or “Class” in WMI terms); the second line is a pipe-delimited list of columns in the same order as the data; the rest of the lines are individual data lines, pipe-delimited. As you can see from the above example, I’ve got a few problems. First, I’m getting more columns than I asked for. I only asked for one column in each query, and the column header clearly shows two columns being returned. Second, the “Name” column (which is the one I didn’t ask for) actually uses the pipe character in it’s result set. This makes it impossible for me to parse the output by column accurately.

I suppose the delimiter issue could be the blame of the WMI client that I’m using, which is clearly third-party, but the fact that I’m getting columns back that I didn’t ask for would seem to be squarely the fault of Microsoft. So, I thank you Microsoft for being the source of my most-recent head-ache. I’m now going to create some horrible hack to get the information that I need…

From the climate change supporters perspective, it’s easy to simply look to an “expert” in a given field and make up one’s mind based on that person’s opinion. If I were to go with that approach, I’d never have to do any real investigation into the evidence itself; I could read someone’s summary and spend the extra time playing Team Fortress 2. Sounds good to me. But I really do enjoy analyzing statistical data. It’s fun, and I get to verify/discount suppositions without having to depend on the word of someone else (of whom I’ve probably never met nor ever will meet).

From the perspective of the climate change opposition, it’s also easy to look at the size and complexity of our world and conclude that we 6+ billion people cover only a small percentage of this planet’s surface, and therefore could hardly make that large an impact on it’s operation. However, the recent oil spill in the Gulf of Mexico is one example of how humans can indeed have an impact on relatively large portions of the globe. So, empirical evidence tends to toss this idea out the window rather quickly.

I’m trying to be fair to both sides of this debate, but notice I haven’t addressed the “experts” on either side, since 99.9% of the debating that I’ve seen is done by people with little or no experience in either climate science or even scientific research in general. I count myself in this category, so I’m not knocking the common man by any means. I have my own opinions, which I’ll no doubt be expanding on below, but I wanted to show that both sides have strengths and weaknesses to their own most-bantered arguments.

Now, on to my own thoughts on the issue:

Climate change is happening; there, I said it. But this statement shouldn’t be a shock to anyone. The Earth’s climate hasn’t stopped changing from the beginning of time. The real questions are:

• Is the climate change we’re seeing today unusual, given our past observations?
• How many years have we been collecting climate data in a consistent format?
• Where is the raw data?
• What steps were taken to analyze the data to reach your conclusion?

As you can see, the most important of these questions is the third on the list above. I still have yet to see honest-to-goodness raw data and a methodology for examining that data to reach a conclusion of unusual climate change. In addition, the “dummy” graphs I’ve seen in articles promoting the concept of man-made global warming, for example, show that we’ve been collecting climate data for only 300 years or so. If the world is as large as many scientists say it is (in the billions of years; though I don’t hold to this idea myself), observing climate changes for this length of time is a terribly small sample size in the Earth’s total climate-changing life.

There is, also, this large ball of fire in the sky that keeps this world heated; you may have heard of it: the sun. Apparently, the amount of energy it casts toward the earth doesn’t always remain constant. Sometimes, it has what the experts call “solar flares” which cast large bursts of heat/energy into space. One would expect that when the amount of energy being output from the sun increases, the temperature of the earth might rise with it in some reasonable correlation. So, given this thought, it seems possible that even if there is something unusual about the climate patterns in the last hundred years, there might be other causes for it than man kind. I’d like to see charts comparing some other variables:

• Chart comparing atmospheric CO2 levels and global population levels (more humans means more breathing)
• Chart comparing temperature and solar flare activity (and not just number of flares, but also energy output as a result of the flares, since different flares have different intensities)

One other thought that’s been floating around in my somewhat-empty head is whether temperature readings on the moon might be helpful here as well. Since I suggested that solar flare activity could be a culprit, if we accepted that we do indeed have unusual climate activity on the earth, one might expect that the Moon’s climate might change also (though on a much different scale, since it has much less atmosphere than the Earth).

So, that’s about it. The primary reason I have chosen to reject the climate change position is that I haven’t seen any real data. All I’ve seen are simple charts (e.g., the “hockey stick”), that could mean just about anything given the proper numeric massaging (see Political Math Blog for ways you can manipulate numbers to say anything you want).

I also was wondering today if I could find a way to see if there were any HD versions of movies that I had already recorded in SD coming up in the listings. I think this MySQL query should do the trick…

SELECT DISTINCT program.title FROM recordedprogram LEFT JOIN program ON program.title = recordedprogram.title WHERE program.videoprop LIKE '%HDTV%' AND recordedprogram.videoprop NOT LIKE '%HDTV%' AND recordedprogram.programid LIKE 'MV%';

That’s a good start, I think. The next step is to wrap that in a Python app that will automatically schedule that HD version of the movie to be re-recorded, and set it up in a cron job that runs on a daily basis. I was thinking about joining the program and recordedprogram tables on programid, but I wasn’t sure if the HD and SD versions of the same movie would have the same programid, and I figured the title should be a sufficient match most of the time (e.g. of the less likely scenario, Ocean’s Eleven is the name of two different movies).

I decided to install Intrepid when it hit alpha stage. It had just been too long since I had gotten any significant software updates, and I was getting itchy (and not just because I tried growing a beard…which didn’t turn out very well). I decided to give Intrepid all of the credit it deserves, and I wiped my desktop system, after cleaning up and uploading my home directory to my local NAS. I hadn’t been doing much reading on Intrepid, so I was slightly surprised to see that KDE 4.1 is the default desktop in Kubuntu now. I was slightly worried after my KDE 4.0.0 experience.

So, after installing the system, I booted it up for the first time to find a not-so-speedy desktop in front of me. Now, folks, I’m not trying to brag here, but I’m running quite a capable system. It’s an Athlon 64 X2 4000+, 4 GB DDR2-800, GeForce 8600 GTS w/ 256MB. Windows, menus, and dialogs shouldn’t lag when requested, but yet, here we are at KDE 4.1.2….lagging. Now, I’ve heard that there has been or are problems with the nVidia driver and KDE 4, so I’m going to also try it out on my T61 which has Intel’s 965 chipset (one of the best-supported video chipsets in the Linux world). So, for the time being, I’ll lay aside my issues with the desktop being laggy.

On the whole, I’m impressed with the number of annoyances that have been fixed in the months between KDE 4.0 and 4.1.2. The panel is now resizeable (though only by the mouse, not by a specific pixel height); the desktop doesn’t move if you click-and-drag in the wrong spots; I can change the window decoration back to Plastik without KDE 4 crashing; running xtermcontrol doesn’t crash Konsole (I reported a bug on that ). I’m sure I’m missing a few things that don’t bother me anymore, but those are the ones that are coming to my mind quickly.

And now, comes the part that would make most KDE developers cry (if any of them actually read my blog). I’m going to give my top 10 list of KDE 4.1 annoyances (as of 4.1.2):

1. You can’t change the font or time format (24-hour vs 12-hour + AM/PM) in the digital clock that’s on the panel.
2. Some KDE applications don’t respect my chosen widget theme (Plastik). This could be some strange Kubuntu packaging issue, but I somehow doubt it.
3. There’s no “Run” option in the classic K menu (nor does there seem to be any way to customize it).
4. There are now two “theme-able” engines in KDE 4: Plasma and Qt. Call me a traditionalist, but I like my desktop to look somewhat consistent, and that’s hard to do when some things on the desktop follow one theme, and the others follow another. Blessings on the person that created the “Plain” Plasma theme that’s on kde-look.org, as it matches my Plastik Qt theme pretty well.
5. I’m all for desktop widgets, but I don’t want my ENTIRE DESKTOP to be built from widgets. Let’s take an example of how the widget-built desktop concept could be annoying (at least in it’s current implementation). The KDE 4 panel is a plasma widget, as are all of the items on the panel (the K menu, the task list, the system tray, the digital clock, etc.). Now, let’s say I want to “lock” the items that are currently on the dock, and I want to get rid of the annoying “cashew” that’s on the far right of my dock. I right-click the dock, and choose “Lock Widgets.” Does it lock my dock? Yes. But, it also locks all the other widgets that I happen to have on my desktop. Bad.
6. I like desktop icons. The desktop is a place that I can stick things that I’m working on currently without having to dig around in an endless hierarchy of folders to get to. If you take away my desktop icons, you reduce my ability to work efficiently. The folderview Plasmoid is a nice, quick fix to the lack of working desktop icons from the KDE 4.0 release, but it’s not a perfect solution. From what I read, KDE 4.2 will allow you to fill your whole desktop with a folderview widget, but you’ll lose the ability to place other widgets on your desktop. That doesn’t sound good either. KDE Developers: it’s time to re-think (not throw away) this widget strategy before you drive away your entire user base.
7. When you try to resize plasmoids, they stretch and skew very slowly and painfully. Just moving them around can be a chore.
8. Dolphin is getting better as a file manager, but it’s still missing some things that I like from Konqueror; perhaps others feel the same way. However, there’s no intuitive way to change my default filemanager back to konqueror.
9. Things just don’t seem to render smoothly. When I see windows and menus appear, I see fragments of things that were on the screen previously for a split-second, and then the item that I requested renders. I’ll see if this is nvidia+kde4 related when I install Kubuntu 8.10 on my Thinkpad.
10. There are obvious user options missing all over the place. I detailed a few missing options earlier, but this is more of a general annoyance, as it affects the way I work all the time. For example, do I want to group similar tasks in the panel’s task list? Maybe, or maybe not, but I should have an option to turn it on and off. I could go on all night about missing user preference things, but I think you get the point.

So, there it is. My most up-to-date list of grievance. Now, I know there’ll probably be a KDE developer out there that reads this and says: “if you don’t like it, fix it and submit a patch.” People in the open source community like to think that it’s easy for anyone to get involved in an open source project, and for some projects that’s true. However, in the case of KDE, the community has been neglected as of late, and I’m starting to wonder if it’s really a community effort anymore. Maybe that’s a separate topic for another post, and maybe I’m just bitter over the path that KDE 4 has taken over the last several years, but I think many of the things I’ve chronicled in this blog post have some value, if only to help me gague how KDE 4 is improving.

• Alpha Release -> Not feature complete; a taste of what the final product has in store; perhaps (very) buggy
• Beta Release -> Feature complete, but still needs polish (and bug-fixing); stable enough for normal users to test (but not use in production)
• Release Candidate -> Feature complete with all (known) major bugs fixed. This could be a final release, if no one finds any major problems
• Final Release -> Everything that the users have been salivating over, and MORE!

I hate to say it, but KDE 4.0.0 seems to fit in what we’ve traditionally classified as an “alpha” release. I’m not saying that because it’s unstable; KDE 4.0.0 is stable for the most part (still can’t get Konsole to start up without crashing). The reason I’d call it “alpha” is that it’s not feature complete.

For example, the first time I logged into KDE 4.0.0, I noticed the panel…OK, well who couldn’t notice the panel? It’s rather large with a massive digital clock on the right side. I mean, a blind man with two missing fingers would notice the panel right off the bat. So, since I don’t like my panel to be so monstrous, I right-clicked the panel and chose “Task Manager Settings.” A configuration dialog popped up with (wait for it)…ONE OPTION!! The one option was weather I wanted tool-tips or not! So, I thought maybe the panel will size down automatically if I reduce the font size of the digital clock. So, I right-click the clock and chose “Digital Clock Settings.” In that dialog I saw options for choosing a font, weather it should be bold or italic, the clock color, weather or not to show the timezone/year/day-of-the-week, and weather to show in 24-hour format, but NO FONT SIZE OPTION!!!

You might say: “Well, Jared, perhaps the people working on the panel just were having a bad year or two…why not give the rest of the release a chance?” Oh, my friend, I have indeed given the rest of the desktop a chance. I set aside my experience with the panel so that I could give the rest of the Desktop my full, undistracted attention. Next, I turned my attention to the icons I saw on the desktop. These icons were the ones that I had there while running KDE 3.5 (the contents of the “Desktop” folder in my home directory). The first thing I noticed was that they weren’t lined up nicely; they were kind of there in no particular order, at least none that I could determine. Odd perhaps, but something I could once again overlook.

Each icon had a translucent backdrop with rounded edges: “A nice touch” I thought to myself. One thing that troubled me, however, was that the icons were on the other monitor from the panel. I have gotten used to having both on the same monitor, so I tried to move them. I tried to select multiple icons and drag them to the other monitor. Much to my surprise, I could only move ONE icon at a time. So, I resigned my self to just move them one-by-one. After trying to move the first icon, I found that once the icon left the screen that it was on, it became invisible on the other monitor. After playing around with it a bit, I suspect that the desktop background on the second monitor was being rendered over top of the icon. But this isn’t the worst of this part of the story: after trying to move the icon back THE ENTIRE DESKTOP STARTED TO MOVE!!!!11111one!

Now that I’ve given you a good horror story of KDE 4.0.0, let’s get back to the topic that I started with. Aaron Seigo, who seems to be the voice of the KDE team these days, has been on the war path of KDE 4.0.0 these days. He has been saying that people aren’t expecting the right thing when they expect KDE 4.0 to be a full replacement for KDE 3.5. They shouldn’t expect that until later releases of KDE 4. He says that people are used to the “closed source” method of software releases. He says that Open Source has generally worked different than closed source software. I agree with Aaron, but only to the point that open source projects are _usually_ a LOT more careful about what they label as a final/stable release. They release early and/or often, sure, but they usually do so with labels like “alpha” or “beta” so that would-be users know that it’s not what they might expect from a final/stable release.

I think that KDE 4.0.0 was released so that the KDE developers could get back into heavy development without users constantly clamoring about when KDE 4 was coming out. It was a bone to those of us in the community who have been waiting for years (literally) for KDE 4. Well, I must say, I’m not impressed with 4.0.0. It has potential, but potential never won any awards, and it certainly isn’t what should constitute a final/stable release of a major Linux Desktop contender.

Now, I wouldn’t be writing this post if I didn’t have an even better work-around. The better work around is to add the following to the /usr/bin/compiz startup script:

Replace this line:
${COMPIZ_BIN_PATH}${COMPIZ_NAME} $COMPIZ_OPTIONS "$@" $COMPIZ_PLUGINS || exec $FALLBACKWM $FALLBACKWM_OPTIONS

… with the following:

${COMPIZ_BIN_PATH}${COMPIZ_NAME} $COMPIZ_OPTIONS "$@" $COMPIZ_PLUGINS &

sleep 2s

if [ $? -ne 0 ]; then
  exec $FALLBACKWM $FALLBACKWM_OPTIONS
fi

sleep 3s
gconftool-2 --type float --set /apps/compiz/plugins/decoration/allscreens/options/shadow_radius 6.0
sleep 1s
gconftool-2 --type float --set /apps/compiz/plugins/decoration/allscreens/options/shadow_radius 6.1

After this, the Shadow Radius adjustment will be made a few seconds after Compiz starts, and you will be on your way to a better Compiz experience.

I’ve been using Amarok solidly ever since, but recently I’ve been itching to return to my Winamp roots in usability/simplicity.  I got this itch again because I saw an article about the release of Xmms 1.2.11, the first update to the aging media player in over 3 years.  As far as I know, it’s simply a point release with bugfixes only (no GTK2 conversion, sadly).  While looking into this, I inadvertently stumbled upon Audacious, a fork of Xmms that I had seen a few years back, but it was as unstable as BMPx was at the time I tried it.

I looked at a few of the screenshots of Audacious, and it looked very promising: a media player forked from Xmms, which supported Winamp skins, and the GUI was upgraded to GTK2.  So, I downloaded and installed.  I was immediately able to pull out my old Winamp 2.95 Classic skin, make a few adjustments to the default settings, and I was in heaven.  The simplicity was wonderful.  I don’t need a media library (my storage server is quite organized already), and it even supported my Microsoft (gasp!) keyboard’s multimedia keys with an included plugin.  I loved it!

So, the moral of the story is, don’t give up on Linux.  If you want something, chances are that someone out there wants the same thing, and all you really have to do is wait (or, perhaps write/fork it yourself, if you feel up to it), and it’ll probably show up.

There were some shorter experiments in there too. I gave Gentoo a try twice. The first time, I was not able to get the system installed. I tried again about a year later, and I was able to get a system installed this time, but I wasn’t happy with the portage system. I know that sounds like blasphemy to those that swear by Gentoo, but it just wasn’t for me.

I eventually settled on K/Ubuntu for my preferred distro. I like it because I don’t have to go all over creation to get certain things working correctly (e.g., Intel wireless drivers), but I still have the power through APT to control exactly what’s installed on my system. Add to that the 18,000+ packages available in the various Ubuntu repositories, and I have everything I need.

Up to this point, it has been hard to give back to the community that I have gotten so much from. Sure, I filed or commented-on the occasional bug report, but I was never able to give back anything tangible. But then I realized one of the things that probably is very hard for Canonical (the corporate presence behind Ubuntu) is shelling out money for download bandwidth. If you think about the thousands, possible millions, of people that download Ubuntu with each release, that could be a significant chunk of change to support.

So, I’ve decided that this time when downloading the latest install ISO, I would use the Bit Torrent method, and I would make sure to leave my Torrent client open to upload at least 150% of what I downloaded. This act is something any user can do to help Canonical defray the cost of providing what I think is the best mix of power with easy-of-use features in a Linux-based OS. In fact, I came home tonight and found that I’d already uploaded 170%, and I’ve decided to just leave my client uploading for at least a few more days to make sure there are enough seeds for those that still want to get ISOs in this manner.

Now, I don’t think that people should be given the right to copy software and give/sell it to whom they choose, but they should not be made to give up their law-given rights (see 17 USC 117) just to use the software they have purchased.  This attempt by Apple isn’t the only example of big software companies trying to control the users by “contract.”  Take Microsoft, for example.  They have stipulated in the Windows Vista Home Basic license that users are not allowed to run the software inside of a emulator/virtualizer.  This is simply outrageous, as Microsoft has absolutely no right to dictate how a user may run the software they have purchased.  Even worse, Microsoft also stipulates that the software may only be re-sold once after the initial sale (so, if you bought Windows off the store shelf, you may sell it to one person, however, he may not sell it).

The validity of these types of stipulations have not been tested in a court as far as I know.  However, this Wikipedia article on software license agreements links to several cases where other license issues have been tested.  Every user of computer software should be aware of these issues, and make themselves aware of the intricacies of the concepts involved.

On the Linux front, I’ve got Xgl/Compiz working on Ubuntu Dapper. This helps me do wonderful GUI effects on my favorite OS. I did with the help of my GeForceFX 5200 (and excellent binary drivers, care of nVidia). I wanted to set this up on my Thinkpad T40 as well, however, Xgl currently doesn’t support the open source ATI driver that my Thinkpad’s ATI Mobility 7500 uses. However, AIGLX (an acceleration architecture similar to Xgl) does support that driver, and Compiz (the compositing manager I’m using) works just as well on AIGLX as it does on Xgl. So, I got it working there as well.

I’ve also been told, at my place of employment, that they’ll be changing my job roll slightly. I’ll be going from PC tech support to Linux/UNIX server management. This, to me, is an wonderful change. I really wanted something like this to happen for a long time, but I never felt like I had enough real qualifications to ask to work in that roll. I still feel like that to a certain degree, but hopefully I’ll be able to learn a lot more than I know right now.

Anyway, that’s my life in a nutshell right now. More to come when I know more.