I'll probably follow this up with more coherent and reasoned thoughts when I get some free time and am better rested, but for now I just felt the need to share my feelings. That is all.

One thing I love about KDE is it's incredible breadth of configuration options. I really like to tweak my environment to best suite my needs, preferences, and habits. I know what works best for me, and prefer to have my desktop environment reflect those preferences. Despite the vast number of preferences in the KDE Control Center, there are still quite a few options for which new GUI preference setting exists. The Hidden Configuration KDE Wiki page discusses some of these options, and is worth a read if you use KDE. In particular, I was looking for a way to disable the listing and expanding of archive files in Konqueror's sidebar. This "feature" was borrowed from Windows XP (Compressed Folders), where it always bugged the hell out of me. If I wanted view the contents of Zip files in Windows Explorer then I'd unzip the damn file. Needless to say, I was rather dismayed and disappointed when I saw this "feature" appear in a KDE upgrade. There is no GUI preference available for disabling it, but after quite a bit of internet searching I found the above Hidden Configuration page, which discusses how to do it. It's a useful resource, and I wanted to make a note of it here both for the benefit of others as well as so I can easily find the page again next time I setup KDE. :-) If you would like to disable this feature as well, first try entering this command (as documented in the Wiki): kwriteconfig --file konqsidebartng.rc --group General --key ShowArchivesAsFolders --type bool false. You'll need to restart Konqueror for the change to take effect. If that does not work (it didn't for me), do this instead:

1. Edit ~/.kde/share/config/konqsidebartng.rc
2. Search for the option titled ShowArchivesAsFolders
3. If you ran the kwriteconfig command, you should find it under the [General] category. Delete that under [General] and instead add ShowArchivesAsFolders=false it to the top of the file
4. If you do not already have that setting in the file, simply add it to the top of the file as described in the last step
5. Save the file and restart konqueror

You should now be rid of those annoying archive folders. Enjoy. Oh, and if you'd like to disable this feature in Windows XP as well, you can easily to so by running the following command: regsvr32.exe /u zipfldr.dll. If you choose to reenable it, you can do so simply by running regsvr32.exe zipfldr.dll.

I came across a couple of very useful Windows apps tonight while doing some maintenance on my systems. The first is Core Mini-SFTP Server. This is a commercial/proprietary app, but it's available free of charge. As implied by the name, it's a stripped-down sFTP server for Windows. No installation or configuration is necessary; simply download and run the executable, specify the username, password, and root directory, then click Start. Any user can now connect via SFTP using the specified credentials. It's very convenient if you simply need quick and easy access to an sFTP server on Windows, but, of course, it does have limitations. It's strictly single user, must be run interactively (ie, it cannot be run as a service when the system starts), and only minimal sftp functionality is included (the sftp client under Linux works, for example, but scp does not). Additionally, it stores the specified password in plaintext within the registry. Keep this in mind when choosing a password, and be sure to delete the key after you're finished if it's a sensitive password (HKCU\Software\FTPWare\msftpsrvr\msftpsrvr). Next up is a fine new FOSS app for Windows. Infra Recorder is a very slick CD burning application based on cdrecord. The interface is very nice and intuitive, functionally it can do just about anything you'd expect of a CD burning application, and so far it seems quite stable (considering it's a beta release). I'm quite pleased with it so far. The audio capabilities are somewhat limited (it can only handle WAV files directly, for example), but given that I use Exact Audio Copy for all of my audio CD needs it's not much of an issue for me. It'll make a great alternative to cdrtfe, my current burning app of choice under Windows. Enjoy. :-) Edit: I'm afraid I'm going to have to take back some of the praise for Infra Recorder. It doesn't seem to actually want to write the disc image that you tell it to burn. Instead, it just pretends to burn it for several minutes, letting you think it's being written to disc. I discovered this after thinking I had burned a freshly downloaded 700 MB Kubuntu ISO, only to find out after I had deleted ISO that it had not, in fact, been written to disc. So, I downloaded it again, checked and double-checked all settings (especially the "simulation" option, and attempted to burn it again, but it still failed. I then fired up cdrtfe and burned it without problem on the first attempt, confirming that the disc image was fine. I'd recommend sticking with cdrtfe for important stuff for now.

One thing I always liked about Windows (compared to Linux) is that it's very easy to add custom actions to the context (right-click) menu for any given file types. For example, I used this ability with Universal Extractor to add UniExtract... entries to the context menu of archive files, and I use it with Open with Arguments to add Open with arguments... to .exe and .bat files. I missed that ability for quite some time once I began using Linux as my primary OS. Something as simple as extracting Zip files, for example, would require jumping to the command line and entering an appropriate unzip command^[1]. However, a while back I stumbled across a tutorial entitled, "Creating Konqueror Service Menus", and was very pleasantly surprised to discover that this allowed me to do exactly what I had wanted for so long.

I setup a few custom actions (called "servicemenus" in KDE) a while back on my home system and pretty much forgot about it since it "just worked", but since I'm now using a new desktop system at home I'm already missing these custom actions. So, I figured I'd document them here while setting them up again. Hopefully this information will help out other Linux users. Much more thorough instructions can be found in the article referenced above - my instructions should be treated as more of a reference.

To begin, you'll need to create a new .desktop file for the action you want to perform. For the purposes of this article, I'm going to add a context menu item that will extract RAR files to the current directory. So, we'll create a new file named ~/.kde/share/apps/konqueror/servicemenus/rar.desktop. The file name is arbitrary, but it must be saved in the specified location, and must end with the .desktop extension. Next open the file in your favorite editor and add the following:

rar.desktop

[Desktop Entry]
ServiceTypes=application/x-rar,application/x-rar-compressed
Actions=unrar

[Desktop Action unrar]
Name=Extract Here
Exec=launch.sh %d unrar x \"%f\"
Icon=package

This code is not very intuitive, so I'll explain each option

• ServiceTypes - specifies the type of files with which the action should be associated. The easiest way to determine this information is to run Konqueror, click Settings, Configure Konqueror, and select the File Associations section. Enter the file extension you want to associate the action with (in this case, rar, and then add the listed file types to the Service Types entry. Repeat for each extension if you want to associate with multiple types
• Actions - specifies the name of the stanza that defines the action. Multiple actions can be specified, but we'll only use one here. Just make sure that the name entered here matches the [Desktop Action xxx] defined below.
• Name - the name of the context menu entry that will appear when right-clicking on the given type of files
• Exec - the action to perform when selected; more details below. Please also see this page for a full discussion of this item, including a list of valid field codes.
• Icon - the name of the icon to associate with the context menu entry (optional). This can point to a real file if you want to use a custom icon, but you have to specify the full path and filename. In this case, I'm telling it to use the package icon from the current icon set. The easiest way (that I know of) to view these "pre-defined" icons is to right-click on any K-menu entry, select Edit Item, and click on the icon button for that item, It'll bring up an icon browser. Find the icon you like best, note the name, then close the windows and add it to the Icon entry.

Now, let's discuss the Exec entry. Ordinarily you'd probably want to call the binary directly; eg., unrar x \"%f\". In this case, however, I want to get feedback on the current progress of the operation, as well as any errors that might have occured. Since unrar is a CLI application, running it from a GUI wouldn't provide any feedback. It would simply run in the background and then exit. To work around this, I created a "wrapper" script called launch.sh that will accept arguments passed by KDE and run the command in a standalone xterm terminal^[2]. Using this method, clicking the the action in the context menu will spawn a new xterm window, which will then display the current status of the operation. It will also allow you to enter any additional information that may be necessary, such as answering an overwrite prompt or providing an archive password.

The code for the wrapper script is listed below. The only dependency is that xterm must be installed in an your $PATH.

launch.sh

#!/bin/bash

# enable support for spaces
IFS=$'\r\n'

# check for number of arguments
if [ "$2" = "" ]; then
    echo "Usage: $0 <dir> <command>"
    exit 1
fi

# set directory and command
DIR=$1
shift
COM=$@

# execute command in xterm
cd $DIR
xterm -e $COM
exit

That should do it. Save both of those files, make sure that launch.sh is copied to a location in your $PATH, then try right-clicking on a RAR file. Under the Actions submenu, you should now see an entry called Extract Here. Click it, and if all goes well the contents of the RAR file should be extracted to that directory.

For reference, here's a list of all KDE servicemenus that I have created:

• audacious.desktop - Enqueue and begin playing all selected audio files in Audacious (originally written for XMMS, and still contains the commented code if desired)
• iso.desktop - Mount an ISO CD-Rom image in a subdirectory of the current folder to allow file browsing and copying; press Enter when complete to unmount the ISO and remove the temporary directory. This service menu requires my mountiso.sh script.
• par.desktop - Repair damaged RAR archives using associated PAR files
• rar.desktop - Extract contents of RAR archives
• tbz.desktop - Extract contents of bzipped tarballs
• tgz.desktop - Extract contents of gzipped tarballs
• vmdk.desktop - Mount a VMware disk image in a subdirectory of the current folder to allow file browsing and copying; press Ctrl-C when complete to unmount the disk image and remove the temporary directory. This service menu requires my mountvmdk.sh script. More details can be found in the How to Mount VMware Disk Images under Linux article.
• xine.desktop - Enque and begin playing all selected video files in Xine
• zip.desktop - Extract contents of ZIP archives
• launch.sh - Wrapper script to display service menu output in an xterm window; most of my servicemenus require this script

^[1] Yes, I know that I can install a GUI archiving utility such as Ark. However, that's not really relevant here for two reasons:

1. I want to right-click and extract directly within Konqueror without first opening it in a separate utility
2. File extraction is just an easy-to-visualize simple example - there are other cases where install a separate utility is not an option or just doesn't make any sense

^[2] Yes, you could theoretically call xterm directly from the .desktop file rather than using a wrapper script, but I couldn't get it to work properly. I had issues with getting xterm and the associated command (in this case, unrar) to accept the correct path, as well as dealing with spaces in the filename. My wrapper script will handle anything that's thrown at it (so far, anyway...).

If it seems that I haven't spent much time working on my website recently, well, I haven't. :-) A whole lot has been keeping my busy for the last few weeks, including:

• Accepting and preparing for a new job
• Tying up loose ends at my old job before leaving
• Purchasing/assembling/installing Gentoo on a new desktop system (which takes a while to work out all of the kinks)
• Shopping/purchasing a new laptop since I lost my work-provided system

That's the highlights, but I've been dealing with some other stuff as well. I have some projects I'm trying to get done before I start my new job on Monday, and of course once I do begin my new job I'm sure it'll keep my busy for a while. So, to be honest I don't know when I'm going to be able to start posting regular updates again, but hopefully it won't be too far off. One important task I'd like to finish up is putting out an update for Universal Extractor 1.5.1. I had mostly completed it about a month ago, but just haven't had time to fix a couple remaining issues and get the updated translation files. I'd like to start working on that again this weekend if I can finish my other projects in time, so keep an eye out for it in the next week or two.

For my third and last post in this series, I'd like to discuss overall site management design changes. Prior to this latest change, I had always run LegRoom off of PostNuke. Now, PostNuke has been good to me over the years. It's been around for a while so it was a pretty mature product even in 2002, it has a huge community behind it, and it's been flexible enough to let me do pretty much anything I wanted during the previous couple redesigns. I'm very appreciative of all the hard work that the PostNuke devs and community have put into the product, and I certainly do no regret choosing PostNuke for my site. With the latest redesign, however, I felt the need for something different. I could've just slapped a new theme on top of my PostNuke install, as I did previously, but I really wanted to migrate to a new content management system altogether to give me a chance to truly redesign the site from the ground up, as well as clean out a lot of the cruft that had been gathered over the years (see Part 2 for some examples of this). Additionally, while PostNuke was a capable and mature CMS, I wanted to move away from it for three main reasons:

• Development progress seems almost non-existent. Take version 0.8, for example. If I recall correctly, initial development of 0.8 began in 2003, possibly even 2002. However, version 0.8 has still not been released. Now, I'm quite sure there are a lot of factors contributing to this delay, and I don't even pretend to know all of the facts, but just from a pure end-user perspective this is ridiculous. There have been minor updates to the 0.7x branch during that time, but to go so long without a major release gives the impression that either development is stalled or non-existent, there are severe technical difficulties involved (which can shake confidence in the developers involved), or there are severe personnel and/or communication difficulties within the developer community (which, again, can shake confidence). Personally, I just got tired of waiting.
• PostNuke is a heavy system, and the generated output (at least from my older version) was just plain ugly. I wanted a CMS that produces cleaner, more efficient, and standards-compliant code, something that doesn't use several levels of nested tables for positioning. From what I read, this situation has supposedly improved significantly in the current PostNuke releases, but that doesn't help me much because of the next reason.
• In order to make PostNuke work how I wanted, I had to make some fairly extensive modifications to the codebase. I completely rewrote the menu generation code and RSS publication module, for example, as well as made various changes here and there to several of the other modules. The problem with this approach is that it makes an in-place upgrade nearly impossible. The end-result is that I was left running an extremely vulnerable version of PostNuke for several years. I was honestly surprised that I was able to hold off the hackers until I was able to complete the Drupal migration. Now, this isn't PostNuke's fault in any way, but simply another factor that has to be considered. I need a CMS that will do what I want without requiring modification of the codebase. PostNuke couldn't provide that.

So, after a rather extensive search, I settled on Drupal. As of version 5.0 it seems to offer the best combination of capability, flexibility, efficiency, and standards compliance out of all of the open source CMSes that I examined. (By the way, I'd really like to thank the admins of OpenSourceCMS for making it easy to "test drive" so many website management systems. If you're a webmaster that's not familiar with this site, check it out ASAP.) So, aside from the CMS change, what else is new? While migrating all web content over to the new site I spent a lot of time "updating" all content to use a specific look and feel. My previous site was something of a testing ground for me, and was originally started when I just didn't know much beyond pure HTML. Each page that I added to the site was essentially created using whatever level of experience I had mastered at the time, resulting in a hodgepodge of styles and techniques. This is especially true of the Tips and Tricks pages, of which some had to be nearly completely rewritten. Now, however, I was able to apply the same coding styles uniformly across all pages on the site. Yay! In addition to the common style, you may also notice a common layout for all of the pages. Each has a navbar across the top that will take you to any location in the page. Each page is broken up into the same sections, where appropriate, for consistency and easy of use. I also added section breaks, along with "return to top" links, to cleanly separate each section. These are a lot of subtle changes, to be sure, but they really do a lot to enhance site usability. Other page-specific changes:

• The Bookmarks page is now properly styles to match the rest of the site (thanks again, Steve)
• As previously mentioned, I added a new Coming Soon section
• DailyStrips has also been restyled to match the site, including rewriting the output engine to work better in the context of a website module (and no more tables!)
• The Metasearch page (aka, Search Internet) has been tweaked and has had a couple more sites added

I think that pretty much covers it. I hope you enjoyed this brief look into the redesign process for this site. Up next - the conversion script I used to migrate from PostNuke to Drupal. It is truly one of the most ugly pieces of code I've ever written, but it got the job done. As promised, I'll make it available to everyone else to use, along with an explanation of the details and shortcomings of the script. I just need a bit more time to clean it up and write the details.

Streaming (aka, embedded) video has been around since the beginning of the internet, and while it can certainly be a very useful technology, I hate companies and websites that require you to stream the video and don't even offer an option to download it. What if you want to view it multiple times? What if you want to show a friend or co-worker? Tough! You have to stream it (re-download it) every single time you watch it. The only possibly reason for doing this that I can think of is that it supposedly allows the hosting site more control over the video, but it creates a severe inconvenience for users, prevents a number of possible customers from watching it in the first place (eg, if they're not running a "blessed" operating system, browser, or plugin), and exponentially increases bandwidth costs for the hosting provider. Additionally, the very fact that the movie has to be sent to the client to allow it to be played and displayed means that the client still gets a copy and can still save it anyway, just with much greater hassle. If you want to save a streaming video from a site that doesn't offer a proper download link, you have a few options. Like most things in technology these range from easy but limited to difficult but extremely flexible. For now, I'm going to discuss the easy approach. :-) I'm also going to assume you're using Mozilla Firefox (if you're not, you really should be). There are a number of Firefox extensions available that can greatly assist with saving embedded/streaming video clips. A simple search for "download video" on the Firefox Add-ons returns 19 extensions. I've personally used VideoDownloader in the past, and it worked well. If you don't want to install a new extension just for downloading videos, and option is to use Greasemonkey scripts. Greasemonkey is a really powerful extensions that allows users to create or install custom scripts that can change the behavior of any web page. In this case, a Greasemonkey script can analyze the page for embedded video clips, then automatically add a download link to that video; simply click the download link to save the video. Two great scripts for downloading embedded videos are Apple Trailer Download and Download Video (though I use this version of Download Video). Note that you must first install Greasemonkey before you can install these scripts. You can find a lot more useful Greasemonkey scripts on my Mozilla Firefox Tips and Tricks page, as well as a massive collection of contributed scripts on Userscripts.org. The primary limitation with the extension/script method is that it's often limited to particular sites. Apple Trailer Download, for example, is limited to just movie trailers hosted on the Apple Movie Trailers website. Download Video and VideoDownloader both support a number of different sites (including the most popular, YouTube and Google Video), but are still limited to only sites they "know" about. I'll follow up this article with more advanced techniques and suggestions that should help you save just about any embedded video.

Someone with too much free time on his hands (not that that's a bad thing) collected a number of scanned test/quiz documents with some rather creative answers. This should make for a rather humorous end to your day (as well as feeling better about your own test grades from school). Remember, there is an elephant in the way.

I've added several new language files for Universal Extractor since the 1.5 release, including (as of this post): Korean, Portuguese, Portuguese (Brazilian), Romanian, and Turkish. These files can be downloaded from the Universal Extractor page and added to your current copy of Universal Extractor if you'd like, and they will be included in the next release. I'd also like to point out that all 18 (and counting!) language files have been voluntarily translated and contributed by UniExtract users that simply wanted to use the program in their native language. I'd like to send out a big thanks to all contributors for improving Universal Extractor by making it much more accessible to foreign users. Thanks! If a translation doesn't currently exist for your own language, and you have a good understanding of English, consider contributing a translation yourself. The process itself is very straightforward; simply copy English.ini and follow the instructions at the top of the file to translate the English strings into your native language. If you're gracious enough to contribute the new language file for all other UniExtract users, I'll be happy to add your name to the list of contributors, as well as link back to your own website (if you'd like). The only "catch" is that I'd ask you to keep the file updated as I release newer versions of Universal Extractor. These updates typically involve only minor changes or additions, and are only necessary as often as I release new versions (which, honestly, is not that often). I can only speak English myself, so I rely entirely on the generosity of UniExtract users to maintain the language files. Thanks again to everyone that has contributed a language file to Universal Extractor.

Several people have e-mailed me regarding a couple bugs in my programs, as well as to submit new translation files for Universal Extractor. I'm behind on a lot of my e-mail right now due to some other priorities, so please forgive the delay in response. I'm reading each and every e-mail that comes my way, and will reply as soon as I get the chance. For everyone else that may have found a bug or wants to suggest a new feature, please consider using the new LegRoom Support Forum instead. I won't necessarily be able to reply any faster, but the information you provide will at least be visible to all other users, which could help them if they're having a similar problem or were thinking about requesting a similar feature. Plus, it helps cut down on my e-mail. :-) I'm going to try to knock out a few UniExtract-related e-mails before going to bed tonight.