Well, things are looking up. On a whim, I re-implemented most of the old gender-bending list’s functionality on the new database-backed platform I’ve been meaning to write for years.

I’m not sure how steadily I’ll work on getting the rest done, but at least it’s somewhat usable now. (I still have to hook in display of and filtering by types of gender-bending and the sorting can be a bit finicky, but it is possible to browse by category, see the degree of gender-bending in each work, and click them to see what used to be footnotes.)

The good news is, I’ve finally cooked up a tool which prevents me from procrastinating or getting distracted for days on end, so it’s only a matter of time before I’ve caught up to my real-life obligations.

The main site should be renovated and back online some time within the next month or two and, while I don’t want to promise anything, I’ll probably resume reviewing fanfics and commenting on interesting geeky stuff shortly afterwards.

Just a couple more fun lists for people who enjoyed If Version Control Systems Were Airlines and Some more geek humor.

Have you ever tried looking up how to use Javascript to get the selected HTML in a browser window in a form that can then be sent via XMLHTTPRequest or fed into an ordinary form field? (eg. for POSTing to a quote-collector tool) It’s a surprisingly difficult piece of info to find. (Hence why I’m partly writing this as a reminder to myself)

First, you’ll probably end up visiting the QuirksMode Range Intro to learn about window.getSelection() in browsers like Gecko/Firefox and the equivalent Internet Explorer scripting, but all you’ll learn there is how to get the selection as plain text… and that’s no secret.

Once you’ve got the selection/range, things become a little trickier because, once again, Internet Explorer and everyone else do things differently. In Internet Explorer, it’s as simple as range.htmlText but in other browsers, it’s a little less obvious.

First, you call range.cloneContents() to get a DocumentFragment object. This is more or less equivalent to Copy (Ctrl+C) when copying and pasting things. Then you create a <div> element, use div.appendChild(clonedSelection), and then grab the HTML as text from div.innerHTML. Don’t go looking for ways to serialize DOM nodes to XML. It’s non-portable and a bit of a red herring.

Here’s the complete code fragment I eventually found on a thread on the FCKEditor forums:

Now that you know, why not create some interesting bookmarklets?

Probably best to read the link above, since I definitely don’t have time to write something anywhere near as good, but in short, it’s an introduction to Dana Hunter’s Carnival of Elitist Bastards which attempts to combat (or at least provide some resistance to) the apparent American tendency towards anti-intellectualism and a tendency to disregard what the experts say if it contradicts what people want to be true.

As I mentioned, the above introduction is much better than this little blurb and I highly recommend that you read it.

Not only do you get to use the VoIP client and high-quality $7 Skype-certified, noise-cancelling microphone of your choice, you can trap telemarketers in a voice-mail maze of your own fiendish design.

In essence, an FXO card is the “voice” portion of a voice modem isolated and made Linux compatible so Asterisk users will buy them. As such, they’re cheaper than an equivalent voice modem and, since voice is the core function rather than an afterthought, their noise/echo-cancellation is usually better.

Oh, and if you want to plug traditional telephones into VoIP… perhaps to use Asterisk as a full-fledged PBX, you’ll want an FXS card (internal) or an ATA (external, Analog Telephony Adaptor). (FXO cards act as phones, FXS cards act as phone lines)

• A faint background image in your terminal reduces eye-strain. (source)
• KDE users can use kioexec to pass URLs as command-line arguments to non-KDE apps. (source)
• set ttymouse=xterm2 will fix mouse support in Vim when you’re running it inside GNU Screen. (source)

Did you ever wonder what arcane incantations went into customizing that “ActionList: viewmode_toolbar” subsection in your Konqueror toolbar? Wonder no more.

First, anyone who has used both Kubuntu and another distro (eg. Gentoo) will notice that it’s possible to change the number of dropdown buttons. I’m happy with the three I have, so I didn’t really look into this very closely, but apparently this patch file will switch between the two configurations when properly applied:

http://launchpadlibrarian.net/4891395/un84.diff

Now that the difficult part is out of the way, here’s the part that is so typically “KDE voodoo”. Easy to do… but you have to know it’s there first. (Which is still better than GNOME in my opinion, but that’s just me)

If you want to change the order of the buttons, remove (or re-add) entries (but not change which ones go into which buttons), or re-order the entries within each button (Keep in mind that this also lets you control which one gets the ‘default for this button’ status), here’s what you do:

1. Go into the file association controls (Configure Konqueror –> File Associations is a nice quick way)
2. Choose inode/directory and click the Embedding tab
3. Re-order, add, and delete KParts.

It’s that simple. My suggestion is to install FileLight and then make it the top-most item in the right-most button with the traditional disk-usage viewer the bottom-most one.

Here are the relevant pages I ran across while googling. The rest was just experimentation on a hunch.

• https://bugs.launchpad.net/ubuntu/+source/kdebase/+bug/43949
• http://strabes.wordpress.com/2007/04/05/make-konqueror-use-your-default-view-mode-in-new-tabs/

Just a blog post I ran across with some excellent tips for evaluating how difficult it will be to learn a given language. Not necessarily something I’d use (since I’d pick a language based on what it would get me and then soldier on, rather than learning languages based on how easy it’ll be) but something to keep in mind nonetheless.

I just pushed my attempts at RAR and GIF handling in Python to my useful hacks page.

The GIF one functions as a partial validator, but it’s main designed purpose was to tell static from animated GIFs without a ton of dependencies.

The RAR one is only really useful for identifying RAR files by header and listing their contents so far, but I eventually hope to add support for extracting files which were stored without compression. (Usually video files, as the docstring says)

Enjoy.