A couple of weeks ago, I stumbled over the CHDK project where some fine people develop additional firmware for a wide range of Canon digital cameras. Since I call a Canon IXUS 800 IS - also known as Canon SD700 in other parts of the world - my own, I was instantly captivated by the amount of additional features that can be added by means of loading an small piece of software onto my cam. Yesterday, I tried to get it to work and was set up within a couple of minutes.
Some of the new choices I am given are that I can now save pictures in RAW image format (one of the reasons I got myself a Canon EOS350D on ebay a couple of weeks ago), a live histogram appears on the display, the state of charge of my battery is shown, and I can download and apply a large number of user scripts that combine batches of actions and permit to trigger them with a single button. And the best thing: The original firmware does not get overwritten!
Ever since I installed Rockbox on a Sansa digital audio player, I'm a big fan of user-contributed, open source alternative firmware. And this one is no exception, it's simply a glorious extension to my digicam. Let me show you a couple of pics of CHDK's new menu and display options:
I just installed the recent beta 4 release of Firefox 3 (Firefox Portable, to be exact, thus making it run alongside my trusty Firefox 2). Firefox 3 is blazing fast, but this may be due to the fact that I've got almost no add-ons installed. My Firefox 2 has to carry the burden of 27 active extensions that may or may not slow down the whole applicaton, who knows. All in all, Firefox 3 seems to be very stable already, but I need my add-ons for web development and so on. That's why I've been looking for other things to do with the new beta. When I found out that Mozilla Labs released a new version of Prism, their stand-alone XULRunner applicication slash Site Specific Browser. If you install Prism 0.2 as add-on to Firefox 3, it allows you to create and export new applicatons from within the browser. The new apps then use the browser's rendering enginge, it seems (I'm not exactly sure how it works). Yet, again, the resulting app feels considerably faster than the old Prism. I'm using an installation of tt-rss on my server as web based feed reader. Now that it's running as a Prism app, it seems to respond so much faster than before - which reduces the amount of time I need to read all my feeds. Great!
Today, I got myself an external USB sound card and tried to install it on my notebook running Ubuntu Studio. It's a Trust SC-5500p because that's the only affordable model my retailer had in stock. To cut a long story short: I didn't work as I hoped it would. I could get heavily distorted sound out of its front speaker output jack, but the rear as well as the center outputs remained silent. I thought the days of insufficient support for hardware on Linux were over, but: FAIL! Since I swore to myself that I wouldn't spent hours and hours an installing packages and hacking config files anymore, I returned the device, and I will start to look for another affordable solution. Suggestions anyone?
I've just learned about SPARQLBot via this posting by Manu Sporny to the uf-discuss mailing list. He writes
This has to be the coolest semantic web related thing I've seen this year. Sparqlbot lets you load semantic data from various URLs and perform SPARQL queries on them via IRC using natural language. [...] The end result is something that resembles the ship computer from Star Trek.
and goes on to cite what probably was his first conversation with SPARQLbot. Funny and playful, but also a very interesting example of what is already possible with semantic web parsers.
SPARQL is a protocol and a query language for RDF data with a syntax similar to other well-known query languages. You can use it for fetching and even writing data that is stored in RDF format. SPARQLBot is a fun project that was demonstrated at Semantic Camp in London earlier this month and has been described as the Semantic Web Command Line. People can contribute their own SPARQL queries as SPARQLBot commands, and so far the count exceeds 30 entries. If you want to talk to SPARQLBot, visit #sparqlbot and give it some commands. Hilarious!
I'm currently overhauling this blog installation, especially some plugins that are in need of updating. This might break things temporarily, and I'm really sorry for that. However, if you can indeed read this entry, everything is working fine, so never mind 
I'll post an update when I'm done and elaborate on the changes in the background. Stay tuned.
Update: I straightened things up a little, introduced two new categories, English Articles and Deutsch Beiträge (German Articles), because I might blog a little more in German from now on, and broke the download manager in the course (it's up and running again, yet I still have to re-up most of the files). All in all it could've been worse so far. Again: Stay tuned.
