Puzzlepieces – February, 2010

Via Alf, I learn that apparently delicious has long supported lookup by domain and path, not just absolute URL.

I took a look through all the bookmarks of my old website, Fagan Finder, and it turned up a few interesting things. The most popular page, with some commentary:

1. URLinfo (688). This makes sense because it is (well, was) a useful tool for web developers, bloggers, etc., the typical audience of delicious, however it is definitely not one of the most popular pages by the site’s own statistics.
2. All About RSS (568), with bookmarks spread over 34 different URLs. Aside from the bookmarks to specific sections of the page, this shows the results of moving the URL by changing the file extension, although I did use a 301 redirect. This page is no longer one of the popular pages in terms of traffic.
3. Google Ultimate Interface (377). Aside from the oddity of this not-that-useful page being so popular, what’s interesting is that a long time ago I created a second version that failed to work in Mozilla, so I had it redirect to the old version for those users which I thought (at the time) would be rare… yet there are almost 6 times as many bookmarks for the Mozilla/Firefox version.
4. Image Search Engines (322) - finally we come to a page that is actually popular among the general public; it’s also the only page on my old site that I updated “recently”
5. Translation Wizard (201) - sadly this hardly works any more but I loved the idea and spent an insane amount of time to build this

Another thing I noticed is that a tool I’d once built but never referred to anywhere, and could only be found by going to a tag page on this blog and clicking “more” in the sidebar somewhere somehow has 10 bookmarks.

Back in early 2005 I hacked up a quick web app to help me study for the Arthropod Zoology course I was taking in university. It helped me so much that in 2006 I decided to remake it in a non-ugly and usable way and I demoed it at BarCampWaterloo in 2007.

There it rested, my “current” yet abandoned project until around September 2009 when my friend Ben began to refactor all the code.

Lately I’ve had the time to work on it more seriously. I’ve moved it to Quizify.com and it is now ready for the general public.

Functionality today is fairly simple but quite useful, at least in my opinion . Input a URL that includes a definition list (as in a <dl> in HTML) and it creates a flashcard-like quiz with the data.

I plan to continue improving it, but in the meantime, feedback is welcome. Oh, and the NLP APIs I was blogging about recently was related to this project, but for a feature that won’t be ready for some time.

I’ve had this draft post about Aardvark for about two weeks now. Now that they’ve been acquired by Google, I guess it’s about time to finally publish it.

I first heard about Aardvark via the Seattle Tech Startups mailing list and eventually got around to trying it. Few things get past my initial attempt, but I’ve still got Aardvark. It’s a question-and-answer service where you can ask questions yourself and answer questions of others.

What I’ve enjoyed the most about Aardvark (beyond it’s ability to send questions to the right people) is how easy to use and friendly it is. I interact with it via instant messenger, and every message it sends me includes all the instructions I need, in a friendly way, without being too verbose either. It’s impossible to not understand how to use it.

Recently they published a paper - Anatomy of a Large-Scale Social Search Engine (the name is a reference to a famous Google paper) - which I found quite interesting. I was expecting more statistics about the usefulness and value of Aardvark than the paper had, however the interesting part is that Aardvark turned out to be far more sophisticated than I’d realized. As I read it I’d think of a way to make it even better, and later on in the paper, find that they’d already done that. One astonishing graphic in the paper is their graph of users over time; that’s some impressive growth.

Now that Google’s bought them, I only hope that they’ll allow the founders to keep doing the good job they’ve been doing… I’ve seen too many excellent products wither after acquisition (e.g. dodgeball and jotspot).

A few weeks ago I was in Florida, around the Fort Lauderdale area, and for the first couple of days, it was very cold, for Florida. Too cold for many lizards, that’s for sure. Here are a couple of my photos of deceased lizards taken after it warmed up. These are mostly iguanas and other large lizards, as I didn’t take photos of any small ones.

In the second photo you can see that something has removed the lizard’s tail, presumably after it died. The second-last photo shows three large iguanas lying on floating plant material, all of whom presumably froze and fell out of a large tree that reached over the water; the last photo is a cropped zoom-in of one of those three.