Just got back from a one week vacation in Algarve. The weather was pretty good, but due to my broken toe I couldn't enjoy the beach and had to stay on the balcony, where I set up a simple but functional work desk. I stayed in the shade all the time and still managed to get mildly sunburned...
You can copy-paste text into a text box and ask Wordle to analyze it and produce a word cloud, where the size of each word reflects how often it appears in the analyzed text. Alternatively, you can point Wordle to a del.icio.us account and it will produce the corresponding tag cloud. Here's my the tag cloud for my top 100 del.icio.us tags:
Pretty obvious what my favorite tag is... Incidentally, I haven't been using del.icio.us a lot lately. I've been using Diigo instead, which in addition to social bookmarking allows you to highlight and save text clippings from a web page.
By default Wordle analyzes the 150 most frequent words, but you can tweak this number to produce sparser or denser word clouds. You can also choose to ignore capitalization and show only lowercase words, which eliminates spurious duplicates.
Wordle conveniently removes frequently used words such as "the" or "is" unless you explicitly request it to show them.
Wordle randomly chooses a font and a color scheme, but these can also be changed by the user. Many of the available fonts are gorgeous, whereas the color scheme offering could be more varied.
After del.icio.us, I decide to generate a word cloud for my Twitter tweets, but I first had to obtain my full set of tweets. Unfortunately I couldn't find a way to download my own tweets from Twitter, which is unfortunate. Twitter, if you're listening, please set my data free (or correct me if I'm wrong). Google Reader and Gmail do it; Typepad does it; Bloglines does it; so should you.
Luckily I found TweetDumpr, a simple webapp that exports your tweets into a CSV text file. With some simple parsing and cleanup, I got a clean text file with all my tweets, which I copy-pasted into Wordle. Here's my Twitter top 100 word cloud:
Unfortunately Wordle currently seems to be down for maintenance, with the following notice: "I am tinkering with Wordle... Don't fret! We'll be back."
The GaWC group uses the term World City, but I prefer the term Global City -- it's less prone to confusion. Besides the list of alpha, beta and gamma global cities, the GaWC group also provides a secondary ranking of cities possibly on their way to Global City status:
I find their positioning of Lisbon in the "Some evidence" category quite puzzling. Ranking Lisbon at the same global-city level as Oslo, Birmingham and Brisbane seems questionable, to say nothing of bundling it with cities such as Brisbane, Bucharest, Bratislava(!), Montevideo(!!) and Almaty(!!!??).
Lisbon is no Manhattan, but it is not exactly a backwater either. According to Euro Monitor's list of Top 150 City Destinations in the world, Lisbon is placed at a respectable 47th place, with 1.7 million arrivals per year, higher than Zurich's 1.4 million and São Paulo's 1.1 million, two cities classified by the GaWC group as beta global cities.
Lisbon is also the 7th most popular city in Europe for meetings and conventions, according to the International Association Meetings Market 2006 survey of the International Congress & Convention Association:
Portugal itself ranks 15th out of 130 in the Travel & Tourism Competitiveness Index for 2008, and Lisbon must surely be one of the two major travel destinations within the country (the other being the beaches of Algarve).
Of course, tourism and conventions alone do not a Global City make, but I would still expect a stronger correlation. I don't know all the criteria the GaWC group uses for their classification, but it seems to me they need to revisit some of them.