Words taken from wikipedia article about New York City. Original image from the same wikipedia article. Itâ€™s a panoramic night view, looking north from the Empire State Building (image from Diliff)
This is an oportunity to test the limits of the algorithm. So Iâ€™m thinking of a really large print, like a building facade or the interior wall of a restaurant (of course Iâ€™m architect). This will be the target, donâ€™t blame me if you try to print it on a little piece of paper.
So there will be new rules: Only uppercase or firstcase words, these must be representative words of the city. Only words with two or more hits, I onlyÂ need the most representative of them.Â Resultâ€¦ there is only 223 words left from (aprox) 3.000.
First twenty words: york 250, manhattan 52, american 48, brooklyn 23, bronx 22, queens 19, staten 18,central 15, jersey 11, lower 11, university 11, county 10, international 10, hall 9, boroughs 8, court 8, liberty 8, african 7, broadway 7, college 7.
After a couple of test I realised that I cannot use the count method to calculate the word size, because I need to put almost 20.000 words to get a nice result. Then I discarded the counter value and redistributed the sizes using an exponential function between two fixed values.
Hope you like it. Images free for personal use.