But there are pleasures to be found in its equally elegant source code. Written in HTML 3.2 code, this code is very clean and enacts practices that became standard in later versions of HTML, such as formatting the <body> text in the <head> of the document: a practice now implemented as CSS. Even the metadata is bare-bones, as can be seen below:
Not only does he not include a long list of keywords that a search engine would use to index the page and allow people to find it, he specifically uses an protocol designed to keep Internet bots away. Human readers of the code won’t get any conceptual shortcuts either, though reading the arrays provides a sense of the grammatical and thematic structures he is aiming for. For example, the array for the first line in the poem contains 60 items, but 10 of them are the same word, “visitors.” This means that every time you generate the poem there is a 16% chance of getting visitors, and a 1.6% chance of getting each of the other words.
How many of the people present in the snapshots inhabit those urban landscapes and how many are just visitors?