On the mechanics of the post itself: I like putting the footnotes off to the side of the body. The alternative - using hyperlinks to jump back and forth from the location tot he end - is more intrusive.

Concur.

I think footnotes are an artifact of dead tree printing; that it has been translated directly to interactive media is ludicrous. I predict its extinction, but the wait is very surprising.

If only there were an easy, standard way to pop up a temporary "tooltip window" within a page. At that point, clicking on a footnote would pop up the actual footnote, rather than scrolling the browser to the bottom of the page.

(Putting footnotes within the page itself causes the page to be wider, which could potentially make the article harder to read on mobile devices. But at this point, a Javascript "popup window" solution would probably be worse.)

<abbr title="it is usually used for abbreviations, but why not?">abbr tag</abbr>

The djangobook does it quite nicely (not footnotes but it would work either way).

I appreciated it, but it took me a moment to notice the asides. After all, I have spent years visually filtering advertising sidebars from page content.

My attention was drawn as I was wondering why adblock was not working.

Just went back to the article to have a look, totally missed the footnotes the first time. Personally, in-page hyperlinks satisfy me.

With Javascript disabled, the sidebar notes are completely disfigured. I love the Noscript plugin, and more webdevs need to think about us.

Aside:

The beauty of that logo!

Clojure's logo is beautiful too.

Didn't pay close attention at first, but the kanji brush-stroked lambda certainly says "Zen of Lisp."

People prefer arrays to singly-linked lists. Python gets it (but doesn't get many other things Lisp gets)

One of the advances of Clojure over other Lisps, is that vectors and maps are just as well supported as lists. They have their own reader syntax (using [] and {} as delimiters, respectively), and all three implement the Collection interface, which means a lot of Clojure functions work equally well on all three. There is also syntax for destructuring. And the use of vectors in places where destructuring is possible (like function arguments and let bindings) makes it easier to identify code structure at a glance (a complaint against other Lisps).

Common lisp sequence functions work on lists, vectors (including those specialized for bits, bytes, etc), and strings.

Some of them do. There are lots of redundant functions in CL, for reasons of backwards compatibility.

All of the sequence functions and special forms work on all of the sequence types.

The "redundancy" is that there are also functions/forms that work on specific sequence types. It would be surprising if there weren't because if two data types have exactly the same set of operations, one of them may be unnecessary.

That's a strange statement. Lists and arrays are different data structures for different purposes. It's like saying that people prefer btrees to structures. Lisp has arrays as well, so if people want to use them they can.