I will first not-completely-but-somewhat-jokingly say that this article’s first author is a past student of EE 295, so of course he’d be doing lovely things with his writing! (I will add that he was a good writer when he started in EE 295, and that his advisor’s students are often excellent writers. The advisor is the third author on this paper. A culture of good writing, of valuing writing, seems to develop in some labs.)

This July 2015 article is “Variable-Length Convolutional Coding for Short Blocklengths With Decision Feedback” by Adam R. Williamson, Tsung-Yi Chen, and Richard D. Wesel. Since it is so recent, I will only photograph one short excerpt from the text, although somewhat more will be cited and described.

Drama is developed from the first sentence, when the authors write something along the lines, of “Although the founding father of our field found that feedback was not useful for x, feedback can be used for other purposes”:

If that’s not dramatic, then what is? It’s dramatic when an important figure in the field is right, or wrong, or just missed noticing something important.

The way that Einstein’s thoughts on the cosmological constant have been cited and argued for and against is perhaps similar. He was right. He was wrong. (Maybe there we even some other back-and-forths, and certainly there were many amendments.) Then, a January 2013 Scientific American called “Right Again, Einstein!” starts, “A new study of one of the universe’s fundamental constants casts doubt on a popular theory of dark energy” (Moskowitz). Almost three years later (Sept 2015), “What Einstein Got Wrong” in that same magazine begins: “Like all people, Albert Einstein made mistakes, and like many physicists he sometimes published them” (Krauss). There’s something exciting and important about great thinkers having limitations, even if they might just be limitations caused by the moment of time in which they lived, and the development of their fields at that moment.

The extensive literature review of the Williamson et al. paper tells a story, too. It’s chronological: this idea was developed, and then that, and then this other one went further. It’s also got characters; the researchers are listed by their names, rather than the papers being listed by reference numbers. Once there are names, some intellectual drama can be introduced: all these authors did stuff in reaction to the work that came before: one name does “pioneering work,” another “formalizes it” or “demonstrates” something else, or “furthers” or “extends” the work. Others “study” or “show” or “provide an overview.” It’s complicated, but there are a lot of people doing stuff, cooperating even, and that’s interesting and appealing.