Blogging with AI

Rationale 🤔

Why do teachers ask their students to write? If I already know x% more about the content of the class than you do, it’s unlikely that I also need y more paragraphs from you on whatever we’re reading for a given day, right?

You’re absolutely right. I don’t need any more to read. Instead, I ask you to write with the goal of helping you be a better reader. Let me explain: I remember college. I know how much you’ve got to read each day. You’ll got the gist of what you’ve read, but it can be hard for it to sink in if you don’t do anything else. But if I ask you to write about whatever you’ve read, you start thinking about it differently. You’ve got to find something to say, which means you need to know more clearly what it is you read in the first place. The easiest comparison here—at least at BYU—is to the difference between the way many of us do daily scripture study and the way we study the scriptures when we are asked to teach a lesson.

But writing isn’t something that you should do in isolation or only for your professor. Instead, writing works best when it’s part of a conversation. Part of the conversation is with the thing you’re writing about; but the other part is toward the people you are writing to. Writing feels less like an empty exercise when you know it will be read by others. And that’s a great reason for us to be blogging. What’s more, as others read what you’ve written, it gives them a chance to see insights you’ve developed that differ from theirs. You will likely challenge their reading in productive ways.

With all this said, we are living in a moment when writing is undergoing rapid change thanks to large language models (LLMs) like ChatGPT, Google Bard, or Perplexity AI. We suddenly have new ways of writing that are worth exploring; can they help us become better readers and writers? How much can adding a machine to the conversation improve our reading? Let’s find out…

The Nitty Gritty đź’Ş

Over the course of the semester, you will write posts in collaboration with LLMs, read your classmates’ posts, and comment on the latter.

Write Posts

  • During the semester, you will write seven, 250-word-minimum blog posts. You may not submit more than one blog post per week for credit. There are 15 weeks of class in our semester; you won’t be able to write during the first week (which is already past) or the week of Thanksgiving, and you only have one chance to write this (the second) week. As such, you have to write during 7 of 12.5 possible weeks. Don’t fall behind!
  • For each post, choose one of the following methods:
    • Write your own blog post about the day’s reading. Then ask a LLM to write a blog post of similar length; post them both and we’ll see how it handles the same topic that you wrote on.
    • Ask a LLM to write a blog post about the day’s reading for you. Then either revise that post into something better and more reflective of your thinking or write a post that responds to the LLM’s post.
  • Four of your posts must use one method, and the other three posts must use the other. But be sure to indicate at the start of the blog post which method you’re using.
  • To ensure that everyone has a chance to read posts before class, each post should be published by 10am on the day for which the relevant text has been scheduled (e.g., Tuesday at 10am for something we are discussing that day at 2pm).

What should you write about? Mark Sample at Davidson College provides the following suggestions:

Focus on an aspect of the day’s material that you find particularly compelling. This could be something you don’t quite understand or that jars you. Or you could formulate an insightful question or two about the material and then attempt to answer your own questions. You might make connections between the course material and conversations we’ve had in class (or ideas that you’ve encountered in other classes).… In any case, strive to go beyond the obvious or what we’ve already talked about.

Read Posts

You are required to read your classmates’ posts. All of them. #worstprofever

Comment on Posts

  • By the end of the semester, you must have left 15 comments on your peers’ posts. I will only count 2 comments per week toward this total.
  • Comments must be posted before the start of class.

Your comments should engage directly with the content of your colleagues’ posts. They should be more substantive than a simple 🙌. Your comments can be short and informal but shouldn’t be flippant. What points do you find compelling? What further questions does the post raise for you? Where do you (politely) disagree with their thinking?

You’re welcome to use a LLM to generate comments, but make sure you’re then commenting on the comment.

Grading 🅰️

Blog Posts

Blog posts count as either Unsatisfactory, Satisfactory, or Excellent. A Satisfactory post must meet all of the following criteria:

  • It is posted on time (again, 10am on the day we’ll discuss the reading).
  • It includes a descriptive title.
  • It refers specifically to the day’s reading, almost always through direct quotation.
  • It is at least 250 words long.
  • It is written clearly. By clearly, I mean less formal than a “normal” paper but still serious and rooted in evidentiary-based reasoning.
  • It contains no more than 3 grammatical or spelling errors.

In addition to the above, Excellent blog posts must also:

  • Include one illustrative media object (an image, a screenshot, a GIF, an embedded video or audio, etc.). You must give credit given to the original source, giving a link and its creator’s name(s).
  • Provide exceptional insight/analysis or make connections far beyond the day’s material and/or with the LLM’s writing. Want some examples of “exceptional insight/analysis”? Try these posts from a previous year’s class.

Your final grade for the blogging portion of this assignment (170 points) will be determined by what “bundle” your 7 posts fall into:

  • 7 Excellent posts is an A+ (100%)
  • 6 Excellent posts with one Satisfactory is an A (95%)
  • 4 or 5 Excellent posts with remaining posts being Satisfactory is an A- (92%)
  • 3 Excellent posts with the remaining posts being Satisfactory is a B+ (88%)
  • 1 or 2 excellent posts with the remaining posts being Satisfactory is a B (85%)
  • All Satisfactory posts is a B- (82%)
  • More than 2 Unsatisfactory posts with the other posts being Satisfactory is a C (75%)

Each student will have the opportunity to rework one Unsatisfactory post.

Comments

The commenting portion of this assignment is worth 30 points. Comments will be graded on a Satisfactory or Unsatisfactory basis. Satisfactory comments meet the following requirements:

  • Posted on time (before the start of class at 2pm).
  • Engage with the original post.

Please note: while I do read all of your posts👓, given the realities of the space-time continuum 🚀⏳, I cannot comment on all of them. While eliciting a comment from me is—in fact—a reason for fist-bumping your friends, 🤜🤛 a lack of comments should not be seen as a criticism of your work.

Credits 🙏

Over the years, my blogging assignments benefited from the work of my fellow authors at ProfHacker. Mark Sample’s 2018 assignment inspired several revisions that year. This time around, I’ve radically reworked the assignment in response to the popularization of large language models. Jason Mittell provided helpful suggestions as I shared an early version.

css.php