An assignment can give students a clear intellectual purpose and still leave a practical question unanswered. Read the same instructions as a student and as an assignment designer, then decide which observations warrant a change.
The assignment
Use Fred Gibbs’s Campus History Part 2: Archive Visit and Campus History Essay, from Making History, HIST 1105, Fall 2026. Students visit the Center for Southwest Research, compare archival evidence with an earlier AI draft, and produce an essay for the Campus History website.
- Original course assignment
- Workshop copy of the assignment (text, September 23, 2026)
- Part 1, which precedes it
- Student publishing guide
The workshop copy preserves the assignment wording and turns its display boxes into ordinary headings. It includes the link to the publishing guide. An assistant should not claim to have read linked pages unless they were supplied or it actually opened them. For a classroom revision, include the schedule and other instructions students receive too.
The exercise concerns the clarity of the instructions. It does not ask AI to write the student’s essay or assess a student’s grade.
Read as a student
Read this assignment as a first-year student encountering it for the first time. Do not complete the assignment or rewrite the instructions.
Explain what I would do first, what I think I must submit, and how I would judge whether my work meets the expectations. Quote the instructions that support each part of your reading.
Identify up to three places where I would have to guess. Separate what the assignment says from what you inferred. Treat information absent from these materials as a question for the instructor, not proof that it was never explained elsewhere.Use the complete Campus History Part 2 assignment. Supply any course schedule or rubric you want included, and list what you supplied.
A recorded reply
Reading the assignment as a first-year student
What I would do first
The assignment assumes I’ve finished Part 1, and it says so near the top: “You’ve already completed the AI draft and critique in Part 1.” So my first move is to locate my AI essay and my critique of it, since I’m told to bring them: “Bring a pencil (pens are not allowed), your phone for photos, and your AI essay critique.”
Then, before going to the archive, I’d search the digital catalog: “Search the New Mexico Digital Archive for your topic. Use exact quotes around multi-word terms.” The same section says to “Note the Collection Number and relevant box numbers from the results,” which I need in order to fill out the request form. I’d also check alternate names (“Names change”) in case my first search comes up empty.
Then I’d go to Zimmerman Library for “at least an hour,” check in at the reference desk, and ask for help.
What I think I must submit
An essay published on the UNM Campus History website, which the instructions describe like this: “Your finished essay will be published on the UNM Campus History website.”
Requirements as listed:
- “~800-1000 words”
- “4-5 images — historical photographs and/or scanned primary sources from CSWR”
- “Informative captions on every image: not just what it shows, but why it’s significant. Cite where you found it (box number, collection name, or URL).”
- “Subheadings to organize your main sections”
- “Bibliography of sources used (CSWR materials + published histories),” with sources specified “exactly” enough “so it can be verified”
There’s also an AI-Archive Comparison, which appears in the grading criteria rather than the requirements list: “An AI-Archive Comparison that’s specific: quotes the AI draft and names what was confirmed, contradicted, or simply absent from the archive.”
The submission mechanism is “submitting a pull request,” per the publishing guide.
How I would judge whether my work is good enough
The grading section says the project is graded “primarily on effort and intellectual engagement, not writing polish or whether you found spectacular archival material.” I’d check myself against the A description: “clear archival evidence throughout; specific sources cited; AI-Archive Comparison quotes the draft directly and names concrete differences; images with substantive captions that explain significance.”
I’d also run the “weak submission” list as a checklist: could it have been written without visiting the archive? Is the AI comparison generic? Are images missing or uncaptioned? Is it “just a list of facts with no narrative thread”? And for style: “smart magazine, not academic paper.”
Three places I would have to guess
1. Where the AI-Archive Comparison goes and how long it is.
- Says: It’s a grading criterion, and the essay should “fuse” the AI essay and my findings. The “Skills” and “What I’m looking for” sections describe it as a distinct thing (“An AI-Archive Comparison”).
- Inferred: It might be a section inside the 800-1000 word essay, or a separate piece on the page, or something submitted to the instructor only.
- Question for the instructor: Does the comparison count toward the word limit, and is it part of the published page?
2. What exactly I turn in to the instructor, versus what gets published.
- Says: The publishing guide explains “how to publish the page,” while “This assignment guide explains what the page needs to argue, document, and reflect on for class.” The word “reflect” hints at something beyond the essay.
- Inferred: Probably the pull request or page link is the submission, but I can’t tell if there’s also a Canvas upload, notes, or a reflection.
- Question for the instructor: Where and how do I submit for the grade, and is there anything besides the page?
3. Deadlines and what counts as enough archival evidence.
- Says: No due date appears in this document. “At least an hour” is the visit length, and the grading mentions “notes that reflect actual digging through boxes.”
- Inferred: My notes might be graded, but nothing says I submit them.
- Question for the instructor: What’s the due date, and are my notes or photos part of what’s evaluated?
A smaller question I’d also keep in mind: how do I choose my topic, and was it assigned in Part 1? This document refers to “your topic” as if it’s already settled.
A fresh incognito conversation at the default settings, with only the workshop copy of the assignment attached. The prompt was sent as written.
Treat the response as a possible reading to investigate. Ask actual students to explain the assignment back to you before concluding that the instructions are clear to them.
Read as an assignment designer
Use the same materials in a fresh conversation. TILT’s framework asks whether students understand the purpose, the task, and the criteria for success. The assignment need not use those exact headings to communicate them. See TILT’s resources, design checklist, and explanation of using the framework flexibly.
Review this assignment using transparent assignment design:
- Purpose: what students will learn and why it matters.
- Task: what they must do, in what sequence, and submit.
- Criteria: how they can recognize successful work.
For each, quote the evidence and distinguish what is clear, partly explained, or not stated in the supplied materials. Recognize strengths as well as gaps. Check whether every graded component is introduced in the task instructions.
Prioritize three questions or small clarifications. Do not invent requirements, dates, grading weights, or teaching activities. Do not treat missing section labels as proof of poor design. Do not rewrite the assignment or give it a numerical score.Use the same assignment and supporting materials in a separate conversation. The review uses TILT's purpose, task, and criteria framework.
A recorded reply
Transparent Design Review: Campus History, Part 2
This review covers only the supplied snapshot (Sept 23, 2026). Part 1 and the linked student publishing guide aren’t included, so anything they might explain is marked as outside the materials, not as a gap.
Purpose
Clear. The assignment names three skills and explains each: structural analysis (“how institutions… shape what knowledge survives and what gets lost”), reading for absence (“What would the AI never know because it was never digitized?”), and source criticism (“who created this document, why, and what can (and can’t) it tell us?”). The “intellectual goal” is stated plainly: “find out what the historical record actually says about your topic and compare it to what the AI told you.” The audience framing, “a curious reader… who knows nothing about its history,” and the real publication venue give the work a purpose beyond the classroom.
Partly explained. The text says what the skills are but not why they matter beyond this course, for instance why comparing AI output to archives is a useful habit. It also doesn’t say how each skill appears in the essay itself. “Source criticism” is a named skill, yet nothing in the task or the requirements asks students to do it in the essay.
Task
Clear. Logistics are concrete and sequenced: what to bring (pencil, phone, AI critique), where to go, how to check in, how to find and request boxes (“about 10 minutes”), and what to photograph versus scan. The requirements are countable: “~800–1000 words,” “4–5 images,” “Informative captions on every image,” “Subheadings,” and a “Bibliography,” with the standard “specify exactly where you found something so it can be verified.” The research tips (“Names change,” “You’re not searching full text”) anticipate real obstacles. The division of labor between this guide (what to argue and document) and the Campus History guide (how to publish) is explicit.
Partly explained.
- The task says to “reshape your AI essay into a UNM Campus History essay, that fuses both the AI essay and your own findings.” What “fuses” looks like (revising the draft, quoting it, discussing it separately) is not stated.
- The publishing section says what the page needs to “argue, document, and reflect on for class,” but “reflect” has no corresponding instruction.
- The submission mechanism is deferred to the student guide (“submitting a pull request”). That seems a reasonable handoff, but no deadline appears in this document.
Not stated. No due date, and no instruction on where the AI-Archive Comparison goes or how long it should be.
Criteria
Clear. The A–F descriptors are specific, and the weak-submission list is helpful (“Could have been written without visiting the archive”; a generic comparison like “AI can sometimes be inaccurate”). The priority is stated honestly: “primarily on effort and intellectual engagement, not writing polish or whether you found spectacular archival material.” That reassurance is a real strength, since archives don’t always cooperate.
Partly explained.
- Evidence of engagement is named (“specific sources cited, real images scanned, notes that reflect actual digging”), but “notes” aren’t requested anywhere in the task or submission.
- “A readable story with a point, even if a modest one” is a criterion, but the task only says “Think about the story you want to tell” as a tip.
- The B and C descriptors rely on terms like “mostly specific” and “thin” without examples.
Not stated. Relative weighting of components. I’m not inferring any, since none is supplied.
Task–criteria alignment check
| Graded component | Introduced in task? |
|---|---|
| Archival evidence, specific sources, bibliography | Yes |
| Images with substantive captions | Yes (“not just what it shows, but why it’s significant”) |
| Readable narrative with a point | Lightly (a tip, plus “Style”) |
| AI-Archive Comparison | No. It appears only in the criteria. |
| Notes as evidence | No |
| Source criticism, reading for absence | Named as skills, but not requested in any deliverable |
Three priority clarifications
- Where does the AI-Archive Comparison live, and what is it? It is central to the A-level criteria (“quotes the draft directly and names concrete differences”) yet is never listed under Requirements or the essay instructions. Is it a section of the published essay, a separate submission, or covered by “reflect on for class”? Does it count toward the 800–1000 words?
- How should students demonstrate the named skills? Reading for absence and source criticism are stated purposes, but no deliverable asks for them. A sentence connecting them to the essay or comparison (for example, whether captions or the comparison should address who made a source and what’s missing) would make the purpose visible in the work.
- What exactly gets submitted, and are notes included? The criteria cite “notes that reflect actual digging,” but the task never asks for them, and the submission is described only through the external guide. Clarifying the full set of items and the deadline (if not in the guide) would let students see how effort will be recognized.
A fresh incognito conversation at the default settings, with only the workshop copy of the assignment attached (making-history-campus-history-assignment.txt). The prompt was sent as written.
It credits what the assignment already explains before listing gaps. Its alignment table marks the AI-Archive Comparison as missing from the task: “No. It appears only in the criteria.” The comparison below disputes that.
Compare the readings
Compare these two readings of the same assignment.
Where does a student's practical uncertainty connect to a design issue? Where do the readings differ? Check each claim against the assignment and quote the evidence. Identify any claim the readings do not support.
Recommend one clarification for the instructor to consider, then state what decision the instructor still needs to make. Do not resolve that decision or rewrite the assignment yet.Supply the assignment and both full readings. Agreement between generated roles is a lead to investigate, not independent confirmation.
Here is a recorded reply, run with the assignment and both recorded readings attached:
Where student uncertainty meets a design issue
The student’s three guesses line up with three misalignments the designer found. The designer’s “Part 1 and the linked guide aren’t included” caveat applies to all of them.
| Student’s practical uncertainty | Design issue behind it |
|---|---|
| Where the AI-Archive Comparison goes and how long it is | The comparison is graded (“An AI-Archive Comparison that’s specific”) but is not in the Requirements list |
| What gets submitted versus published | The page must “argue, document, and reflect on for class,” but “reflect” has no matching instruction, and submission is handed off to the external guide |
| Deadline, and whether notes count | “notes that reflect actual digging through boxes” is named as evidence, but notes are never requested, and no due date appears |
The first is the strongest connection. The student can’t tell how to meet a criterion because the task never names it as a deliverable.
Where the readings differ
- Frame: The student asks “what do I do and how do I know it’s good enough?” The designer audits purpose, task, and criteria for consistency.
- Skills: The designer treats source criticism and reading for absence as a gap (“nothing… asks students to do it in the essay”). The student never mentions them.
- Notes: The student wonders if notes might be graded. The designer says outright that they are never requested.
- “Reflect”: The student reads it as a hint of “something beyond the essay.” The designer reads it as an instruction with no counterpart.
- Topic and Part 1: Only the student raises these (“your topic” as if settled). The designer sets them aside as outside the snapshot.
Claim check
Supported by the assignment:
- All of the student’s requirement quotes (word count, images, captions, subheadings, bibliography) and the grading quote (“primarily on effort and intellectual engagement”).
- “Bring a pencil… and your AI essay critique,” and “Note the Collection Number and relevant box numbers.”
- The designer’s “about 10 minutes,” the “curious reader… who knows nothing about its history” audience framing, and the weak-submission examples.
- No due date appears, and no relative weighting is given.
- The B and C descriptors (“mostly specific,” “thin”) have no examples.
- The tip “Think about the story you want to tell as you go” is the only task-side mention of narrative.
Not fully supported:
- Student: the “Skills” section describes the comparison as “a distinct thing.” The Skills section never names an AI-Archive Comparison. Only “What I’m looking for” does. The Skills section only gestures at it (“What would the AI never know because it was never digitized?”).
- Student: the catalog numbers are needed “in order to fill out the request form.” The assignment says to note them and to “fill out” a request form, but never links the two.
- Student: “reflect” “hints at something beyond the essay.” The sentence says what “the page needs to argue, document, and reflect on,” so the subject is the page. Both readings treat the word as ambiguous, which is fair, but it may point inside the essay.
- Designer: the comparison is “No” in the task. This is overstated. The task says “compare it to what the AI told you” and asks the essay to “fuse” the AI essay and findings. What’s missing is the named deliverable and the quote-the-draft requirement, which appear only in the grading section.
- Designer: reading for absence is “not requested in any deliverable.” This is also overstated. The criteria ask the comparison to name what was “simply absent from the archive,” which is related to absence, though not quite the skill as defined. The grading section also acknowledges “sometimes there isn’t much there.”
- Student: the pull request is the graded submission. The assignment says the guide covers “submitting a pull request” for the page, not that this is what’s graded.
One clarification to consider
State, in the Requirements list, where the AI-Archive Comparison appears and whether it counts toward the ~800–1000 words. This clarifies the student’s first uncertainty and the designer’s top priority. It likely also clarifies the second and third items, since the answer shapes what “reflect” means and whether anything beyond the page is submitted.
The decision still open
The instructor needs to decide whether the comparison is part of the public Campus History page, written for the “curious reader” who “knows nothing about its history,” or a course-only piece read just by the instructor. That choice determines the audience, the word budget, and whether notes or other evidence are submitted. I haven’t resolved it or drafted any wording.
A fresh incognito conversation with three files attached: the workshop copy of the assignment and the two recorded readings on this page. The prompt was sent as written.
It finds claims in both readings that the assignment doesn’t support. The designer’s “No” is overstated: the task does ask students to “compare it to what the AI told you,” but never names the comparison as something to hand in. It ends with a decision only the instructor can make, which is whether the comparison belongs on the public page.
An apparent agreement between AI roles does not turn a diagnosis into a fact. Return to the instructions and ask whether the quotation supports the concern. The instructor may have explained a detail in class or put it in another document.
Prepared observations for discussion
These observations come from reviewing the assignment for this workshop. They are prepared teaching examples, not transcripts of separate AI runs.
The student has a workable starting point. The page describes an archive visit, what to bring, how to request boxes, and how to collect images and citation details. The finished essay has an approximately 800–1000-word target, four or five images, captions, subheadings, and a bibliography.
The designer should recognize existing strengths. “Skills you are practicing” explains learning purposes. The archive instructions provide a sequence of actions. “What I’m looking for” names evidence of strong work and distinguishes grade levels. A review that claims the assignment has no purpose or no criteria has missed explicit information.
A useful question joins the two readings. The grading section expects an “AI-Archive Comparison,” but the requirements list does not explain where to put it or whether it counts toward the essay’s word target. A student could reasonably ask whether it belongs inside the public essay or in a separate reflection. The designer can trace that uncertainty to the connection between required work and grading criteria.
The next move belongs to the instructor. Decide where the comparison belongs and how it relates to the word target; then clarify that in the task instructions. Do not have the assistant choose a submission format or add a word count on its own. Due dates and course-wide submission arrangements may appear elsewhere, so their absence from this page is a bounded observation.
Return to the reader-role exercise or presentation.