research on urogenital microbes in women's health, glycobiology research and training.
Joined January 2012
- Tweets 11,349
- Following 1,799
- Followers 2,277
- Likes 11,875
453 Photos and videos
Mar 8
Is life closed book? Did I miss something?
I just gave a closed-book, pen-and-paper midterm exam in my 300-level course at UBC with 100 students. All exams were graded by an experienced graduate-level TA according to a rubric.
*** The average was 64/100.***
My class averages at UBC are usually 80-85.
Context:
• This was the first midterm, covering ONLY 4 weeks of material.
• Students had a list of possible questions in advance: no surprise questions.
• Questions included (a) 3 concept definitions, (b) 3 paragraph-long questions, and (c) a 1.5-page essay.
• I have taught this class multiple times. Nothing in my teaching style changed this semester.
• We read entire paragraphs of text in class, so students don't have to do something on their own that wasn't covered during the lecture.
• Students take a 10-question multiple-choice quiz at the end of every class (30% of the final grade).
• Attendance is 95-99% every class. Attention during lectures and participation in pair-work activities are very high → anticipating the end-of-class quiz.
*** But unfortunately, I suspect many students are not reading the material on the syllabus. They are asking LLMs to summarize it instead.***
After the midterm, students reported:
• They thought they knew concept definitions but couldn't produce them on paper.
• They thought they understood the arguments but struggled to connect them or identify points of agreement and disagreement.
My view:
It might be “cool” or “innovative” to teach students to summarize readings with ChatGPT or write essays with Claude. But we may be doing them a disservice: reducing their ability to retain material, think creatively, and reason from what they know. If you only read what AI has summarized for you, you don’t truly "know" the material.
Moving forward:
We have a second midterm coming up. I don't know how to convey to students that the best way to do better on the exam is to rely on and improve their own reading skills.
130
Mar 8
I know all my followers here are totally into the incompressible Navier Stokes problem
73
Mar 8
Good thing all our scores are above the nonpayline of 4% @drugmonkeyblog
Mar 6
the US government is proposing these very stupid changes for international students
96
Mar 7
That’s funny… Claude says I need human experts to do most projects.
Mar 7
Anthropic researcher: Even if all AI progress stops now & algorithms don’t improve, current models already can automate most white-collar jobs within 5 years.
Manual task-feeding to AI model is already more economically viable than human labor.
1
90
The Lewises retweeted
NIH is shutting down some study sections. E.g.,
3
10
32
9,891
Feb 25
Gosh, it’s absolutely brutal to get fundable scores that are not fundable.
2
96
Feb 18
Grok 4.2 is indeed based “historical evidence overwhelmingly shows [removal] was not [based]. The process involved systematic displacement, broken agreements, legal manipulations, outright violence by European settlers and later the U.S. government, driven by expansionist policy
66
Jan 11
Spoiler alert it’s studying rich source of failure modes, and gooners
Jan 11
BREAKING: Another all time high on 𝕏 🔥
New users are spending 43% more time in app over the last 6 months, an all time high.
76
Jan 11
Call me when you want to know what they are saying. I want to see this for dolphins
Jan 9
we’re less than 10 years from decoding all animal communication and BCIs that let us talk back
an entire kingdom of minds we’ve been ignoring for 300,000 years
going to be really fun
83
Jan 11
X={pretending to be truth}
47
Jan 11
The Tao te Ching begins with a 2,500-year-old warning not to confuse the Address with the Data.
36
Jan 3
Of all the famous nonscientists I have met, PSH was the nicest to us kids—he only wanted heroin
34
Jan 3
Hypothesis: English embedding spaces carry a systematic anisotropy that privileges administrative-commercial frames (classification/governance/credentialed discourse) and compresses or distorts “liberation” frames that are non-agentic
37
The Lewises retweeted
30 Dec 2025
I am a Product Manager at Meta.
Last quarter I was assigned to "understand teenagers."
We built a museum.
Not a digital one.
A physical museum.
Inside the office.
McKinsey designed the exhibit flow.
It cost $2.3 million.
Photos of Taco Bell and mall food courts on the walls.
Instructions on how to take "wacky, teen-style selfies."
I am 34.
I stood in the exhibit called "Teen Hangout Zones."
It was a photo of a parking lot.
I was told to "internalize the lifestyle of my teenage targets."
Targets.
That's the word we use.
There was a display about TikTok trends.
Written by people who don't use TikTok.
The Slang Translation Wall still includes "on fleek" and "YOLO."
I sat in a bean bag chair labeled "Authentic Youth Seating."
I took a selfie in front of the Chipotle poster.
My manager said it "lacked authenticity."
We could have just asked a teenager.
But our user research team said that's "not scalable."
We have a Slack channel called #teen-insights.
340 members.
Zero teenagers.
Someone suggested we "touch grass."
That's now a $450,000 outdoor exhibit proposal.
My VP said the museum "understood the assignment."
The assignment was understanding teenagers.
We did not understand the assignment.
I am now 11% more aligned with Gen Z.
That's what the dashboard says.
The dashboard was also built by people over 40.
The exit survey asked if the museum was "bussin."
I checked yes.
I am 34.
Next quarter we're adding a "Teen Slang Zone."
It will be curated by our VP of Youth Strategy.
He has a teenager.
He hasn't spoken to her in three months.
But he's "very passionate about the space."
19
13
206
16,839