There is still time to register for CIDR 2026 in Santa Cruz! If you need a roommate for the conference, there is also a spreadsheet you can use to find someone! www.cidrdb.org/cidr2026/reg...

why i pay for the verge

Kyle Kingsbury is not a journalist. He is not an op-ed writer. He is a computer safety researcher. And he has written one of the most compelling, comprehensive accounts of the ongoing hell in Chicago that you could possibly imagine. In under 1600 words. aphyr.com/posts/397-i-...

As an academic who has named papers after the classic mortal kombat ninjas (eugenewu.net/ninjas.html), I am obligated to share that the new Mortal Kombat II trailer IS AMAZING! www.youtube.com/watch?v=a9pH...

1/ Where does *academic* database research go from here—in a world dominated by AI, $100B in private investment, and an NSF budget cut in half? Where should focus go? At SIGMOD 2025, we hosted a panel to discuss, summarized on the SIGMOD blog: wp.sigmod.org?p=3828 🧵 Key takeaways from the panel:

Columbia's Data Management Group just wrapped up @sigmod2025.bsky.social with a strong showing: 4 papers, 2 demos, a keynote, a tutorial, and a panel. Highlights include a test-of-time award, best paper at DEEM, and best industry demo awards! A full write up is at: eugenewu.net/general/2025...

The US government is aiming to get rid of a quarter million people involved in science research and education by 2026. I have no words.

software craftsmanship in action.

Working on [email protected] keynote slides. Thinking about human data interaction, AI, and a virtuous cycle. Looking forward to seeing folks in a week!

the following 30 seconds from perfect movie REAL GENIUS are a rock solid vision of the future of education youtu.be/wB1X4o-MV6o?...

Almost every db faculty that has received an nsf award ad benefitrd from hector and sylvia’s work. Well deserved!

Go columbia db! Congrats to luis and papa john!

Could not have been clearer about this www.theverge.com/24282022/kam...

Excited for Phil Cascado from Outropy to join our class to talk about systems problems that arise when building agentic systems for real. AI is great for demos and hard for products. Class: W6113.github.io

Excited for Phil Cascado from Outropy to join our class to talk about systems problems that arise when building agentic systems for real. AI is great for demos and hard for products. Class: W6113.github.io

Dishonest charts are coming from all directions, and they're only going to hit harder. Unless we push back. So, an interactive guide: Defense Against Dishonest Charts flowingdata.com/projects/dis...

Really excited for shreya to speak about docETL, and in general LLM powered data analysis and the role of agent-based optimization, in our graduate seminar Agents For Real Class website: w6113.github.io

Exciting opportunity at the Columbia DAPLab! We are looking to hire a postdoc to work on AI Agents at the intersection of systems, AI, and HCI for Fall 2025. 🔥 Check out the details in the job post here: daplab.cs.columbia.edu/positions #ColumbiaUnversity #AI #AIAgents

The disparity between the ease to write an EO and the effort needed to push against it seems like a problem.

Mechanisms for user agency and control over their data will continue to grow in importance and visibility!

Looks like science giving us the finger. Sometime certainly feels like it

End of semester faculty reviews from McSweeny’s. www.mcsweeneys.net/articles/a-f...

ok last one. i swear

Mad Lad in class wrote this story about Codd with an odd bod.

I asked my students in introduction to databases to create database memes. Here are some bangers: I remove all data redundancy, Might not preserve some functional dependency, Cluttered databases I do reform, 'Am your good ol' friend Boyce-Codd Normal Form.

A helpful gift shopping algorithm that I made for @theguardian.com a couple of years ago.

the best present for your kids: white lies for a white christmas?

We are excited to be starting a new lab that rethinks the systems, ML, and HCI stack to support agent-based automation! Columbia's PhD deadline is 12/15 -- please share with potential applicants! columbia-dap-lab.github.io

Values need to involve a tradeoff. If they don't come with a cost you're willing to accept then they're not actionable. e.g., "Move fast and break stuff" = 🔥 "Move fast with stable infra" = 😐

Planning and arguing for a project (whether a grant or paper) is hard. One reason is that writing and speaking is a linear process, while the arguments for a project is a graph. This blog post describes a mental model and its implications to writing. eugenewu.net/writing/2024...

Thanks for the shoutout! Our perspective is first figure out how to separate complexity from the data out of the visual representation before trying to attribute complexity to the person.

Really happy to invite Yoonjoo Park from UIUC to virtually visit Columbia's data management seminar to talk about really nice work on checkpointing data science systems.

For better or worse, some quick tips on the PhD application from a faculty's perspective (N=1). eugenewu.net/tips.html

Really excited for C Mohan (inventor of ARIES recovery) to visit Columbia and share his wisdom about cloud dbms systems!!