What we’re reading, March 27, 2026
Farewell to Asimov Press, clinical trials value, and Austin's housing success
Hope you’ve had a great week! Here’s what caught our attention:
The very best biology magazine in the world, Asimov Press, announced they’re winding down (for now). I’ve been an advisor to them since before they started, and I’m very sad to hear the news, even though I’m looking forward to what the team behind it – Niko McCarty and Xander Balwit – are doing next. At Asimov, they published some of the deepest and most thoughtful writing on biology and actually made it feel enchanting – so unlike the textbook descriptions most people encounter at school. It’s the kind of writing I’ve tried to achieve in my own work. They covered topics across the history of technology, biological curiosities, ideas for science policy reform, and visual explainers of how molecular processes work. Here are some of my favourite pieces: Origins of the lab mouse, Making the micropipette, Where’s the synthetic blood?, Before they hatch, Driving Toward Nanopores, and you know what, I’ll even include a piece of my own, Measuring the Black Death. At the end of their farewell post, they leave open the possibility that it might be revived in the future; I really hope it happens one day! — Saloni Dattani
How about one more from Asimov before they go: this week Alvin Djajadikerta published “Designing AI for Disruptive Science.” In it, he explores the possibility that (current approaches to) AI-for-science might enhance our ability to predict within current frameworks while simultaneously weakening our capacity to identify and shift into entirely new paradigms. He lays out histories of important paradigm shifts, discusses some interesting papers on technology-based scientific narrowing (arising from both AI and, interestingly, the internet), and closes with a call for more and better study of the mechanisms of science – perhaps using AI agents as a metascientific model organism for better understanding the effects of different institutional and incentive designs (I, for one, would immediately subscribe to a science-focused AI village-style experiment). It’s a thoughtful and thought provoking piece, and well worth the read. — Jordan Dworkin
I’ve been revisiting some older papers that might shed light on the benefits of faster clinical trials. Let’s start with a short 2007 paper by Frank Lichtenberg, which looked at the correlations between deaths from various diseases over time, and the share of medical prescriptions for that disease relying on newer medicines. Lichtenberg found that mortality was lower for a given disease when more of its prescriptions relied on newer medicines, and he used this to back out an estimate of the consequences for US health if those new prescriptions had not been available. For example, in 2003, he estimates there would be roughly 1.6 million fewer life-years if drugs invented since 1990 had not been available. With roughly 380 new molecular entities approved between 1990 and 2002, that works out to 4,200 life years lost, on average, per drug: i.e., if a drug invented after 1990 had not been available during 2003. Just one year later, Philipson et al. (2008) looked at the impact of the Prescription Drug User Fee Act of 1992. Philipson and coauthors argue the act sped up drug approvals by 6-7%, and generated a surplus (the gap between what consumers would have been willing to pay and the cost of producing medicine) equal to $14-31 billion. — Matt Clancy
In most product markets, transitory supply and demand shocks can spike prices, but as long as there’s open entry to the market allowing elastic supply, prices come back down near marginal supply costs. This is rare in American housing markets because infill supply responses are usually illegal under local growth control laws–but not, per Pew’s new analysis, in Austin, Texas. Austin enjoyed among the highest permitting rates in the country for several years; as in-migration has slowed somewhat, supply finally caught up to demand and rents are back down to pre-pandemic levels in nominal terms and down significantly in real terms and as a share of incomes. Austin’s biggest zoning reforms actually came during the boom, allowing Austin to keep permitting at pre-2019 rates in 2025 despite substantially lower real rents. That said, once rents fall to near construction costs plus a competitive gross margin, development will slow substantially as it has in Minneapolis (where median market rate rents are now at federally subsidized LIHTC levels). — Alex Armlovich
More progressives in the housing space are coming out against the Senate’s build-to-rent ban. The growing concern is that the ban could reduce rental housing supply when affordability remains a pressing issue. It’s a tough situation and we hope all stakeholders come to a pro-housing solution with sustainable politics. — Alex Armlovich
“You will be hard-pressed to find a true admirer of Excel,” writes David Oks (who I first noticed as one of the Gravel Teens and is now at a16z, truly an ideological evolution for our time). But Oks is, if not an admirer, then surely a believer that Excel and predecessors like VisiCalc and Lotus 1-2-3 are inventions of deep historical importance. The most intriguing part of his argument is the claim that the rise of finance in the 1980s, particularly leveraged buyouts, was driven in large measure by the dawn of the spreadsheet. Before, “analyzing a single company would take weeks”; now it might take a few hours, and your model of that company could be amended in seconds. I’m not fully convinced (is the spreadsheet a driver of financialization or did financialization provide a larger market that spurred better spreadsheets?) but it’s an intriguing case study of the ways in which a particular technology’s diffusion can help reshape the economy, or perhaps solidify a reshaping already occurring. — Dylan Matthews
The nation’s largest offshore wind farm to date, the Coastal Virginia Offshore Wind project - which clocks in at 2.6 GW - sent its first commercial power to the grid this week (it joins Rhode Island’s Revolution Wind, which sent its first electrons onto the New England grid earlier this month). In stranger wind news, Interior announced a ~$928 million settlement with TotalEnergies to surrender two Atlantic wind leases. Jake Bittle & Rebecca Egan McCarthy have a pragmatic take on the deal - the payment mostly reimburses lease fees TotalEnergies paid in 2022, and the “redirected” oil and gas investments (Rio Grande LNG in South Texas and conventional production in the Gulf of Mexico) were already committed. Neither offshore lease was under active development or even near to completing federal permitting, and the deal doesn’t preclude future re-auction. Encouragingly, permitting reform negotiations in the Senate have apparently continued undisturbed. — Willow Latham-Proenca
Meanwhile, Jordan Dworkin published Who Will Program-Manage the Program Managers? this week on this blog, exploring how science funding agencies can build capacity to support ambitious, unconventional research programs. If you’re interested in this space, ARIA (the UK’s Advanced Research and Invention Agency) is currently hiring.
Our grantee Greater Greater Washington published their take on DC's draft Future Land Use Map. The plan would add only 15,000 housing units by 2050 and confuses theoretical "zoned capacity" with economically feasible development sites. By carrying over sites from the previous Future Land Use Map, DC's Office of Planning is banking on lower-quality development opportunities left after the best sites have already been built. The silver lining: the draft was bad enough to draw mass public protest and condemnations from both leading mayoral candidates.
The Irish government is seeking volunteers for their Clinical Trials Advisory Council. If you have relevant expertise and interest in shaping European regulatory frameworks, consider applying.