“I have seen the future of science. It is ruled by bitter competition instead of collaboration, pageantry instead of exploration. . . .Those willing to toil over genuine questions will necessarily lose out to those that can furnish cheap answers. . . .”
statmodeling.stat.columbia.edu/2026/04/20/94/
There’s lots of great use cases! I’ve had similar experiences and seen interesting outputs. I also don’t confuse that with a formal service-ready software.
I’ve had multiple people try to sell/pitch their vibecoded agronomic whizbang app. Those are security nightmares, to say nothing of the quality
I’ve already seen enough vibecoded websites to last me a lifetime. It’s like the last stop on the enshittification train.
youtube.com/shorts/CkXvh...
Pretty obvious that tariffs and the war in Iran are increasing the prices of farm inputs. But Admin doesn't want anyone to think about that 🤷♀️. farmpolicynews.illinois.edu/2026/04/usda...
The best version of Bob Dylan songs are the ones performed by others, without exception.
One reason soil erosion is so tough is that it occurs over long timescales. In these long term plots in Wooster, OH, you can see that reduced tillage and adding grasses in the rotation has prevented ~4” of topsoil loss in the past decades.
This always reminds me of the Wendell Berry quote…
“My understanding of what is best for [the soil] is the tragic understanding of hindsight, the awareness that I have been taught what was here to be lost by the loss of it.” -Wendell Berry
It’s hard to show a negative. Avoided erosion or avoided nutrient losses hard are tough, but oh-so important.
One reason soil erosion is so tough is that it occurs over long timescales. In these long term plots in Wooster, OH, you can see that reduced tillage and adding grasses in the rotation has prevented ~4” of topsoil loss in the past decades.
This always reminds me of the Wendell Berry quote…
Saw a thoughtful thread about AI, don't want to QT or argue. But. The biggest rage factor with LLMs is the people who, because genAI is transformative for coding, think it's transformative for everything else, because they devalue every other form of work and labor and knowledge.
It has left agricultural science stuck in-between, at least for anyone who is interested in being applied. The more basic scientists have (somewhat) migrated here and anyone that is farmer-facing -is either on still Twitter or just looking for other avenues.
Perhaps you received a mysterious noreply email asking you to evaluate some publications 'for novelty'. Looked kinda dubious? Yup, that's the one.
So what's up with this 'metascience novelty indicators challenge'? 🧵
“Everybody’s got something to hide except for me and my monkey” -The Beatles
#AshaniThilakarathne 's latest publication showing long-term, diversified #croprotations improve #soilhealth and help soybeans better withstand extreme #drought, reducing water stress and yield losses, compared to conventional systems. 🌱🫘💦
🔗https://doi.org/10.1016/j.agee.2025.110180
Ahhh yes, “frictionless” tech
Never thought a science journal could have a clever, relevant, and on-topic holiday message, but here we are! 🤓
It’s good to see papers start to address LLMs as structural plagiarism — provenance, more hidden than the original words or training data. www.nature.com/articles/s42...
Nobody is even willing to give Papyrus a chance these days 😢
Agricultural soil carbon sequestration gets a lot of attention but let’s not forget that the best way to keep carbon in soils is to stop further expansion of agriculture into natural ecosystems.
#WorldSoilDay
Regenerative agriculture is mostly overhyped as a climate solution. But in Brazil, I saw some regenerative grazing practices produce higher beef yields, which meant more money for the rancher and less eating of the earth.
Development vs climate action is a false choice.
In the agriculture sector, there are many actions that reduce climate pollution AND help farmers adapt to a changing climate.
An underappreciated one is helping smallholders sustainably boost productivity, reducing ag land expansion & deforestation.
Didn’t realize Don Quixote worked at CDC now…
Whatever your research question, I can assure you it is *NOT* answered by a combination of linear mixed models, random forest, and structural equation models.
Whatever your research question, I can assure you it is *NOT* answered by a combination of linear mixed models, random forest, and structural equation models.
We've somehow settled into an equilibrium wherein *all* the social media sites suck & are unbearable.
A photograph of a random forest in the Canadian boreal forest
Random forest
PNAS – Profile of Johannes Lehmann, Cornell University, University of Bayreuth, Germany. National Academy of Sciences Member.
PNAS highlights Johannes Lehmann’s pioneering soil science—from revealing the secrets of Amazonian dark earths to advancing biochar for fertility and climate solutions. His work is reshaping sustainable agriculture. Read the PNAS Profile: www.pnas.org/doi/10.1073/...
You mean to tell me that a model trained on average quality code is producing code that is of average quality? I thought we could spin straw into gold!
WE ARE THE ALGORITHM
