QUERY_

Machine vision for poor people, or yet another way to misuse Quantum GIS! I wanted to know if color of DNA extract correlates with sequencing success, and to get the colors I imported a photo of extracts into QGIS as if it was a map, placed sample points, and measured their underlying RGB values.

There is no "science-wide replication crisis" because every word in that phrase carries unfounded assumptions. 🧵 1. There is no evidence that progress has generally stalled in the sciences. Also, no one has actually tried to estimate replicability across fields.

the beauty of humans is that even if we explicitly attempt to copy something, the noise of our lives and our brains seeps into the work. you still end up with a New messy synthesis of things AI doesn't live a life, it may have system noise, but it doesn't have a story to tell; because its software

A press release is out about the Biodiversity Heritage Library. From 1 January 2026 the Smithsonian will no longer host the administrative functions of BHL. BHL is looking for a new home. For further information see blog.biodiversitylibrary.org/2025/04/new-... #ILoveBHL #BiodiversityHeritageLibrary

metsamaja on liminaalnemaja

Spent the last 2 weeks sampling dozens of type specimens in a well maintained, almost fully digitized, and just lovely TAAM fungarium (Tartu).

Six years ago, we published a (thus far underappreciated) study where we showed how scientific communities may generate a literature populated by irreproducible results or may converge on many perfectly reproducible yet false findings *in the absence of QRPs*, challenging popular narratives.

Typology of matchboxes from Leiden's fungal type collection

Evergreen reminder that clades and taxa cannot really be "basal" or "early diverging". doi.org/10.1080/0027... Also see these nice blogposts (and other works!) by @iochromaland.bsky.social - for-the-love-of-trees.blogspot.com/2018/04/tree... - for-the-love-of-trees.blogspot.com/2018/04/tree...

two-dimensional specimen 💌

‪When did herbarium in Leiden become this big? Trends in collection of regular specimens vs. types; fungi and lichens; notable collectors and expeditions. (Mind that only ca. 20% of fungarium is digitized.)