A Critical Flaw In Scientific Standard Operating Procedures
In 2019, Pew reported a pediatric medication error that began with one small mismatch. A baby’s weight got entered in pounds rather than kilograms.
The system did not reject the number. It interpreted it in the wrong unit.
A weight-based calculation then used the doubled value, and the child later received twice the appropriate dose of acetaminophen.
No crash. No obvious alarm. One wrong node entered quietly, was transformed confidently, and emerged downstream as a critical error.
The same situatuon occurred causing the aviation incident when an aircraft was fueled up is pounds instead of kilograms causing it to run out of fuel halfway to destination over Gimly airport.
Get Dan Cleary’s stories in your inbox
Join Medium for free to get updates from this writer.
It landed safely due to excellent piloting but it shows again the danger of using the wrong unit VALUE.
What we do is build little isolated “quasi-consistent”, single use systems. Much of what we currently call “interoperable” is not truly interoperable.
It’s internally coherent enough to run, but isolated from the broader SI-based web. It can be locally compared to other things thst used the EXACT same process to form a ranking, but it’s dimensional information in SI units is gone.
A dimensional analysis isn’t enough to tell you anything. Once on the consistent, physics-wide, SI Planck system, dimensions are all but trivial. What’s important is the value. It will show you if you are in SI or are not.
For instance, any time an energy has been calculated using E=mc², the answer has always been broken unless Planck mass was used or the proper conversion unit was used.
“m” does not stand for ANY MASS. it’s Planck mass. “q²” is not e² or “any charge unit”. It’s Planck charge.
Use anything else and technically it’s no longer “math”, it’s just drawing. It’s art.
