3 ms·
It looks like an example of the "I have a PhD in X, I don't need to learn from CS or other fields to find a good solution". In my day job as a research software
by bafe 3y ago
It looks like an example of the "I have a PhD in X, I don't need to learn from CS or other fields to find a good solution". In my day job as a research software engineer I encounter plenty of these cases
- contravariant 3y agoLast time I looked there's at least several tens of papers of chemists rediscovering that getting a chemical equation to balance is just finding the null space of the corresponding matrix. Maybe I'm being too critical, but really I don't understand why papers on the subject exist in the first place. Finding null spaces is solved, you just need the one paragraph pointing out the equivalence between the two problems, there's no real need to investigate much further. I mean there's only 90-ish elements that can reasonably participate in a chemical reaction in the first place so the largest linear equation is pretty small, there's not even any need to go looking for a more efficient calculation.
- bafe 3y agoI recently experienced something similar with people who wanted to find certain functional groups in a molecule with a regex search against the SMILES representation. It took a lot of persuasion to explain them that string search is a terrible idea for subgraph matching