We have had this project at work that has not been working I won't go too much in the details for obvious reasons

We are facing a contradiction, we want to organize and extract relationships between extremely long, specialized, texts (and sometimes extremely short texts), find how they relate, how their parts relate internally, and so on. The higher ups had a vision and it simply didn't match reality.

So I told them that instead we should look at all real data we have, study them to be able to find what kind of structure they embed and see if some of them are reccuring and systematic (because we want to be able to do some logic reasoning on said structure), hence going from the particular to the universal. And that if we cannot find said structure (resolve the contradiction between having a bunch of unstructured text and the need for something upon which we can reason systematically), then our approach to the problem is wrong and needs to change.

And they agreed

I know it's not much, but it's something, and I felt like I was applying a tiny bit of On Contradiction in real life.