Comment by mikeskim

I would not be surprised if most academic CS research is not reproducible. This is true for many other fields outside of CS, and I've seen it first hand in machine learning. It's a problem but it's also just how things are.