I'd suggest that the basic framework discussed here is good, but you need to be willing to be flexible. For example, not all experiments are testing a hypothesis, so this may not be relevant. Also, the important safety precautions may relate more to the method than the materials, so the order might (at times) need to be changed. Sometimes the calculations will need to be incoprorated into the discussion, sometimes they may be appropriately separated (or even tossed into an appendix). Basically, you need to be willing to modify some things to suit the experiment you are reporting.
