You know what the secret horror of every scientist is?
That they cannot, try as they may, reproduce their work.
Actually wait, I take it back- there are two horrors. First is that of reproducibility (if you can't reproduce what you did a few months ago, then who can?).
The second is this: getting brilliant results, then realizing there was a problem with the experiment design, expecting that results with the new and improved version will be even more brilliant and then realizing that nope, your results with this version are exactly opposite to what you predicted and what you got last time.
Ethical dilemma! Do you stick with original design because it gives you what you want, even though you cannot entirely explain why it worked that way? Or should you be good and swallow the bullet and admit that perhaps your entire hypothesis is incorrect?
But then, what if the assumptions made while forming the hypothesis were incorrect? Then, the results from the two designs should give you an unanticipated insight into the whole process. But then! What if you can't reproduce the damn results in either case? AAARRRRGGGHHHH
Where does that leave one? Cursing oneself for not writing down EVERY single detail in lab notes.
Humph.
3 comments:
i'll add a third one.. getting scooped!
I know the feeling of 1 and 2. A third nightmare is completing analysis on something, arriving at a conclusion/result. Then while writing up a paper/abstract etc, you go back and then realize that you don't know who you had excluded or why you had excluded one data point and you cannot get the same results back again!!!! :(
I think every researcher re-invents the wheel when it comes to maintaining lab-notes. I've heard this so many times now, 'I wish I had written down all the details.' It's happened to me too.
Did I tell you I've not been able to reproduce my initial crystallization conditions after the first successful attempt? I have no clue what was different in that one trial.
-Gotti
Post a Comment