One of the features I always like in the print edition of Seed is the lab notebook pictorial. Every month (or, at least, all three of the months that I've looked at the print edition), they publish a reproduction of a page or two from the lab notebook of a working scientist. It's sort of cool to see how they differ from one field to another, while remaining largely the same.
Back when I was doing the "A Week in the Lab" series of posts, somebody asked me about my own lab notebooks. I present here the reason why Seed is never likely to ask me to supply notebook pages for their monthly feature:
Those two pages are about as cool as my notebooks ever look. They're certainly never as neat and well-organized as what you see in the pages of Seed. More examples, and some explanations below the fold.
The most useful pages in my lab notebooks are usually like the left-hand page above, which shows the re-configuration of the vacuum chamber I discussed back in December There's two things on the page here: at the top, there's a diagram showing how the various bits of vacuum hardware are all put together. The point is to have a diagram showing what all the parts are, and where they fit. This is primarily of use for future students who need to learn, as I can identify all the pieces by sight, but sometimes, I need to reconstruct what the system looked like at a particular time in the past.
At the bottom of that page is a diagram showing the arrangement of pieces I used to align the lasers and detectors so as to guarantee that the detector is looking at the right part of the chamber. Again, this is primarily of use to future students, who might need to have the alignment technique clarified for them. (Though I'm more likely to draw a new diagram on the whiteboard in the lab than flip through the book to find this page...)
The right-hand page from the spread is more of the same. This time, it's showing the arrangement of laser beams going into the chamber. There are two different lasers that need to cross in the interaction region of the chamber, and they come in by slightly different paths. The paths are marked by two adjustable irises, one in each beam. The scrawled note saying "NOT CO-LINEAR" is a reminder that no one laser is supposed to pass through both irises (which is a fairly common trick, as two irises define a beam path more cleanly than one). I can easily see some future student spending hours and hours moving mirrors around to put one beam through both irises, and then wondering why it goes off and hits the wall, without going anywhere near the chamber, and I'm trying to head that off.
(The two irises are also labeled on the table itself, with pieces of masking tape with "811 nm" and "819 nm" (the wavelengths of the two different lasers) written on them. Not that I expect that to help.)
The bottom of the page shows a couple of sketches of the detector. I probably should've printed out copies of the pictures I put in the old blog post, and pasted those in, but that would've been a hassle.
Those pages are probably the most visually interesting in the notebook for that experiment. Pages with nothing but text are much more common. In this case, we have a calculation attempting to determine the number of photons hitting a particular detector, which I'm trying to use to determine whether my incredibly expensive krypton resonance lamp is still working properly, or if it's got crud on the front window that's cutting the flux down.
The calculation on that page is very much a back-of-the-envelope job. I did this two other times, on the backs of odd scraps of paper found lying about the lab, and had trouble reconciling the results each time. I finally decided to be all responsible and stuff, and actually write the whole thing into the lab notebook, with enough detail that hypothetical future students could follow the process.
Much more typical than the written-out calculation are pages like this one. There's useful information here, but it's not necessarily obvious what's going on, or even what these things refer to. The gibberish down at the bottom of the page is a list of file names (data traces saved off our Tektronix digital scope, hence the "TEK" at the start of each name), with the parameters for each run. The actual data is in a folder on my hard drive somewhere.
It's sort of the worst of both worlds-- the data is no use without the lab book, and the lab book really isn't much use without the data. That's ok here, though, because the data from that day was all pretty useless-- we didn't see a damn thing. when we actually get useful data, I'll sometimes print the graphs out and tape them into the book on later pages, but I didn't scan any of those in for this post. If you want to see the sort of graphs that get put in the book, check out this Week in the Lab post.
And then there's this...
This page is part of the data that went into the graphs I mentioned. At this point, I was working pretty quickly, and just scribbled down tables of numbers as I made measurements. It's a little more comprehensible in context-- the facing page, which I didn't scan, has a more clearly labeled table-- but really, this is kind of a mess no matter what you do with it. Happily, these were calibration measurements only, and not the sort of thing I'll need to repeat, so the really sketchy information here isn't that big a problem-- it only needed to make sense long enough for me to carry the book upstairs and fire up SigmaPlot.
I could probably reconstruct the meaning of this page pretty easily, but any student looking at it would be hopelessly lost. But then, I don't expect students to need to do anything with this information, and they're going to produce their fair share of pages just like this one (that leave me wondering "What the hell was he doing here? I mean, seriously, are these phone numbers?") in turn.
So, there you go: the art and science of lab notebooks. Or, at least, the slightly disorganized lab notebook of one particular physicist with really bad handwriting and no artistic ability to speak of...
(Sorry for the poor image quality-- they're blue-lined graph paper, and written in blue ink. It's clearer on the actual pages, but the scanner doesn't pick it up that well...)
- Log in to post comments
Hey, that one at the top left is the Flying Spaghetti Monster, isn't it?
Don't be to hard on yourself...our lab books are just text and numbers, not even any pictures! :)
Hey, that one at the top left is the Flying Spaghetti Monster, isn't it?
Must be a miraculous apparition!
My lab book has been blessed by an appearance of the Flying Spaghetti Monster! That's even better than seeing Jesus on a grilled cheese sandwich...
That's a great feature! One of the phases every bench scientist goes through -- and it can last years -- is "how the hell am I going to keep my notes?". There must be nearly as many systems as scientists -- I must have tried four or five myself.
These days I use graph pads like yours, only with holes for a ring binder; I number my experiments and keep an index in the front of the binder and a few appendices (whereabouts of valuable biologicals, info on my shorthand for any unfortunate who needs to read my notes, supply info for reagents) in the back; and I stick all my raw data into the notes folder, or if it has to go in a separate place (like big chunks of autoradiograph film) I cross-reference that storage location in the notes folder, in an appendix titled "Dude, where are my data?". Using ring binders makes it possible to keep the expt index and various appendices organized, and it sounds horribly complicated but in fact it's a very simple system that still prevents me from ever losing or forgetting anything.
I was all set to post an entry of my own like this one, then I remembered -- it's probably an hour's work to find a page that isn't covered in profanity and frank admissions of mistakes spotted, total gobsmacked inability to understand the results, or both.
Over here at the Wonder Drug Factory, our notebooks are basically all electronic. We have to print the experiments out and have a physical copy taped into a notebook, to get it witnesses, but all the experimental details are done on the computer.
This helps out in a lot of an organic chemist's work: calculating weights and volumes of reagents, etc. And when you repeat experiments, it's much easier - just hit repeat, and enter in the amount of your first reagent this time (or a factor to multiply by) and the whole thing redraws and fills in (except the written details, of course). Since all our drawings are chemical structures, which have been done by software for 20 years now, we don't have any problems there, either.
And in industry, they teach you not to editorialize in the lab writeup, so as not to cause trouble in potential patent disputes. "Product not seen by TLC" or "NMR doesn't show methyl group" are OK. "Reaction didn't work" isn't. 'Cause you might be wrong.
Have you noticed that your illustrated notebook pages look a lot like Voynich manuscript?
And in industry, they teach you not to editorialize in the lab writeup
That's not "teaching" anything; it's a demand that the whims of the moneygrubs be allowed to determine the work practices of scientists.
'Cause you might be wrong.
Of course I might be wrong; in what bizarro universe is that a problem? I'm a research scientist. Being wrong (enough times to finally be right) is practically my stock-in-trade. Telling me not to collect my day-to-day ideas in the most convenient and useful place for them is not going to help me get it right.
Process chemistry in pharma manufacturing is 90% engineering and 10% science.
There is a huge regulatory burden so everything you do there has to have a paper trail. If you make a bad batch of the drug active substance, or fail manufacturing practice certification inspection, it can have incredibly expensive consequence. A small fire or spill from bursted feeder can shut down the plant unit for weeks (=>investigation) even if there is not much loss on the equipment. Etc. So it is not just a whim of moneygrubbers and pencilpushers when you manufacture stuff that makes perhaps $600M/y profit for the company.
I think bigger problem with bossy cover-my-ass non-scientists can be in medicinal chemistry research - it takes for example just one difficult safety officer (who is determined to "go by book") and he can make your life miserable and prevent you from doing any meaningful work.
This article is very useful for me as I plan to create my own ELN and needed examples of real paper-laboratory notebook pages.
Thank you very much, I'll be able to create template for my ELN.
Regards,
Roland