Lab Notebooks (2020)

The article mentions, but doesn&#x27;t elaborate, on the purpose of the notebook as a legal document. This was much more important when notebooks could be used to show priority in patent disputes, but only if well kept. My parents were both research scientists. Where my dad worked, the company patent lawyer read and critiqued your notebooks, and maintained the discipline.<p>Many of those rules became unnecessary when the US patent system joined the rest of the world by adopting &quot;first to file&quot; priority. Rules such as: Always writing in pen, and crossing things with a single strike-through.<p>Notebooks are still a great way to preserve the little details of &quot;what the hell was I thinking?&quot; A colleague told me a good notebook prevents you from having to repeat a study because you can&#x27;t figure out what you did the first time.<p>There are still industries that require legally defensible record management -- you&#x27;ll know if you&#x27;re in that boat. If not, you can DIY whatever method works for you. The success metric is if you actually use it. On paper, I write in pencil because I&#x27;m more concerned about readability than a precise narrative. On the computer, Jupyter, just because nothing else really lets me document &quot;thinking in code.&quot;

I have seen meticulous, beautiful lab notebooks before. Paper ones. With numbered oversize greenish pages, faintly quadrille ruled, filled with exceptional INK penmanship, paragraph after paragraph of cogent technical observations, theoretical musings, responses to papers, thought-experiments and real experiments, taped graphs and images. Everything dated and signed. I&#x27;ve seen stacks of them spanning YEARS of work.<p>I tried. Mine look like shit. If it&#x27;s in ink, it&#x27;s peppered ALL OVER with correction after correction after correction. Perhaps it&#x27;s a matter of practice, but then I&#x27;ve not been at one place long enough where projects weren&#x27;t constantly re-shuffled, killed, or lasted long enough to get a grip (sometimes the job and the company itself doesn&#x27;t even last long enough).<p>Perhaps these are becoming a relic of a lost era? Paper is too much of an &quot;air-gap&quot; when everything is done on computer. Moreover, the standard 2-week Agile &quot;sprint&quot; leaves no time for careful thinking and writing. One can&#x27;t put epiphanies gained from writing up notes into &quot;the deliverable&quot; for a freaking Jira Ticket.

I recommend the Leuchtturm 1917 Notebook for this kind of stuff. It comes with spine labels and a table of contents pages in the notebook already, and the paper is extremely high quality. They&#x27;re the perfect daily-use notebook, IMO.

The main problem with paper is that it&#x27;s not searchable, so I&#x27;ve been using [0] since 2018. Each invocation just appends a timestamped line of text. I have a crontab that pops up a box every hour and I type in what I&#x27;m working on. Sometimes I&#x27;ll add extra notes if I&#x27;m working through a particularly thorny problem. It&#x27;s always right there on the terminal.<p>I&#x27;d say it&#x27;s saved my ass at least a dozen times over the years, where I hit a problem that I&#x27;ve seen before but can&#x27;t quite remember how to solve it. I also page through it quarterly to get a good summary of what I&#x27;ve worked on for that quarterly 1:1 with the boss.<p>[0]: <a href="https:&#x2F;&#x2F;gist.github.com&#x2F;mfischr&#x2F;f1ee2967ec0181b934639c30f4e68f17" rel="nofollow">https:&#x2F;&#x2F;gist.github.com&#x2F;mfischr&#x2F;f1ee2967ec0181b934639c30f4e6...</a>

I&#x27;ve started writing a markdown file every day, with a separate heading for each project, and a tool that automatically creates an index page that says what topics (based on headings) were present each day.<p>Apart from the format of the headings, it is deliberately informal and unstructured. I&#x27;ve found it very useful.

One advantage of lab notebooks that the article does not mention is that by design, the pages are firmly bound to the book, so you cannot lose them. I started to use lab notebooks because I would otherwise just write thoughts on loose-leaf paper and have a difficult time keeping track of the pages and their order. The lab notebook (or any notebook with firmly bound pages) solves that problem.<p>That said, I don&#x27;t use lab notebooks for coding&#x2F;software development. I tend to use Markdown files in a Git repository for that. That setup serves the same purpose, since I write down what I intend to do and whether I was able to achieve it or not, but it is much easier to use when coding in my experience.

Habit of lab notebook has stuck with me since college. Shoutout to reMarkable tablets and Neo Smartpens, for helping combine the analog process of handwriting with digital search. Likely a personal thing, just like to take hands off keyboard from time to time. Seen hints of research re benefits of handwriting, but never really dug into it.

I have started using Obsidian for daily notes (which you can categorize better than the default so it&#x27;s more findable) and you can tag &#x2F; create back links. I am thinking that these tips plus the tools Obsidian provides would make an excellent combo. Then the files can also be placed in git and backed up for protection!

Lab notebooks have saved me many times. The days run together in experimental work and having a record of what I tried or saw and on what day has paid off weeks or months later when a collection of observations snapped into focus. Never been able to keep a table of contents, though.

I wonder if we soon will get wearable lab cameras&#x2F;microphones that can log what you do while keeping your hands free. More than once, I have been working with both hands and suddenly been distracted for a brief moment just to discover that I no longer remember if I really added stuff to that well. Also, having more exact timestamps at each stage of an experiment might be useful for later

We call them “Engineering Notebooks” and after 5 years at a company they tell an interesting story of the projects worked on. I’m often reviewing notes from the past couple of months and occasionally have to go back a year or two to dig up some information that I know is there somewhere.

Highly recommend Muji’s gridded notebooks (€1.95 for A5-size, 30 sheets):<p><a href="https:&#x2F;&#x2F;www.muji.eu&#x2F;product&#x2F;recycling-paper-notebook-dark-grey-a5-7160.muji" rel="nofollow noreferrer">https:&#x2F;&#x2F;www.muji.eu&#x2F;product&#x2F;recycling-paper-notebook-dark-gr...</a>

I know of two places to buy genuine engineering lab notebooks, for anyone interested.<p>1. Scientific Notebook Company &lt;<a href="https:&#x2F;&#x2F;snco.com&#x2F;" rel="nofollow noreferrer">https:&#x2F;&#x2F;snco.com&#x2F;</a>&gt;<p>2. BookFactory &lt;<a href="https:&#x2F;&#x2F;www.bookfactory.com&#x2F;" rel="nofollow noreferrer">https:&#x2F;&#x2F;www.bookfactory.com&#x2F;</a>&gt;<p>Of the two, I slightly prefer BookFactory&#x27;s offerings. Note that there are many, many variants. Unless you are employed somewhere that mandates a particular type – and if you do, they are likely available from your office supply closet – pick one you like. Do go with the bound hardcover style, though. They&#x27;re much more durable than spiral or softbound.

I&#x27;ll probably just use an app, rather than real paper, but I&#x27;ll definitely keep this in mind next time I&#x27;m doing something experimental and not just &quot;Stuff copilot could probably do&quot;.<p>Preserving a record in detail seems like a natural fit, we&#x27;re used to doing the same for our source code, I&#x27;m surprised it&#x27;s not more common to do the same for the source of the source code.

I started working at a lab where I am the only SWE and everyone else is a scientist. I have a lab notebook, but I don&#x27;t use it properly (thankfully nobody gets on my case about it). I usually just use it as a short-term working memory, but it helps me keep track of the progress I&#x27;ve made. It&#x27;s nice to look back on something I was working on two months ago and see all my thoughts written down about which library to use or architecture to use. Looking back at my past struggles gives me a sense of accomplishment and progress.<p>That said, I usually try to preserve the results of my &quot;experimentation&quot; in documentation with the software, so I can share it with others. It&#x27;s a good sign when people never talk about X to me, and I end up asking them if they ever used X and they tell me &quot;I just used your documentation&quot;. Best feeling ever.

One issue is whether to organize lab notebooks by project, or by time. In the latter case, this means the notebook is a record of your day-to-day activity, which has certain advantages (like having a single record to refer to), and disadvantages (reconstructing a project&#x27;s timeline can be time-consuming if you&#x27;re working concurrently on different projects). A decent compromise is to have a daily notebook, but never put work on two projects on the same page. Then if you really want to reconstruct a project&#x27;s history from the daily journal, you can at least just copy pages.<p>The primary reason for having a lab bench notebook is that, if properly used, you can always reconstruct the process you were documenting. A lot of bench research involves protocol development, for example. In comparison with software engineering, a protocol is like an algorithm, only you&#x27;re applying it to physical samples. In a well-run lab, once someone has developed a protocol, they &#x27;publish&#x27; it (stick it in the lab&#x27;s manual of protocols, where other people can have access to it). So, you don&#x27;t share your lab notebooks with others, you share the completed protocol, and then if there&#x27;s some question about if the protocol can be improved, you can go back to the lab notebooks to see what was already tried (this seems similar to software library or device driver development).<p>Some people don&#x27;t like lab notebooks much, because properly recording what you did in enough detail that you can go back and repeat it eats up time and requires a certain discipline, and it often takes some unfortunate catastrophe (comparable to losing all your data backups) to learn the value of the practice.<p>Incidentally, if a research lab, public or private, has no guidelines at all on lab notebook practices, it&#x27;s an indication they may be engaging in shady behavior (e.g. Theranos) - it&#x27;s comparable to a crypto exchange outfit like FTX not having an accounting division, and should be a red flag for investors. This is also a reason for the ink-only rule, with pencil you can go back and fudge data more easily.

Does anybody have a reference&#x2F;example of a lab notebook for software engg ?<p>If I&#x27;m forced to log my thoughts&#x2F;actions it comes out as either too terse(everthing feels obvious) or it&#x27;s too verbose and it&#x27;s often difficult to find the right balance.

A huge downside of writing this down on paper is team collaboration. Imagine that instead this is kept in a tool like Notion which the entire team can access. At first this feels incredibly uncomfortable - your notes are there for everyone to see. However it’s a massive force multiplier because often the work is picked up by someone else later on, and instead of asking you they can self-serve. It’s pretty common that by the time they need it you would no longer be at the company.<p>I don’t have experience writing in a lab journal format, but for documents like growth experiments and how they worked, or RFCs, this is a godsend. It takes a lot of work to keep it tidy, but it’s worth it.

It reminds me an approach to data science and ML. It&#x27;s way more common and essential for ML teams to log their attempts (experiments). There is a whole set of tools for this - experiments trackers. Primarily to even being able to compare and pick the best direction, but also to ensure reproducibility (in some areas it might be required).<p>ML&#x2F;DS always seemed to me closer to science in its nature vs software engineering. Can be because of its nature as well - in a lot of cases it&#x27;s a process of incremental improvements (vs - simplifying this a lot - let&#x27;s say in SE we do a button that just works or not).

I do this for my personal software projects, research or otherwise. I have more stricter rules about paper and pen I use though.<p>A single fountain pen, filled with the same permanent, pigment or cellulose reactive ink, on a hardcover, ink-safe notebook.<p>However, the rest “at the speed of thought” rules are almost the same. Theorize, design and detail on paper, implement, document results, GOTO 10.<p>It’s a greet speed regulator, clarifier and keeps everything documented. As a bonus, you get “That weird guy” merit badge, which I don’t care.

I always have one of more pages of loose-leaf on my desk. It’s messy, but basically the lowest barrier to entry. And if I’m somewhere else and write something down, it goes in the pile with the other notes when I get back.<p>A few times a year, I open an org-mode file and start summarizing my notes. Those sheets then get put away, to help keep things a little less messy.

I use lab notebook inspired process to share information across our engineering organization. It is like your daily standup, but written throughout the day.<p><a href="https:&#x2F;&#x2F;bobek.cz&#x2F;blog&#x2F;2021&#x2F;written-standup&#x2F;" rel="nofollow noreferrer">https:&#x2F;&#x2F;bobek.cz&#x2F;blog&#x2F;2021&#x2F;written-standup&#x2F;</a>

I use lab notebook inspired process to share information across our engineering organization. It is like your daily standup, but written throughout the day.<p><a href="https:&#x2F;&#x2F;bobek.cz&#x2F;blog&#x2F;2021&#x2F;written-standup&#x2F;" rel="nofollow noreferrer">https:&#x2F;&#x2F;bobek.cz&#x2F;blog&#x2F;2021&#x2F;written-standup&#x2F;</a>

100% yes.<p>And a new page per day (paper is cheap), with the date always on top outer corner of the page. Never share a page across dates.<p>Thoughts, however bizarre, around the topics are very useful. Elaborate where possible. Names, keywords and classifications will be useful later, so do consider the future as you write.

I get these, modified to be regular notebooks, from <a href="https:&#x2F;&#x2F;snco.com&#x2F;" rel="nofollow noreferrer">https:&#x2F;&#x2F;snco.com&#x2F;</a><p>You can get them custom printed with whatever paper you like, custom names, titles and numbering on the books, for no more cost than a regular notebook.

A good reminder to get back to keeping lab notebooks - they are extremely useful for being able to look back in time and go &quot;What was I thinking and where did that number come from?&quot;

Tangential: I find the website design to be particularly nice !