Re: Using org-mode for laboratory notes.

[John Hendy <jw.hendy@gmail.com>]

On Wed, Sep 19, 2012 at 8:44 PM, Torsten Wagner
<torsten.wagner@gmail.com> wrote:
> Hi,
>
> one way which works rather differently is the combination of git and org-mode.
> You could write your protocols in separate org-files and link to them
> in your records.
> org allows (at least it did a while ago) to link not only to a file
> but also to a specific version of a file.
> You could do small modification in the protocol-files as you need them
> and check them into the git system.
> Link to them and you will see the version you used for exactly this experiments.
>
> Actually using something like git and a git sensitive link is
> important if you might plan link to a lot of external files. Imaging
> you overwrite a file by accident or because you can't remember you
> referred to the original file already. A normal link would quietly
> point to the new file and would not be in-sync with anything you
> mentioned in your org-file.
>
> Other benefits are gits diff, merge and change-recording capabilities.
> If you set-up the git repro with entire lab-book on a server (a PC
> reachable from all your other devices) you could easily add data from
> within the lab, go to your office to add more data and at a certain
> point merge all this together. Both PCs could work offline and only
> need to be online for check-in and check-out new data.
> Another benefit of combining org-mode and git... you can tag certain
> versions of your lab book. E.g. tag them whenever you write a paper
> and make a notice in org-mode. This enables you to get back to all the
> measurement and reps. data evaluation results as you found them during
> writing your paper, even years and many many changes later (e.g. you
> might improved your data analysis method over time but for the paper
> you still want to see the old stupid way how you dealt with the data).
>
> Recently we got an org-file sensitive git-module, which makes merging
> org-files much more nice.
> Check here:
>
> http://orgmode.org/worg/org-contrib/org-git-link.html
> http://orgmode.org/worg/org-contrib/gsoc2012/student-projects/git-merge-tool/index.html

Thanks for re-bringing this up. The video demo of org-merge was
*amazing.* I should really try that. Git scares me a bit, but it seems
amazingly powerful. I really like the idea of being able to have tons
of versions to look back to, especially if I start rearranging and
wonder where I put some tidbit. I've done a few org file overhauls
which have both helped and hurt for this reason. I know something
existed, but then can't find it.

>
> Albeit I have to say I like to do all kind of stuff in org-mode, I
> faced problems using only org-mode as lab-notebook. Sometimes things
> in a lab are to numerous and to verbose to type them all in as they
> happen. Sometimes a little sketch, some quick scribbled note, etc.
> contains the real important data.... sentences like "Damn, Joe slammed
> the door AGAIN", during an AFM measurement. Or in your case "Uhh..
> what are the funny little flakes in my buffer solutions".... I think
> sometimes a keyboard still filters to much and hence org-mode might
> not really contain all the necessary info.

True, though one could document via voice recorder and link to the audio file.

I also sketch on a whiteboard, take a pic with my tablet, and then put
the file in a folder called "ip-pics" which serves to hold a huge
repository of any experiment, equipment, or sketch pics used for my
intellectual property notebook. I'll trade having to do a bit of file
and device maneuvering (taking pictures/sketch on x, getting it into
the right directory for Org to find) for all of the other stuff Org
provides.

In other words, imagine gaining the ability to sketch on pen and paper
and losing the ability to
- easily rearrange
- tag
- log time
- create awesome exports with minor modifications to notes vs. typing
from handwriting
- the list goes literally on, and on, and on

> As a summary:
> org-mode as a lab-book will work fine if you are strict in using it
> and force yourself to be verbose enough.

Yup. I think writing more than you think you do and trying to
anticipate one's future self is huge. When I file contacts, for
example, I try to think of any possible words I might use. For a
machining shop, things like "machining, machin, mill, lathe, molding."

>
> Hope thats helps
>
> Totti
>
> CC. There might be some legal issue with real lab books and electronic
> once. Back in the good old time where scientists didn't publish each
> and every result and where it could take many many months until some
> discovery reached the other side of the planet, the laboratory books
> where the legal evidence of the original work. If someone made a wrong
> claim or someone accused someone else of falsify reporting, the
> laboratory books where used to proof those claims. That is the reason,
> you find e.g. nicely archived laboratory books of all the great
> scientists of the Bell laboratories. Not sure how much this is still
> relevant today.
>

The large R&D company I work at is currently experimenting with a few
e-notebook vendors. They still serve much of the same purpose.
"Filing" an entry electronically timestamps it. I'm sure there could
be issues with fraud, but paper notes can have the same issues. After
all, I write my own dates on my own notes. This is the point of a
witness, and the witness signing date is what really matters, not when
I enter it. We're encouraged to get things witnessed asap.

The US is undergoing the move from first to invent to first to file,
so most of this will be irrelevant within the next 1-2 years. The old
practice was that one needed to prove first inventorship. Now it wil
just matter who gets to the patent office first, so long as they've
reduced the invention to practice and have actually been using it.


>