Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: [Phys-L] Upper Division Labs and ELN



On 3/30/21 1:39 PM, John Sohl via Phys-l wrote:
1. What are other physics (or chemistry) departments doing in this
respect?
2. If you are using an ELN, do you have a preferred one and why?
3. If you are going digital but not with an *actual* ELN what are you
using? (OneNote for example.)
4. Any general thoughts? Flame wars? Advantages? Disasters? Whatever.


To answer some slightly different questions:

A) IMHO at the most elementary level, an ELN is not necessary.
It's a complication that beginning students mostly don't want
to deal with. OTOH if they do want to, that's OK.

B) At the opposite extreme, in a real-world research situation,
theory and/or experiment, it's all electronic. The volume of
data is too big to deal with in any other way.

Also almost anything in the real world is a team effort, and
electronic systems facilitate sharing without fratricide.

C) Any ELN worthy of the name should provide an audit trail.
That is, it's OK to change stuff, so that you can see the
latest & greatest version at a glance ... but all the previous
versions have to be kept available.

Hint: If you invent anything of important, the patent and the
inventorship will get challenged in court. Being able to document
the process and the intermediate steps that led to the invention
is priceless. The schnook who is challenging won't have all that.

D) Conversely, almost anything that has a solid audit trail can be
adapted to the purpose. In particular, I've been using the git
program for about 15 years, and similar source-code control systems
for many years before that. Such things are optimized for text,
which is a big part of what I do, but drawings can be stored as
text (e.g. svg), and data can be stored in the same system with
not much penalty.
https://en.wikipedia.org/wiki/Git

E) Be sure to keep enough metadata so that you know what the numbers
mean. This applies to notebooks in any form, electronic or otherwise.

F) Regardless of format, the notebook is not enough. You need two
things:
-- Real-time visualization of what's going on, via gauges, scopes,
strip-chart recorders, et cetera.
-- A file containing the raw data.

Neither is a substitute for the other. And it's not optimal to create
the data file by having humans digitize the gauge readings. It's better
to use an A/D converter and store the data automatically into a file.

You would think this would all written down somewhere, but I haven't
seen it. You would think this would all be common knowledge, but it
isn't.