In research, proper maintenance of scientific data and records is necessary. Which way one could preserve his/her own records suitably so that it can be recalled at any time in future.
Keeping everything in files with highly descriptive file names, in directories labeled by date helps quite a bit. Usually if trying to retrieve a record you have some idea of the month it was done.
Drawing from programming principles, the concept of "gardening" where one is constantly tidying up ones own work as you work on it probably applies quite strongly here. Much of my data often starts life as a rough quick look and as time goes on I make a point of spending time adding in the important context details so that in a few years time I will still understand it. Principles like always labeling axis on graphs, always including units and headers on columns, and most importantly, attaching to any dataset the conditions used to create the dataset. Often can lead to replication but digital storage is cheap.
Some kind of tag system would be ideal but I am not aware of any software for doing this efficiently.
Small advice. I try to keep the original date of experimental data. That is, processing of the data I produce with copies of the original files. Thus date of the treated files can be much later than the original, but the original is always left with the start date. When you need to find your data you specify when they were received reading the thematic entries in the laboratory logbook. Knowing this date it will be easy to find the required entry in the computer.
- Badges
- Science topic
- Similar topics
- Engineering
- Computer Engineering
- Computer