I'd like to find a best practices document to share with the developer of a internal DB interface. Specifically I'd like to show that the approach that has been used for 15+ years results in output files that require a lot of editing to easily read into R, python, gis.
see screenshot


  • The screenshot looks like an Excel sheet. Is it an Excel sheet and you have problems on reading the header you want to read? Or am I missing something? Is it CSV? Jan 31, 2017 at 20:14
  • 2
    When people write answers staring with I think you are asking about, you really need to edit your question and make it clearer. Why are you mentioning the DB interface? What are we looking at there? What is require a lot of editing to easily read (describe the edits, describe 'easily')? What kind of output files?
    – user4293
    Feb 1, 2017 at 7:45

1 Answer 1


I think you are asking about what to do with headers in tabular data, if so, here are my thoughts:

Headers should be one column/one row and only one column/one row. If you have to label across rows, simply add the name of the label to the rows you want to apply it to.

This does lead to extremely long headers, so also another thing to apply here is some deductive reasoning. Do you truly need to say all of that in the header? In some cases, headers are redundant. And in these cases, I simply delete the text entirely. When they are necessary, I go with the longer headers.

Just my opinion. I've found one row/column headers to be much more easy to work with across many platforms and technologies.

Not the answer you're looking for? Browse other questions tagged or ask your own question.