Address comments on E05

[korenmiklos]

• What's Etherpad?
• '[This] URL has to be in quotes because it is full of strange characters.': technically, this particular URL works without quotes, at least in UNIX. However, I would also consider it to be good practice to but all paths, absolute or relative, into quotes because it is more robust. Also, the option en causes an error for my Stata (Stata 16, Mac) and does not seem to be a standard option at least in my version (see help import_delimited##import_delimited_options).
• 'Note that reshape changes the dataset in memory and you cannot undo it. Make sure you know what you are doing.': You could consider to mention preserve and restore at this occasion which allows to play around with the dataset without having to save it to disk first.
• I would suggest to qualify the 'Never do a many-to-many, m:m merge.' statement by writing that many-to-many relationships are hardly encountered and that the student should be well aware of its applicability in their particular use-case.
• Possibly it is too obvious, but you could point out that
  • merge commands are typical steps in which there is a chance that you (significantly) reduce your sample because the information you wish to merge is not available for every observation unit. merge should to be applied with care and lost observations should be tracked.
  • merge can also be applied to explore (1) how many observation units are in dataset A and in B or (2) how many observation units are in dataset A but not in B. This allows to, for example, learn about selection in the dataset.

[korenmiklos]

1. Changed to "shared notes".
2. Sorry, this is an unfortunate line break. Leaving it for now.
3. Duplicate.
   @andrasvereckei Please break out 4 and 5 into separate issues.

[andrasvereckei]

1. Changed to "shared notes".
2. Sorry, this is an unfortunate line break. Leaving it for now.
3. Duplicate.
   @andrasvereckei Please break out 4 and 5 into separate issues.

Done