[bluGill]

There are too many different exemptions to make a blanket statement. Sometimes user entered a wrong value is an exception and you need to handle it. Sometimes the exception is an out of memory error which realistically will never happen in the real world so you can ignore it.

[samiv]

User entering wrong input is a logical condition and typically handling a logical condition (point being it's not an error from the software's operational point of view) using tools that allow easier logic flow (i.e. error codes) etc.

Out of memory error can happen in many environments because for example in C++ it's about the allocator. You might have a custom allocator with some limit that has no bearing whatsoever what the OS does.

  * Use asserts/panics for bugs.
  * Use "error codes/values/enums" for logical conditions (that are errors to the user, not errors to the software)
  * Use exceptions for unexpected errors in the execution of the software (normally more or less just resource allocation failures)

[popdu]

Working under a simple structure to expect the best of people I am mandated to presume you read the article leaving me confused as to how your comment relates to social sciences. This causes me great discomfort and although objectively irrational, leads me to wish that the authors soon conclude their future work on exception aggregation to resolve the situation. This also reinforces my past evaluation that the Japanese sure are an interesting bunch.

[bluGill]

The abstract is in Japanese and the paper is unreadable on a phone (PDF needs to die) so I have to react to the only information I have : a headline and posted on hacker news

[popdu]

No worries, no harm no foul. Thanks for being an inspiration/springboard for my strawman larp at the article. It amused me greatly, hopefully amuses someone else as well. All in good fun.

[tappaseater]

Incredibly well said.