Why I don't file bugs

[nunfetishist]

Today, I encounted two reminders of why I don't submit bug reports to free software projects. We'll ignore the conversation I had last weekend with a LibreOffice developer at Collabora's office why I don't bother.

Firstly, I got a mail from Launchpad. Somebody had asked "It's been five years. Why is this bug still here?" on something I filed, well,

Comments

[spodlife]

Meanwhile over in the commercial software world the same thing happens - the bugs most likely to get fixed either bother the developers or a customer threatening to take their business else where.

[nunfetishist]

Commercial software's exactly the same, except the patch you can create adds lines to their bank account, not their source code. (I have two bugs open in Zeus that have been for nearly a decade, languishing unfixed in the "well, don't do that" state, too.)

[gerald_duck]

The sqlite team at least considered the bug I filed (which was that sqlite3_bind_text(myStmt, index, NULL, 0, SQLITE_TRANSIENT) shouldn't be rejected because a null pointer is fine when the number of characters is zero) but they decided by some baroque reasoning that the current behaviour was correct. )-8

[nunfetishist]

Changing the specification to fit the bug is a common approach to "WONTFIX", yes. Did it make you feel sad and miserable? I can't remember a single experience of bug-filing that had a satisfying outcome.

[pwaring]

The only positive bug-filing experiences I've had have been when the project is on GitHub, then I can fork, patch, commit and open a pull request. Seems to work pretty well for the documentation bugs I filed - 100% accept rate so far, compared with about 10% elsewhere.

My policy now is that if the project isn't on GitHub and the bug software doesn't support OpenID, I won't file a bug.