level of pain?! (new bug report guidelines)

[bitmode]

The new bug report guidelines have added yet another few steps to fill out. It's gotten to a point where I wonder how PDX actually processes these bugs.

10. Level of pain: The audacity of this bug to cause you pain! Tell us how annoying you’re finding it, on a scale of 1-5.

Why would QA ever use that info instead of their own judgement or the upvotes and other mentions? If "the reporter is angry about this" has any bearing on how a bug is handled, the process is seriously flawed. IMHO that has no value in a report. It is understandable that bugs are not judged just on technical merits but there are clearly better ways to estimate the impact.

8. Steps to Reproduce: If you are able to reproduce the bug, please write down steps needed to be done in order to achieve the desired result.

This point really covers almost half of what a bug report needs. (The other half being how the result is not working as designed; the description.) In very rare circumstances the reproduction steps might involve using a particular operating system, delivery platform, DLC combination, or particular modded value. If they are not needed for reproduction (i.e. almost always) they should not be there in my opinion. If the bug says "OS: Windows" that should mean I can't reproduce it on Linux. Rules 2, 3, 5, and 6 should not be separate points at all.

6. Do you have mods enabled?: Incredibly important! Let us know if you used mods when you experienced the issue.

No it's not important. This is meant to filter out bug reports that should not have been made in the first place or reported elsewhere. Clicking yes on this one, you might just as well throw the report away. I think this should rather be a checklist item like "Does the behavior result from a mod? Report to the mod author if so. If in doubt, reproduce it without mods". Same with rule 9, saves from previous versions.

11. Upload attachment

For graphical bugs, screenshots are obviously crucial. But other than that, if QA does not know what to do with a bug after having read the description and validated the repro steps, why waste time scouring over screenshots or save games instead of having the community clarify the report?

 

[Corpse Fool]

bitmode isn't the hero we deserve, but the hero we need right now.
In bitmode we trust.

 

[KubiG37]

I just find it hilarious that bug reports about a typo in an event description are marked as Level of Pain 5 - Ragequit

 

[Arheo]

The new bug report guidelines have added yet another few steps to fill out. It's gotten to a point where I wonder how PDX actually processes these bugs.

Why would QA ever use that info instead of their own judgement or the upvotes and other mentions? If "the reporter is angry about this" has any bearing on how a bug is handled, the process is seriously flawed. IMHO that has no value in a report. It is understandable that bugs are not judged just on technical merits but there are clearly better ways to estimate the impact.

This used to exist on our internal reportsa, so it probably crept in there. We don't use it internally, and shouldn't exist here - will remove when the forums feel like playing ball.

This point really covers almost half of what a bug report needs. (The other half being how the result is not working as designed; the description.) In very rare circumstances the reproduction steps might involve using a particular operating system, delivery platform, DLC combination, or particular modded value. If they are not needed for reproduction (i.e. almost always) they should not be there in my opinion. If the bug says "OS: Windows" that should mean I can't reproduce it on Linux. Rules 2, 3, 5, and 6 should not be separate points at all.

It helps us parse reports if they're clear. Operating system matters more than you'd think.

No it's not important. This is meant to filter out bug reports that should not have been made in the first place or reported elsewhere. Clicking yes on this one, you might just as well throw the report away. I think this should rather be a checklist item like "Does the behavior result from a mod? Report to the mod author if so. If in doubt, reproduce it without mods". Same with rule 9, saves from previous versions.

This field is very necessary. There are numerous bugs we've fixed that only occur in specific mods, but the standard to consider fixing them does differ. Using combinations of mods, homebrewed mods etc, can make it difficult to justify trying to find the faulty behaviour.

For graphical bugs, screenshots are obviously crucial. But other than that, if QA does not know what to do with a bug after having read the description and validated the repro steps, why waste time scouring over screenshots or save games instead of having the community clarify the report?

Screenshots usually make things quicker to repro, not easier (though sometimes both). Individually that might not seem like it matters much, but when you're talking dozens of issues in sequence, the time adds up. Savegames on the other hand, are 100% necessary unless the issue can be seen clearly from game-start. You'd be surprised how many reports have repro steps like 'start as germany, play the game until 1942, observe X' - someone has to sit and watch the game for an hour on the offchance that the game falls out the same way and something breaks. Think of our games like super complex state machines - the conditions for a bug to reproduce are often unique to the state that is achieved after n-billion potential variations: it's often impossible to reproduce the exact conditions.

 

[eastcoastceojam]

Thanks to @bitmode for highlighting this new policy, as I hadn't noticed it yet. A couple questions....

1. Does the rollout of these new bug report guidelines mean that older bugs should be resubmitted if they lack the newly recommended details?
2. Several of my bug reports are around errors in national focus or event code. Are save games, OS information, screenshots, etc. really required when I've cut-and-pasted the code in the bug report?

For the 2nd question, the devs and QA team may disagree with my assessment, but they wouldn't need to waste time examining saves to isolate the complaint. Here's an example of what I mean when I say a code error bug report:

Italy's "Aid for the Spanish Republic" doesn't set volunteer game rule

 

[Com]

me looking at the ace generation and air mission advisor bugs

 

[Herennius]

This used to exist on our internal reportsa, so it probably crept in there. We don't use it internally, and shouldn't exist here - will remove when the forums feel like playing ball.

Ok, didn't expect that...I thought that this "pain scale" is something being introduced on purpose and probably gradually for all pdx GSGs (reason: both Vic3 and Crusader Kings bug reporting forms have it as well and at least for the latter I can confirm that it is part of the form for a quite a long time already)