Ability to leave reason for closing the error

[gmile]

Is your feature request related to a problem? Please describe.
When resolving the error, it would be convenient to have a way to leave a short comment on why it was closed. Nothing special, just a comment in plain text. Next time someone discovers the error is resolved, they could see the reason why & have some context about it.

Describe the solution you'd like
Have a pop-up with ability to enter plain text.

Describe alternatives you've considered
Have a small optional textarea available right next to the "Resolve" button.

Additional context
Not sure if this request is within the scope of the error-tracker. It could make for a nice addition though.

[kelvinst]

Have a pop-up with ability to enter plain text.

Can I suggest something: instead of a more specific "reason for closing", why not a more general "comments" feature? So people can use it to register a link to the ticketing system, or to articles about the error, or anything they would like to?

[crbelaus]

I still have to think more about this, but at the moment I also agree that a generic comment feature would be more flexible.

[odarriba]

In order to properly do comments we need to support a way of authenticating the user - maybe not having auth on ourselves, but receiving a user name via a hook or similar.

[kelvinst]

In order to properly do comments we need to support a way of authenticating the user - maybe not having auth on ourselves, but receiving a user name via a hook or similar.

Although I agree it would be nice to have info on the commenter, wouldn't it be a good first step to just add the comments features, and later introduce the user hook as you suggested?

[odarriba]

In order to properly do comments we need to support a way of authenticating the user - maybe not having auth on ourselves, but receiving a user name via a hook or similar.

Although I agree it would be nice to have info on the commenter, wouldn't it be a good first step to just add the comments features, and later introduce the user hook as you suggested?

I generally agree on that, but we need to think about it because splitting it in two phases would require:

• two migrations needed on two different releases (which can generate issues twice for people that doesn't check if a migration is needed and got an initial break)
• having to migrate old comments to have something like an anonymous user (?)

Also, adding support for users allow having other extra features on top of comments, like a history of changes (who closed it, when it was reopened, etc) that may be interesting.

I'm not saying users are 100% needed, but it is interesting to think about it to improve the developer experience on updates by reducing the number of migrations 😄