Talk:Community Wishlist/Wishes/Centralized Incident Management

This page is for discussions related to the Community Wishlist/Wishes/Centralized Incident Management page.   Please remember to: Sign posts using the four tildes (~~~~) Remain civil and polite during discussions. Place new text under old text (start a new post).

Thanks!

@Xaosflux this may sound pedantic, but could you share any good examples of an incident management system in an open source context? I see how reporting bugs or incidents could be useful, I'm not sure who'd be responsible for declaring something an incident. JWheeler-WMF (talk) 21:26, 16 September 2024 (UTC)

The license is of no concern, this has nothing to do with development. The mission of our projects also isn't primarily software development, it is about bringing knowledge to the public - which puts us apart from general OS software projects whose primary goal is to produce a product, not a service. To that end, please keep in mind I'm specifically referencing users of WMF production services here - not users of mediawiki software. Now, do I expect us to maintain a distinct incident record for every instance of a problem (something that causes incidents) - certainly not as we have millions of users. There are many incident management frameworks, but I think we have quite of a lack of w:en:IT service management when it comes to our end users. When these sort of users report problems they don't really care that it may be fixed in the next release, they care if their own personal experience has been fixed (they can't apply the next release). I'm not trying to be preachy about concepts here, most of the attention to this will be from tech people - please don't take any of this tone to be negative. Our phab tool could possibly be the ticketing system for this, but would need some workflow enhancements. Open->Claimed->Resolved doesn't mean that service is restored, or that a new feature is actually available. — xaosflux ^Talk 22:37, 16 September 2024 (UTC)

(Indefinite hold) - it's not indefinite, it's (in the vast majority of situations) only a week. * Pppery * _{it has begun} 16:26, 17 September 2024 (UTC)

Could be - but the reporters are left out of the cycle at that point; the point of this challenge is that it isn't a devops issue - it is a service management issue. I'm not trying to solve for our customer that is "users of mediawiki software", but our customer that is "users of WMF hosted services". They are never followed up with that their problem (using a production service) is supposedly resolved. — xaosflux ^Talk 09:29, 18 September 2024 (UTC)