Should we delete topics that end up not containing any useful information?


(Csaba Pinter) #41

Agreed, a notification about deletion and at least a link to the now-private archived page for a few days would be welcome. At least I got an email with my comments after I complained that I couldn’t reuse my meticulously written feature request. I’ll create a new topic from that soon.

It is definitely interesting that they mercilessly hard-delete topics even with potentially useful discussions, while we debate an occasional deletion of “obviously useless” topics. I accept a never delete policy though, as long as the search results are not cluttered with those.

And yes, being able to use Discourse for Slicer for free is a huge perk for sure.


(Andras Lasso) #42

Immediate deletion is clearly a mistake. Anyone who were following the discussion but not getting email notifications could not see the answer and I could not even add a like to the answer. Also, valid questions were raised and got answered, so I don’t see the point in erasing this information. Finally, deleting an active topic feels like shutting the door in someone’s face while he is still talking - quite rude.

Anyway, this was a useful experience. It’ll help me to be a bit more empathetic towards people who post in the Slicer forum.


(Andrey Fedorov) #43

It’s interesting how my initial characterization was spot on. I’ve been communicating on forums and email lists well over decade now, but I don’t think I have ever seen such a hostile reaction from a forum leader. That was shocking to me.

But it was indeed a useful experience, even though it did not convince me in the advantages of the deletion, quite the opposite. It is not easy to define what “the topic” is, and how not to deviate from it. It is also difficult to assign the value to a conversation, since it is subjective. I still believe combination of tag-based curation, “thumbs up” and activity count is a better way to self-regulate the content.


(Csaba Pinter) #44

I added a new topic in the feature category on meta about ordering by the total number of likes: https://meta.discourse.org/t/ordering-search-results-by-total-number-of-likes-in-topic/89804
I hope the category and the description is acceptable, and we get an answer.


(Csaba Pinter) #45

Should we establish guidelines for tag-based curation?
The only thing that I can think of right now that would help me find a useful topic in the future is to add a specific tag in addition to the ones describing the topic for just this purpose. For example “FAQ” or something similar that marks a topic that contains a comprehensive and well-illustrated answer for a question that tends to come up every now and then.


Tag curation guidelines and best practices
(Jean Christophe Fillion Robin) #46

Then, would we migrate the following pages into discourse ?


(Csaba Pinter) #47

I’m talking about a tag called “FAQ”, not a topic. Many times when people don’t use the search or cannot fnd what they are looking for, but post a question that several others have asked before, then it’s easy to point to the best of those previous topics. I propose marking those “best previous topics” with a tag. It can be called something else if FAQ is not good.
Of course if ordering by aggregate likes worked, then they would be the ones that are on top of searches. I am just looking for alternatives, and following up on @fedorov’s “tag curation” suggestion.

I wouldn’t migrate wiki pages to discourse, because it’s a different modality: community discussions vs official documentation.


(Andrey Fedorov) #48

I agree.

In spirit of “staying on topic” :wink:, I am going to start a new topic to specifically brainstorm tag curation ideas.


(Csaba Pinter) #49

FYI we got an answer from the Discourse guys:

“This would require a major rewrite of the way search works. I’d estimate this would take a week of time for an engineer on our team who has been working with Discourse for years. For someone on the outside, who is not already intimately familiar with Discourse, probably more like 3 or 4 weeks of engineering time.”