TLDR
This post follows on the learnings from the Comments experiment. We’re taking a specific look at what we saw in Discussions and detailing some observations.
We’re proposing a further experiment with the current Discussions functionality, making Discussions tightly focused on one type of conversation, clearly separate from Q&A focus.
The post seeks feedback on options for that area of focus.
This is not a proposal to persist with the current functionality for the long term, but rather to use what we have to learn more as we consider the long term path.
What did we already know and what did we learn?
There’s been much said about the Discussions experiment since it launched, and the Meta community’s enthusiasm for it has ebbed and flowed. There is recognition of the potential value of such a space – potentially to house relevant content that doesn’t quite fit with Stack Overflow Q&A, or to provide well-intentioned users with a space to have subjective, open-ended conversations on technical topics.
There have been challenges stemming from the “scrappy” nature of the experimental feature, with its limitations as well as the level of accessibility to spammers. By design, there was never a clear definition of what Discussions is for, since the idea was to see what participants wanted to do there. There has been more focus on what Discussions is not for, which is informative but does not necessarily offer the clarity that users need to find success.
With the inclusion of Discussions in the Comment experiment, we sought to explore whether the space could alleviate the burden from the comment section. The assumption was that there was some overlap there, with Discussions as a way to start those conversations adjacent to (or spinning off from) a question or answer that needed space to flourish.
A key learning was that, while there is something to be explored in that realm, sending users from one somewhat ambiguous space (comments) to another (Discussions) was not offering many users the clarity they needed. And the overlap in use cases may not have been as great as we initially thought.
Next steps & your feedback
The original approach for Discussions was to start broad and to see what emerged. We’ve probably reached the limits of what we can learn there, with this implementation at least.
In the near term, we are:
Increasing the reputation requirement on Discussions to 2, temporarily as has been done in the past. This is to alleviate pressure from spam (as other teams explore platform-wide anti-spam measures) and reduce volume of posts as we look at next steps
Deleting Discussions posts generated from the comments experiment that did not have any replies. Those posts will get a reply (still visible to the author after deletion) noting that the post is being removed due to inactivity.
In the medium term, to maximize our learning with Discussions, we’d like to define the space’s focus — in other words, focus the types of conversations on a specific topic or theme. We have some thoughts on what that topic could be, but we’d like your input on where you think the current version, with a little fine-tuning, could be effective as a testing ground and learning tool.
Please note: This does not mean we’re “doubling down” on how Discussions works today. Clearly some changes will be needed regardless of direction, and we don’t consider the current Discussions product to be a foundation for a long-term solution. Rather than rehash the many missing aspects (which are well known), the intent of this post is to help determine the focus of a better-defined conversation space. With that narrower focus and structured guidance for this experiment, we will learn more about the potential of Discussions.
Also note that this post is not defining a long-term focus for a future Discussions space, this is only in service of the next experiment.
Our criteria for potential topics are something focused, and easy to explain and something likely to generate conversation.
These are options for the area of focus we’re considering for the experiment:
Each of these is expanded on in a linked answer. Feel free to comment and vote on each as you see fit and/or offer feedback in your own comments and answers. Please note that the name and description for each area of focus is meant to clarify for further discussion here, it’s not final wording by any means.
The feedback on the post will be one input into selection of a focus area. We’ll be using other tools to gather opinions from contributors and readers.
