Okay, let me start this with a disclaimer: I’ve been as part of my role/job in the same situation as the NC team, where one existing feature request tool after a while didn’t do the trick (anymore), and where a new tool has certain advantages and tied in nicer with the priotisation and development processes. And I’ve been part of the team that facilitated the move from old to new, by doing the migration work itself, and the communication with the people who raised the requests in the old system.
That being said, I wanted to make a few concrete proposals on how this migration could be done in a slightly more inclusive way, while maintaining the feeling of people that their observations and wishes matter, and at the same time still not having to migrate tons of outdated and “dead in the water” entries.
Prioritising
I get that it’s not doable to auto-migrate just the whole forum section, and that it’s also not a good idea. But the forum supports sorting by votes, by age, and by responses. Identifying a top10 or top50 or top100 is a matter of minutes. To copy over at least the “hottest topics” should not take days.
Involve people more to migrate their stuff
Even if it turns out that there is no NC capacity to copy over the top50 requests: at least nudging the creators of those top feature requests with either a copy-pasted message per thread or an automated @-ing would send people the message that their specific contribution is being seen, and would make people aware who didn’t notice this generic thread about the topic.
Get the community on board
The HA community doesn’t just include people who can contribute by developing code. With more than 2mio installations, there is certainly a number of people with a background in product management, communication, stakeholder management etc, who would be happy to do some of the leg work for a more inclusive migration effort.
Motivate people to move their stuff by highlighting the impact it can have
Current wording around feature requests includes e.g. “Suggest a feature, get fellow users to vote on it and see it happen” or “Sure thing, we hear you”. Moving to a new FR tool without moving stuff at all can be very demotivating. Yes, people can move their stuff over themselves. But how this is handled so far I’m afraid it sends a message of “well, if YOU still care about this one thing, then bring it along.”. The whole aspect of giving people the feeling and experience that their FRs actually also make a difference, get’s a bit torpedoed with such an approach in my opinion. I understand from a previous answer that there is no hard data on how many of the requests raised here actually made it into HA. But maybe there is some highlights that could be mentioned as part of the communication? Example: “Your input really can make a difference. Did you know that feature X, Y and Z that were released in the last 3 years actually started out as a feature request? Yours could be next!”
Long story short: I think there is options on the scale between “hard cut” and “we auto-copy EVERYTHING” that would enable transferring at least a top selection, while giving people a bigger sense that their input actually is being seen and matters, and without blocking NC capacity too excessively.
And to come back to the beginning of my text: I fully understand and support using the FR tool that does the job the best. And I know from own experience the pain of being stuck with an old tool
I am of mixed feelings about the move to GH. You WILL lose many users who have never heard of GH. Many of you know my feelings about GH- “The place where issues go to die”
More than a quarter-million unresolved issues starting 20 years ago does not give one any confidence that feature requests on this new forum won’t be equally ignored.
OTOH, we are told that the developers don’t browse the forum anyway which explains the dearth of feedback from the FR forum.
I share your concerns… Just looking at number of comments and upvotes vs. what we could observe in the Forum, I’m afraid this move will causes some ideas to loose the momentum.
In particular, in previous setup we were able to receive lots of workaround solutions, that people might have to achieve similar functionality without waiting for HA to support it.
Hopefully over time culture of periodic checking of GH will increase and there will be more discussion over new functionalities that community would like to see implemented.
You would go to GitHub and add a feature request.
Only other people who want to add feature requests will be there (and the developers).
The discussions, explanations and workarounds that is “answer” to most feature requests will not come from regular users that roam the forum.
So the end result on GitHub will more likely be a closed feature request with s very short comment like “already exist” or “not planned”.
Which is understandable, don’t get me wrong.
But I believe there is a big part that the regular users of this forum fill, that won’t be there on GitHub.
If my fears become reality the “not planned”/“already exist” messages will look cold and not be enough of an answer for neither new or semi old users.
Well, what I’m really afraid of is that judgement if feature requested is worth to be implemented will be in hands of Devs rather (who might, or might not, asses it only from technical perspective), than users, who might really like something to be implemented, but won’t even know about idea…
FRs in Forum that had 100+ comments are not uncommon. In GH the most commented one has 8 comments and probably 90% has no comments at all.
In Forum we’ve seen lots of FR with 100+ upvotes, in GH there are 7 with double digit upvote.
I know that the migration just started and it is not apple to apple comparison, but I’m afraid this separation will disconnect actual development from desires of community.
I have a better idea. While recovering from back surgery, I had time to think about this. Nothing else to do since ION network as all that was on the hospital TV.
Keep the new Github Feature requests. The devs may look at them once in a slow week- certainly more often than on the forum.
Here’s the idea.
Keep the Feature Requests but rename it “Feature Requests Discussion”.
Here’s why. Few forum browsers will open GitHub to see what new feature requests have been opened. This means the author really has no way to promote their feature request. Post it on GitHub and it disappears from the forum users forever. The discussion forum gives someone enthusiastic for the feature can promote it as a discussion and encourage people to go to GitHub and vote on it if they agree. Any activity here would show up as a new post. In the meantime, other users might offer workarounds or better ideas prompting the original author to drop the request. (I assume there is a way for the original requestor to drop the request).
This is a win-win. The devs who never visit the forums would have a better view of requests and the requestor would have a forum to discuss and promote their request or discover better solutions.
The new system notifies code owners of every feature request. It will get a lot more dev attention than the current system. In some instances, it already has fixed long standing feature requests. I suggest you get used to GitHub because I don’t see feature requests coming back here.
I see the merit in Steve’s idea, because I had a similar thought yesterday: a structured space on the forum for discussion purposes only may be good. I know the architecture discussion are very technical by nature (as they should be), so if the same style is followed, that may deter the layperson (this is not a criticism, but a possible reality where discussions may spiral into implementation rabbit holes). Having a structured place to flesh out ideas and test them beforehand may help the creation of an FR on GitHub and keep the GitHub side cleaner.
Devs don’t watch the forum posts- they have made that quite clear. Users don’t watch GitHub or Disturb in any regular fashion.
If a user posts a feature request on GitHub, that’s the end of it. Other users don’t know there is a new features request. Unless the posting user further clutters the default “Configuration” category with discussion of the request.
If there is no “Features Request Discussion” category where users, not devs, will be able to discuss their request with other users, possibly finding support from other users (the purpose of user forums?) and even workarounds or alternate solutions.
There is NO CONNECTION between GitHub and the Forum, so most users will never know of new feature requests or even have the opportunity to discuss them and decide if they want to endorse the request with a vote. The entry into the Home Assistant world has been lowered to the point that the average user has never heard of GitHub.
As planned, anyone who feels passionately about a feature request posted on GitHub will promote it on the forum. Where? Where will the thread land?
This is why a FR Discussion category on the forum would benefit everyone. (Especially the moderators).
Because we have to moderate forums as it is. Right now feature requests are moderated across 4 or 5 places. This will unify discussions and organization in 1 place. That’s much less moderation.
Sorry, GitHub is the space where things are organized and tracked. It makes sense for issues and feature requests to reside there. FRs being over there is no different than issues and users have no issue finding and using GitHub for that. I understand you don’t like it and your feedback has been noted.
Again, a quarter million unresolved issues on Github is not confidence inspiring.
This change will result in far less user participation and add to the unrelated clutter on the user forum. I am NOT suggesting that feature requests be created on the user forum, only that there would be a category where users could discuss their requests.
