What do you guys think about a Trello Roadmap with Feature Voting?

[Regenwasser]

Today there was this exchange:

@shinuito

@Regenwasser

Btw why does OGS not have a community Trello to vote on issues or something?

I suppose there’s the forum which any user with an account on the main site can use to comment about or bring up issues. Comments/hearts, general agreement on what should be done does tend to get things done, but some things not as quickly as others it’s true.

There’s also the GitHub issues page for the repo, and that has the ability to upvote proposals also. It requires an account, but I suppose trello probably does too?

I imagine if they wanted they could use the new news page to ask to vote on a feature with the emoji reacts Play Go at online-go.com! | OGS Or just a poll on the forum, but it would need a list of requests gathered together anyway.

Maybe it could be something to do in future, have a list of popular requests that might be viable but would need priority.

I personally think that something like a Trello Roadmap with Feature Voting could potentially be very useful.

I know that some indie game developers are using these. One example would be this Trello from the game Barotrauma.

It seems that a while back OGS used UserVoice for feature requests and user suggestions. It seems that this effort has died down.

Key Advantages of using a Trello Roadmap

1. Accessible to non-technical users
   • “Cards” and “Votes” are easy concepts
   • No need for labels or github issue etiquette knowledge
   • Anecdote: Even I feel reluctant to engage on GitHub even though I’m an engineer
2. Prioritization via voting
   • Users can signal demand for what they want
3. Clean Roadmap
   • Higher ups control what appears on the board
   • Avoids the clutter of low-quality input (not saying that is the case here but there are 111 open github issues)
   • Users get an overview of what is currently being worked on
4. Strong community engagement
   • Users feel heard when voting
   • Encourages lightweight participation by simply clicking on an upvote button

Possible Downsides

• Might not be used by enough people as the OGS community is rather small, see the deprecated UserVoice effort
• Somebody has to do the curation of the board
• Not developer-centric as it doesn’t integrate with commits
• User votes could be misleading as popularity doesn’t always reflect importance
• Might feel like double maintenance to have Trello cards and GitHub issues

Possible Flow

Idea (from the OGS forum or other channels if existing) → Trello vote → Selected → GitHub issue → Development

[Samraku]

@Regenwasser

Might not be used by enough people as the OGS community is rather small, see the deprecated UserVoice effort

This. I liked the uservoice page, but I assume there was some cogent reason for retiring it

[_KoBa]

I don’t know about trello (i’ve never used it) but uservoice was taken off because it wasn’t very useful; it didn’t allow users to properly discuss about the feature ideas, or properly explain and elaborate what they would want their features to actually do. Uservoice also had some limit on the votes people could cast (iirc it only allowed 10 votes per user), so most people just ran out of votes pretty quickly.

So it was deemed to be better to just use the forum, so that people could hone those ideas together and give some input on what they would prefer from any new feature suggestion.

There’s also github, where people can also actually work on the code together and build those features ^^

But like i said, i’m not familiar with trello at all… Is it better for discussing about feature ideas than the ogs forum?

[Regenwasser]

@_KoBa

But like i said, i’m not familiar with trello at all… Is it better for discussing about feature ideas than the ogs forum?

I somewhat tried to elaborate that in the initial post but let me try to rephrase it.

The purpose is a different one:

• Forum → exploration & discussion
• Trello → (pre selection), prioritization & overview

The ideation process should happen in the OGS Forum (or other channels like Discord, Reddit or whereever else people are active).

But these ideas often end up being unfocused, repeated, fuzzy, lost in the ether, not clearly prioritized, and we lose track of them. They come up are discussed, then nothing happens, they disappear, and then 10 months later someone posts the same idea again.

A Trello Roadmap would show all ideas and their popularity at a glance. I encourage you to look at the Barotrauma example I linked above. That said the Barotrauma Trello was more active when the game was still being very actively developed. Now since the game is finally released and activity has died down somewhat so has the action on the Trello but I think it’s still a good example.

Trello is not good for in depth discussions but for transparency, community interaction and providing an overview. I think it would enable the community to get a deeper inside into what’s going on short and long term on OGS.

Few concrete examples and thoughts

• I consider myself to be a relatively frequent OGS user (although I’m a little on and off with Go) but I have absolutely no idea what is currently being worked on or what is planned for the rest of the year or next year
• Was the sound change intentional, is the community happy with it, will we ever talk about it again or is a thread with 10 users and 14 likes just not enough backlash? Did anything change about the sound of stones?
• What about the chat issue, will we forget about it again, how many people actually want a fix? Private chat blocks game chat
• What about the proposal to change the behavior of the First move timer for white in correspondence tournaments
• What about the Feature Request: SGF Comments with four upvotes. Is that being considered? New people interested in that idea will have a hard time expressing their support because the last comment was over a month ago so the thread has gone to Valhalla by now

[_KoBa]

Ahh right! So trello is more like a site where developers can tell what they are working on, instead being open for all users to make their wishes for new features?

I can see how ogs would benefit for something like that! Now it feels like all those wonderful people coding new features just code the feature and push it to the main site, but theres no coherent system to let users know what is being worked at, or what has been done recently.

ps. ogs has a link to changelog in the “about” page, but that link seems to be to the wrong section of github? The current link is for Commits · online-go/online-go.com · GitHub, but i think it should go for Commits · online-go/online-go.com · GitHub?

[Uberdude]

Without an increase in development resource, you might just end up with a beautifully prioritised list of hundreds of items and only the resources to do the top few. Also anoek is likely (and within his rights) to want to work on things he wants to work on, with some bias to what the community wants, but not controlled by it.

On my other screen:

and we have 8 developers.

[_KoBa]

@Uberdude

anoek is likely (and within his rights) to want to work on things he wants to work on

Hmm i think there’s a ton of things he wants to work on, but then theres constantly new bugfixes and other things popping up which become more urgent ://

But yeah, he def has a long to-do list he wants to work on, but i don’t think it’s written down anywhere…

Thank god he’s not the only developer working on ogs, but i guess everyone else has their own stuff they wanna focus on too, so it may feel kinda random sometimes

[GreenAsJade]

Koba has largely nailed it:

@_KoBa

there’s a ton of things he wants to work on, but then theres constantly new bugfixes and other things popping up which become more urgent ://

I know for a fact that this is a major reason why anoek doesn’t commit to “roadmaps or timelines”, because experience shows that any such commitment will inevitably be “wrong” and later “used as evidence in a negative way”.

And of course we are also subject to this one:

@_KoBa

everyone else has their own stuff they wanna focus on too, so it may feel kinda random sometimes

BUT

this doesn’t mean “the developers who are contributing would not like to see a better roadmap with clear discussion and preferences from the user community”.

Personally, I’d love to see that.

There is one missing ingredient, and one wrong assumption, that prevents this from happening: assuming that anoek will “run with it and own it”.

What it would take for “A Trellow Roadmap”, or any other “organised effort to collect user input and curate it” is someone actually doing that over time.

A proposal that says “would it be nice if we have this” is largely worthless in a practical sense.

Doing it … consistently and in a sustained manner, directing conversations from here to it, summarizing and keeping it up to date: that is the hard work that is needed to make such a thing actually useful.

The tool (Trello or whatever) doesn’t matter too much, as long as the person driving it can use it in the way that’s needed.

If the proposal relies on “anoek needs be involved in operating this tool” then it’s not going to fly.

If the proposal is “I’m going to collect all our input, and curate it and summarize it back on an ongoing basis”: frickin go for it.

anoek listens to discussions here, curation of the discussion could totally add value.

But it’s not a “tool idea” its an activity someone would have to do ongoing to make it work.

[benjito]

I believe the closest we have to sonething like that which works today is GitHub thumbs up. The list is sortable by reacts it is possible to prioritize that way.

+1 to all the comments to the effect of this not being a tooling issue, but a question of resources. If devs were to divert some energy from development to planning (which isnt a bad thing) i think id be more interested in a lightweight “here’s what were working on” blog post rather than adoption of trello (or whatever sprint board tool)