As announced last week, we discussed on how to involve the community in the development of Pontoon.
The first part is actually creating this category on discourse. We used mailing lists in the past, but they’ve not seen a lot of traction, so we’ll have the future conversations here. We’re hoping to get feedback on proposed developments as well as ideation and problem reports. The latter we’ll digest and then file bugs for.
For actual feature development, the plan is to involve community of different technical background at different phases. The earlier in the life cycle of an idea, the more abstract and technical the involvement will need to be.
In the earliest phase, we’ll work with rough mock-ups, and ad-hoc prototypes. Finding the right balance between sharing things early and sharing something that’s educating people about the proposal is going to be the hard part here. Involvement at this stage will require that you anticipate on what could be good about the proposal, and ignore the things that are bad about the prototype. We’ll also try to open up these experiments to people that want to create their own variations or alternatives.
Once we distilled the feedback from this cycle, we move on to a second stage, where we can create a pre-production version of the proposal, implemented in Pontoon, on a dedicated staging instance. This will be visually much closer to what we want, and also functionally. Often, this will be set up against dummy projects, so this is still a bit of an extra effort.
The next stage is to actually develop the feature (see, I’m not saying proposal anymore here) in a code quality so that it can land in the Pontoon code base, and go on the production web site. It would still be behind a feature flag, that is, we might only switch it on for some users, or for all, and we’ll take more wide-spread feedback. This is the point where anybody in the community can share their experiences.
Not all features and proposals will go through this process, and those that do might skip a step or two.
We also contemplate doing a design sprint or two. We’ll need to find the right topic and the right people to be involved to make that a good experience, though. And we need to keep an eye out that we’re not creating bubbles in such a sprint.